0:00:07 The energy grid and electrical grid of the future, it’s not just going to be the dichotomy of generation, transmission, and storage.
0:00:12 This sort of next generation of what the grid looks like is going to be in a much more decentralized way.
0:00:14 Why are delivery costs such a big problem?
0:00:20 The grid is aging now and brittle. The workforce has aged out.
0:00:24 Should we just leapfrog the grid? I need this power now, today.
0:00:27 The United States needs to get better at megaprojects.
0:00:29 Things that are a billion dollars, things that are at scale.
0:00:35 There is no safety, there is no national defense, there is no national security without a reliable electrical grid.
0:00:40 U.S. energy usage per capita peaked in 1973.
0:00:42 Since then, it’s been flat.
0:00:46 Meanwhile, China’s per capita energy use has grown ninefold.
0:00:54 Today, with AI, EVs, manufacturing, and data centers demanding more power than ever, America’s electrical grid is buckling.
0:00:57 We haven’t just underbuilt it, we’ve forgotten how to build it.
0:01:06 In this episode, I’m joined by A16Z general partners David Yulevich and Aaron Price-Wright and investing partner Ryan McIntosh from the American Dynamism team.
0:01:11 We talk about how the U.S. energy system broke, why fixing it is about more than megawatts,
0:01:15 and what it’s going to take from new tech and talent to faster permitting and smarter software.
0:01:25 As a reminder, the content here is for informational purposes only, should not be taken as legal, business, tax, or investment advice,
0:01:32 or be used to evaluate any investment or security, and is not directed at any investors or potential investors in any A16Z fund.
0:01:37 Please note that A16Z and its affiliates may also maintain investments in the companies discussed in this podcast.
0:01:45 For more details, including a link to our investments, please see A16Z.com forward slash disclosures.
0:01:53 U.S. energy peaked in 1973 in terms of per capita usage.
0:01:56 China’s has increased ninefold over that same time period.
0:02:00 We have some reasons to be optimistic now that things have started to change or will change further.
0:02:02 Why don’t you give some context there?
0:02:04 What’s happened and why should we be excited about what’s coming forward?
0:02:09 The history of the grid in the United States was build big power plant industry formed around it.
0:02:12 The grid grew incredibly fast through the 20th century.
0:02:15 Then around the 80s and 90s, things started slowing down.
0:02:19 And then through the early 2000s, the grid effectively froze in the United States.
0:02:25 A big piece of that was a lot of the energy generation, a lot of the manufacturing, a lot of sort of heavy industry moved to Asia.
0:02:29 And so for the last 20 years, effectively, the grid has ossified.
0:02:32 We forgot how to build new power plants.
0:02:38 We forgot how to build new power projects, new loads, sort of large data centers, large factories, large megaprojects.
0:02:41 And you say we forgot, were we not allowed to, or we actually just lost the know-how?
0:02:42 We were allowed to, but we lost the skill set.
0:02:45 And I think you can see it in more extreme examples with nuclear power plants.
0:02:47 All of that sort of transition happened decades before.
0:02:54 But basically, the grid itself, the grid operators, forgot how to plan, how to move quickly, how to do it cheaply.
0:02:57 And so now we’re at this point in time where we are reshoring.
0:03:00 And we are bringing back manufacturing, we are bringing back data centers.
0:03:06 And there’s this highly concentrated demand, and it’s now, now, now at sort of any price, but they cannot move fast enough.
0:03:07 And so that’s what we’re seeing today.
0:03:10 We talk about data centers and the grid being inflexible to this.
0:03:12 It’s playing catch up.
0:03:16 We need to do a lot of the growth that happened in China and bring this here and do it incredibly fast.
0:03:20 How did this forgetting happen, and how can this relearning happen or this retraining happen?
0:03:22 It’s a good question.
0:03:24 I think a lot of it’s like a workforce issue.
0:03:25 I think a lot of it’s a policy issue.
0:03:39 I think the United States, historically, was a bunch of regulated utilities, sort of the top-down, big thermal power plant, big transmission lines connecting to substations, then distribution lines going to individual factories, homes, things like that.
0:03:45 And I think some of the newer technologies don’t necessarily benefit from scale in the same ways that these large thermal plants typically did.
0:03:50 And so this sort of next generation of what the grid looks like is going to be in a much more decentralized way.
0:03:53 So there’s also an element of relearning of what is the grid actually?
0:03:57 Is the grid these large sort of power systems and large infrastructure projects?
0:04:04 Or does it look like a lot more decentralized way where we can eliminate a lot of the wires in between, things like delivery costs, which have increased exponentially?
0:04:07 And can we do it in a more dynamic and flexible way?
0:04:10 So things like solar and batteries, they don’t need to be massive.
0:04:11 You can put them anywhere.
0:04:12 You can put them next to load.
0:04:16 And so this is also sort of an element that grid operators are thinking through.
0:04:22 How do we do that while also managing frequency, voltage, things like that, without causing sort of a grid to go down?
0:04:23 So there’s a lot of challenges.
0:04:28 Well, if you think about what our grid is, I mean, it’s a piece of technology that was designed about 100 years ago.
0:04:33 And very little technology on the grid has changed in those 100 years.
0:04:36 And you look at why are delivery costs such a big problem.
0:04:38 Grid is at capacity.
0:04:42 Getting a new project onto the grid today, you know, you sign up for interconnection, it could take a decade.
0:04:46 There’s a backlog of 20 plus years to get a new transformer.
0:04:51 The transformer technology we’re using today is kind of bananas if you actually look at what makes a transformer.
0:04:58 First of all, there’s one company that makes these, and there’s one plant in the U.S. that produces the right type of steel that you need in order to make these transformers.
0:05:03 And it’s 100-year-old tech, and it’s like the wait list for transformers is insane.
0:05:12 So I think we’re getting to the point in demand starting to rise again and the calcification of grid technology where it’s like should we just leapfrog the grid?
0:05:15 Like do we really need to wait in line and wait for this to catch up?
0:05:17 I think there’s like two versions of how you do that.
0:05:23 One is how do you get power gen and power storage essentially as close to demand as possible?
0:05:34 And that’s a problem for new tech to really help solve because we’re talking about instead of these like mega projects that we’re used to building, well, not used to building anymore, like massive nuclear plants, massive new natural gas plants, etc.
0:05:40 We’re talking about much smaller and more distributed sources of power bypassing interconnection altogether in many cases.
0:05:48 We’re seeing that as a pretty big trend with data centers as data centers are just building power directly on site and co-locating power with the data center because they’re like, I can’t wait.
0:05:52 Microsoft’s like, I can’t afford to wait 10 years to get an interconnection with the grid.
0:05:54 I need this power now today.
0:05:59 So how do you get power more tightly coupled with the load that it’s actually going to serve?
0:06:10 And that’s a really interesting problem for tech also from a software perspective because if you get generation, storage, and usage all co-located very closely together, like that’s a very good problem for AI to solve, like reinforcement learning.
0:06:14 Stick that in there and suddenly you get massively efficient systems that you couldn’t get at grid scale.
0:06:20 I think an interesting point to add on to that is that there’s very little visibility into the grid itself.
0:06:29 So like they understand sort of power plants are operating or not, but especially at sort of the distribution level, the things, the power lines you might see outside your home, there’s very little understanding of what’s actually going on there.
0:06:35 And so there’s like a reluctance, especially when you have things like net metering where I’m sending from my home a battery back to the grid.
0:06:37 Things get incredibly complicated.
0:06:44 And so the grid operators don’t have a very good understanding of when can they allow new projects to go online, how much power, when to actually cut people off.
0:06:53 And so there’s a lot of these policies, interconnections, sort of the general term, but like states like Texas have much more lenient policy of you can build wherever you want.
0:06:55 If we need to cut it off from the grid, we’re going to do so.
0:06:57 And so it’s sort of this connect and manage approach.
0:07:09 Whereas other states, like they will do these incredibly long feasibility studies in like a variety of sort of scenarios with like the entire grid is at peak capacity, but they want to make sure this specific project can stay online 24-7.
0:07:13 So that ends up creating these massive delays of being able to study every single possibility.
0:07:16 So there’s a lot of policy approaches here as well.
0:07:27 And there’s a bunch of these technologies called grid enhancing technologies, which are effectively like, you know, an average power line might be used at 50% capacity, but it needs to be designed for the peak capacity for the summer when everyone has their AC on.
0:07:31 And so there’s a lot of sensors or other technology that could be placed there.
0:07:35 So you have a much more dynamic view of what our infrastructure actually was looking like.
0:07:39 And so when we have these new technologies, then we can much more efficiently use the infrastructure we have.
0:07:42 What are your reactions to this conversation thus far?
0:07:45 Where are some areas you’re particularly excited about or reasons to be optimistic?
0:07:57 Well, I think the reason we’re having this conversation is we’re touching on a bunch of these topical themes, which is that we’re in a moment in time where, exactly like Ryan said, the grid is aging now and brittle.
0:07:59 The workforce has aged out.
0:08:01 We’ll talk about that more, I think, in a minute.
0:08:14 But we had to go out and hire and train entire specialized crews, specialized people that work with cement and concrete, specialized people that work with steel to go build the large Vogel reactors in Georgia.
0:08:17 We put them on Vogel reactors three and four.
0:08:18 We turned them on.
0:08:19 Huge win.
0:08:22 And then those people that went back to building highways or like bridges or something else.
0:08:31 And like instead of just going and put them on Vogel like five, six, seven, eight, nine, ten and building just like this massive crescendo of nuclear power, we just put these people back into the general workforce.
0:08:33 And so we’re not learning our lesson there in the workforce.
0:08:51 And at the same time, we have this insatiable thirst for energy, whether it’s EVs, whether it’s data center compute for AI, or just generally a shift toward more and more consumption of electricity, or even just like the reshoring and manufacturing, all these things that are just very, very electron heavy.
0:09:01 At the same time that we’re seeing this, you know, what Aaron brings up to me is a piece that we hardly ever talk about, which is resiliency and not having people be as dependent on the,
0:09:03 the interconnectedness of the grid.
0:09:15 People that you deploy solar, you know, we talk about like distributed compute for those of us that are in the tech world and like how important it is to like have distributed compute and have networks be able to suffer and survive through segmentation things.
0:09:17 But like the grid is very interdependent.
0:09:22 Even in the US, there’s really only a few major regions that can segment themselves off.
0:09:34 But when you deploy solar or you deploy batteries, or you deploy an SMR reactor or your own power generation on site for your own data center, you don’t have to worry about how brittle the grid is because you’re fairly resilient from it.
0:09:44 And I think that’s a component is that the energy grid and electrical grid of the future, it’s not just going to be the dichotomy of generation, transmission and storage.
0:09:52 But as Aaron brought up, you might do all three of those things in the same place and not have to worry about how robust the grid is or how capable the grid operators are.
0:09:56 And I think that’s a dimension that was never important to people before, but it’s important today.
0:10:06 And you can certainly imagine if you’re the military, you certainly care about having reliable access to power at all your forward operating bases and even at home at your home military bases.
0:10:09 Like you just cannot lose your ability to have electricity.
0:10:11 And so I think all these things are just coming together at once.
0:10:14 And it’s really exciting moment in time.
0:10:33 And I think it’s buoyed by the fact that we’re also at this sort of technology inflection point where AI can help some of these things, not just be a consumption driver, but even be an enabler and facilitating more efficient use of electricity, better monitoring of the grid, better ways to even go through the regulatory and permitting process, which is onerous for many cases.
0:10:42 Building on that, I think Texas, literally today, we’re recording this during a massive heat wave that’s affecting most of the eastern and southern United States.
0:10:59 And if you compare the grids of Texas with New York, Texas famously, historically, had massive grid failures several years ago when a big heat wave came through, the grid couldn’t keep up with all the air conditioners that were going on, and people saw massive power outages.
0:11:00 Everyone was really mad.
0:11:02 People were like, oh, ERCOT doesn’t work.
0:11:03 DREG doesn’t work.
0:11:05 And what has Texas done in the couple of years since that happened?
0:11:09 They have absolutely flooded the grid with solar capacity.
0:11:13 Texas has doubled their solar capacity in the last approximately three years.
0:11:17 And with that, they’ve just deployed thousands of batteries.
0:11:26 One of our portfolio companies, Base Power, is one of the players here, but there’s many battery power companies deploying all across Texas to provide storage for that solar power.
0:11:38 And if you look at the performance of the Texas grid versus the performance of the New York and surrounding area grid during this heat wave, I must have seen 10 news articles this morning about how well Texas grid has done.
0:11:45 The elasticity and ability to react to very quick changes in demand without having to change kind of baseload power.
0:11:50 Like, you can’t build a new natural gas plant or a new nuclear reactor overnight.
0:11:53 But solar is just so insanely cheap.
0:11:58 Like, it’s basically having a giant, massive, huge nuclear reactor in the sky that will go forever.
0:12:00 And Texas isn’t a green state.
0:12:01 This isn’t a political issue.
0:12:11 But it’s like, why aren’t we deploying the world’s cheapest form of power literally everywhere we possibly can and then just putting batteries everywhere?
0:12:12 Like, there just should be batteries everywhere.
0:12:19 It’s bananas to me that batteries as a topic has, like, recently gotten caught in the sort of political crosshairs.
0:12:23 We really, as a society, need to be good at power storage and batteries.
0:12:25 Like, this shouldn’t be a controversial topic.
0:12:27 We invented the lithium-ion battery.
0:12:42 And yet today, if you want to buy a battery, whether it’s for a drone or for the grid or for your car or for whatever it is, like, you’re either buying a battery made in a lights-out factory in China or you’re buying a battery produced in Vietnam by a Chinese company.
0:12:45 And, like, there’s no meaningful effort in the U.S. to change that.
0:12:54 And I think that this is a really critical problem, not just to manage power load on the grid, but for power for all of the things that we need to power the next generation of innovation in the United States.
0:13:03 I think we’d be hard-pressed on the American dynamism team to think of a company that we’ve met with an interesting technology in the last two years that doesn’t have a battery in it somewhere.
0:13:08 So, as a country, we need to be investing in battery technology and battery manufacturing.
0:13:21 By the way, if China decides that whatever your company is doing that’s using batteries doesn’t align with what they like or they want to punish you, being cut off from being able to buy batteries from China is incredibly punitive to a company.
0:13:23 And we’ve certainly seen that happen with some of our startups.
0:13:31 And then you find out quickly that the ability to procure and source batteries from places that are not in China is very difficult.
0:13:39 If you extrapolate that out to what would happen to our whole country if we just were unable to buy batteries from China, it could be catastrophic in a very short period of time.
0:13:49 Just to add on to a quick point on the grid side of batteries, if the rest of the country, which is presumably watching what’s going on in ERCOT, which is the grid operator in Texas,
0:14:02 if Texas can prove that you can deploy these sort of decentralized distributed energy resources and to sort of flatten these peaks, provide more resiliency and ultimately lower price of electricity, then every state should go and do this.
0:14:07 There’s a very complex web of deregulated and regulated entities when it comes to the grid.
0:14:13 Of course, there are a lot of different policy and workforce and political reasons why not everywhere is this decentralized world.
0:14:18 And it’ll probably be more complex than just these deregulated energy-only markets that Texas works with.
0:14:23 But I think this is going to be very obvious if it isn’t obvious already.
0:14:29 And I think the United States needs to move incredibly fast to make this happen and hook up batteries, solar panels, make it easier and cheaper to do it.
0:14:33 Even co-location for large loads is still a very politically fraught issue.
0:14:35 Utilities are pushing against this.
0:14:38 It’s still really hard to hook up solar and batteries to your home.
0:14:43 I think it’s actually cheaper to put residential solar on your home in Germany than in the United States.
0:14:46 And that’s largely a permitting, largely an installation issue.
0:14:47 That’s crazy.
0:14:48 That should not be the case.
0:14:51 Erin, I believe the quote in your college yearbook was,
0:14:52 drill, baby, drill.
0:14:57 How do you think about your sort of love for oil and gas with other sources of energy?
0:14:59 Oh, my parents will be shaking their heads if they hear this.
0:15:04 I think, broadly speaking, our approach to energy in the U.S. just needs to be yes and.
0:15:11 You look at the sort of atrophy of our power build-out over the last 30, 50, whatever, you know,
0:15:14 you name your time frame years compared to, let’s say, China.
0:15:19 And if we want to accomplish the goals that we’ve set out as a society to accomplish over the next decade,
0:15:20 like, we need more power.
0:15:22 And it’s a matter of yes.
0:15:25 And I think solar and batteries, extremely important.
0:15:29 And D, you should talk more about the exciting things that are happening around nuclear.
0:15:32 But, like, there is a place for oil and gas.
0:15:35 Like, I cut my teeth at Palantir working in oil and gas.
0:15:37 My husband worked in oil and gas.
0:15:40 Like, the first check I wrote at A16Z isn’t an oil and gas company.
0:15:43 So, this isn’t me coming with a particular agenda around carbon.
0:15:47 This is me coming and realizing that, like, we basically need every tool in our toolkit.
0:15:54 And we should be using technology to deploy whatever makes most sense, wherever it makes most sense at scale.
0:15:58 If we’re talking, like, energy mix of where we’re at today and where do we think we’re headed,
0:16:05 if I were to make, like, a personal bet, solar batteries are just the ability to be incredibly cheap and deploy incredibly fast.
0:16:07 Spin up and spin down.
0:16:07 Yeah.
0:16:09 And I think that already is a way, but I think that will continue.
0:16:12 But I think to be very clear is that you need all different types of energy.
0:16:15 You’re going to need true sort of baseload, dispatchable power.
0:16:16 It’s going to be gas.
0:16:17 It’s going to be nuclear.
0:16:18 It’s going to be geothermal.
0:16:20 It’s going to be a lot of hydro as well.
0:16:25 And I think as you attach more of these sort of renewable resources or these non-reliable resources,
0:16:28 while they’re incredibly cheap and works most of the time,
0:16:34 this long-tail risk, once you get to, like, 50% to 75% of the grid, is going to become very, very expensive.
0:16:36 You need a lot more battery backup, things like that.
0:16:39 And so I think it’s going to be very complex and it’s going to be different for many different regions.
0:16:43 But certainly it’s not all of any given resource.
0:16:50 Yeah, when you look at, like, the changing nature of load over the next decade, some of that is going to come from data centers.
0:16:51 Some fraction.
0:16:57 I would say it’s probably overstated how much data centers contribute to the growing load in the United States over the next decade.
0:17:00 Data centers generally are baseload.
0:17:05 If you’re training a model, you’re largely using a dedicated amount of power for the long term.
0:17:08 Maybe there are some fluctuations if you’re doing more inference.
0:17:11 But I would generally say, like, data centers represent baseload.
0:17:13 But then you also have things like electric vehicles.
0:17:15 You have things like heat pumps and air conditioners.
0:17:20 You have industrial autonomy, which may or may not be running 24-7.
0:17:24 So you’re going to have some increase in the base level of power we as a society need.
0:17:29 But continuing to increase the size of the peaks and troughs of how we use energy on a day-to-day basis.
0:17:37 And we should be thinking about designing our grid and designing our energy mix and power sources around what those loads look like.
0:17:41 And not over-solving for either baseload or variable power.
0:17:50 I think just to put this in more tangible terms, the peak summer load in places like California might be half of what it is in winter or something like that.
0:17:51 And it depends on what climate you’re in.
0:17:58 And so the concept of baseload is, like, do you build all the plants you’d need for the 100 gigawatts of power you’re going to need when it’s wintertime?
0:18:01 Or in the summer, you know, half the year, you’re only going to need 50.
0:18:02 So, like, what would be baseload?
0:18:08 What you need to do in sort of modern civilization is, every time you turn on the switch, like, the power is working.
0:18:14 And so how you actually match supply with that very, very fluctuating load, both daily and seasonally, is very complex.
0:18:21 And so, like, today, you might have to build a natural gas, what’s called, like, a peaker plant, that might only operate, like, a week a year.
0:18:28 And so that’s an incredibly expensive asset that is going to only be delivering very expensive power, but is only needed when all the other resources are tapped.
0:18:31 And it’s, like, that last couple megawatts of power.
0:18:44 The alternative now you could do is what’s called, like, demand response or with batteries on the grid is say, okay, well, instead of doing this $10,000 sort of a megawatt hour plant, I can just make it so everyone’s thermostat in this area turns down a couple degrees.
0:18:49 And so then an aggregate means that I don’t need to build that large asset or pay that expensive premium.
0:18:51 Okay, pushing back on that.
0:18:59 Like, I think that the American consumer will fully organ reject that level of dictation over how they use their power.
0:19:10 I think a more likely outcome is that you can do it on the compute side and just say, look, these three racks of the data center are just going to go offline during the peak summer heat when you’re running your AC.
0:19:13 This is not a critical job.
0:19:13 Right.
0:19:14 It’s a non-critical job.
0:19:15 It’s not a mission job.
0:19:17 It’s a back office job.
0:19:19 And you’re just going to run it at night instead of during the day.
0:19:22 And you’re going to pay less electricity for that benefit.
0:19:26 I don’t know if I agree with Aaron that AI is not going to suck up all the compute.
0:19:30 I think that Constellation just turned up a new nuclear reactor or is reactivating a reactor.
0:19:40 I think Meta immediately sucked up all of the power that they’re going to generate or nine-tenths of it or something from the new Constellation reactor that Meta signed the contract extension for.
0:19:50 And so I think we actually probably are underestimating the amount of compute that we’re going to soak up with electricity over the next 10, 20, 30, 40, 50 years.
0:19:52 The amount of data we’re going to start storing.
0:19:53 Just look at video.
0:19:57 The amount of video we create per minute has just ballooned way beyond anyone’s expectations.
0:19:59 I’m sure the same will be true for AI compute.
0:20:05 And I think once you start getting into like robotics and autonomy, if you think about compute expansively, I totally agree.
0:20:05 Yeah.
0:20:12 And so like those things are going to be much more responsive than do I want to go have my room be 74 degrees instead of 71 degrees.
0:20:17 Well, let me tell you, anyone that’s done business in Tokyo in the summer knows as a nation, by the way, Japan has done this.
0:20:19 It is absolutely terrible.
0:20:20 It’s horrible.
0:20:21 We’re not going to do that in America, please.
0:20:23 We’re not going to do that.
0:20:24 We’re not going to do that in America.
0:20:30 You’re in like the 40th floor of a Japanese building wearing a suit, by the way, because you have to wear a suit.
0:20:31 It’s swelteringly hot.
0:20:34 Everyone is walking around like they’re not miserable, but they are miserable.
0:20:36 And you watch them out a window, but the window won’t open.
0:20:38 It is one of the worst.
0:20:39 Why doesn’t France work in the summer?
0:20:40 This is why.
0:20:41 It’s terrible.
0:20:41 Exactly.
0:20:43 So we hate this idea.
0:20:44 I’m spoiled living in California.
0:20:49 I will say like one of the biggest proponents of this or current users is like crypto mining.
0:20:50 It’s like these are flexible.
0:20:51 Yeah, but that’s right.
0:20:53 But those people are demand responsive, right?
0:20:56 So those people will just turn off their compute when it’s not cost effective.
0:20:56 Yeah.
0:21:01 But it’s important for the grid is that you can build these assets and you have the demand for power here.
0:21:02 Like they’re going to soak up that demand.
0:21:06 But if it gets far too expensive, they will also shed that demand.
0:21:10 In the U.S., my guess is that this already is reflected in the fact that you have peak load pricing.
0:21:14 Like for me, my power is 10x more expensive between the hours of 4 and 9 p.m.
0:21:15 So we don’t run the washing machine.
0:21:16 Yeah, right.
0:21:20 And so leave it to organizations or individuals to figure out how to manage that.
0:21:22 But just like charge people for more power.
0:21:26 It’s a little bit of a non sequitur, but I like that we keep talking about oil and gas.
0:21:27 We’re talking about natural gas.
0:21:28 We’re talking about batteries.
0:21:29 We’re talking about solar.
0:21:30 We’re talking about nuclear.
0:21:34 Ryan even mentioned hydro, which of course is totally viable in some places.
0:21:39 The thing that nobody ever brings up anymore, except for I think a very fringe group, is wind power.
0:21:42 And I’m very happy to hear that nobody here is jumping for wind.
0:21:45 I think wind is incredibly cheap when it’s working.
0:21:48 You kind of know solar is going to work and like sort of this reliable schedule.
0:21:49 The sun’s going to be out.
0:21:50 It’s going to be working.
0:21:53 Spare some cloudy days, but there’s still always something coming through.
0:21:55 But the wind might not blow for a week.
0:21:56 I think it’s worse than that.
0:22:01 I think I read that globally, one third of all wind turbines are out of service at any given time.
0:22:07 The other thing is, I think wind is the only power generation mechanism where when you get too much of the input,
0:22:10 the blades of a wind turbine feather and turn off.
0:22:13 Whereas there’s still something that’s too much sun for solar.
0:22:16 Too much water and hydro, like that’s not a problem.
0:22:18 But like too much wind and the wind generator turns off.
0:22:21 Well, who wants a system where you get more of the input you want and then it stops working?
0:22:25 It’s also just extremely hard and dangerous and specific to service.
0:22:29 Like you see those videos of people climbing the ladder up to the top of the wind turbine.
0:22:35 Grid operators look at wind and it’s like great when it’s working, but they can’t plan for it.
0:22:37 They have to build other capacity to supersede that
0:22:38 if it’s not going to be there when they need it.
0:22:39 So fine, I cede wind.
0:22:41 We can move past wind.
0:22:42 Yeah, I agree.
0:22:42 So no wind.
0:22:46 But I do think the demand response, this went to the point that Ryan brought up,
0:22:47 that monitoring the grid is really important.
0:22:50 Being able to send signaling on the grid is really important.
0:22:52 And you have to remember, we’re all used to the internet,
0:22:55 which has bidirectional communication and messaging.
0:22:57 It has data layer and control layers.
0:23:00 And there’s like a full control plane and things like that for the internet.
0:23:02 The electrical grid doesn’t really have that.
0:23:04 And so to be able to send messaging and things is very, very difficult.
0:23:07 And now a lot of people just do it out of band using the internet.
0:23:13 To actually send messaging and do monitoring of the grid itself without an overlay network is very hard.
0:23:16 And that’s one of the challenges that people are now, I think, starting to address.
0:23:21 Yeah, it’s wild how much of a mystery what’s happening on the grid is at any given time.
0:23:23 Like we really have very little visibility.
0:23:31 And it’s very hard for, I think, centralized utilities to deploy meaningful software to understand that.
0:23:36 So when we think about as VCs, like what types of things do we look at and what do we get excited about?
0:23:40 I think companies that kind of are going at this monitoring from the opposite direction.
0:23:43 Like how do you get software almost insidiously on the grid?
0:23:49 Like how do you start learning more about demand and generation as close to the source as possible?
0:23:52 And then try to feed that information back from each other.
0:24:00 Like the idea that you’re going to go sell a software tool to a PG&E or similar and have a reasonably speedy top-down implementation
0:24:04 where you actually get good signal and metrics and can actually do interesting things with that data.
0:24:08 To me, I find like a little bit unbelievable.
0:24:11 Something very interesting that I learned is a lot of the load forecasting,
0:24:15 which is basically like the tasking of when plants need to go online.
0:24:19 So there’s usually a 24-hour ahead sort of market, day-ahead market that’ll basically say,
0:24:21 you need to run your plan at this time.
0:24:23 And then they sort of supply and demand match to a price.
0:24:24 And there’s like a merit order.
0:24:25 It’s complex, but that’s how it’s done.
0:24:29 But most of this load forecast is done by just looking at the weather.
0:24:31 They look at basically one of the best indexes.
0:24:35 They look at where the homes are, how many people are there, and then what the temperature is going to be.
0:24:38 That often is sort of the largest factor that goes into this modeling.
0:24:42 But if we have all these sort of connected resources, we have solar, we have EV chargers,
0:24:45 all of this stuff is spitting off data, telemetry, and things like that,
0:24:50 we’re going to get a much better look of how like load is actually being forecasted real-time,
0:24:54 which is going to help a lot of understanding like where do we actually need to build?
0:24:55 Like what is the actual price of power?
0:24:57 And then you can start making these markets, I think, a lot more efficient.
0:25:02 Well, when you look at energy desks for the big hedge funds or energy trading companies,
0:25:06 their weather guy is usually the highest paid person on the desk outside the portfolio manager.
0:25:10 Those climate and weather PhDs that are working on a trading desk,
0:25:16 they are just absolutely raking it in because they’re like God right now because there’s very little other data.
0:25:19 That’s why you see when the stuff that goes on in Texas, like heat waves and things like that,
0:25:23 if they even get it wrong by a couple of degrees where it’s like they think it’s going to be hot,
0:25:26 but it gets actually really hot, that’s when you get these crises,
0:25:28 like crises that end up causing a lot of strain on the grid
0:25:32 and then you have to turn off all these very expensive plants and then you get the headlines.
0:25:34 That are usually also the ones that are the worst for the environment as well.
0:25:35 Yep.
0:25:39 Let us know if you have any reactions to this or otherwise give us the state of nuclear.
0:25:40 Where are we right now?
0:25:41 What are the major bottlenecks?
0:25:42 What are we excited about?
0:25:48 I think that the biggest thing that’s shifted in the last two or three or three or four years in nuclear
0:25:51 is that everybody now acknowledges that nuclear energy is clean energy.
0:25:55 I think that’s been one major shift in public sentiment and perception.
0:26:01 Nonetheless, there’s still major headwinds politically with nuclear that need to be overcome.
0:26:04 Taiwan, for instance, turned off their last nuclear reactor.
0:26:05 Insane.
0:26:06 It’s unbelievable.
0:26:11 This is an island country that is seven days away from a total energy blackout
0:26:13 if they get an oil and gas blockade from China
0:26:17 so that they can’t bring in ships to deliver oil and fuel.
0:26:20 So at any given time, they’re like seven days away from a total blackout
0:26:22 and they turned off their last nuclear reactors at all.
0:26:22 Why’d they do it?
0:26:27 Because they caved to political, like very loud vocal minority groups.
0:26:28 Like environmentalist activist reason?
0:26:29 Yeah.
0:26:33 This party ran on a commitment to turn off the reactor before they realized how stupid it was.
0:26:34 It’s just like colossally stupid.
0:26:39 By the way, turning off a reactor, like a real full-scale reactor, it’s not like an SMR where
0:26:42 you can just flip it on like a few days later or a month later.
0:26:45 With these large reactors, it could take years to turn them back on.
0:26:46 That was just terrible.
0:26:51 But broadly, I think the tailwinds for nuclear are just getting stronger, where people recognize
0:26:52 realize that it is clean energy.
0:26:55 I think there’s still messaging work to be done.
0:26:59 We should stop calling the spent fuel nuclear waste because it’s really not waste.
0:27:01 Almost all of it can be recycled and reused.
0:27:04 People do need to recognize that those tailwinds are shifting.
0:27:05 So that’s happening.
0:27:08 I think that people understand that it’s baseload power, right?
0:27:10 So it’s not dependent on it only working during daytime.
0:27:15 It’s not like hydro where you have to be around an appropriately configured water source.
0:27:21 And then I think that one of the largest inhibitors to creating new power plants in this country,
0:27:23 it’s not that we can’t do it.
0:27:24 We can.
0:27:29 There’s a huge regulatory and permitting, I would say, morass that has to be swam through
0:27:36 that is incredibly expensive, requires an army of consultants, many tens of millions of dollars,
0:27:41 many, many thousands of pages of applications and documentation and process review.
0:27:45 And again, this has to do with building the power plant, getting the fuel, transporting the
0:27:46 fuel, storing the fuel.
0:27:52 Each step along the way is extremely laden with regulation and policy.
0:27:54 And some of that’s for good reason.
0:27:59 But finding ways to better navigate that to make it more efficient is really a step in areas
0:28:02 that a lot of companies and people are focused on now.
0:28:06 And actually, I think the government now is also focused on how do we streamline the approval
0:28:08 process for a new reactor?
0:28:10 How do we start approving new reactor designs?
0:28:16 And then I think the last thing I guess I would say is that right now, if you’re going to put
0:28:19 a lot of energy and work into building a nuclear power plant, you want to build a really big one.
0:28:23 We largely only see really big power plants in this country, like the AP1000s that we turned
0:28:24 on in Georgia.
0:28:28 And those again came in, I think they were 10 years late and multiple billions over budget.
0:28:31 We only do that because if you’re going to put in the effort and time, you want to get the
0:28:33 most bang for your buck and generate the most power.
0:28:38 We are now starting to see movement from the government in the DOD, in the Department of
0:28:43 Energy, and from the national labs to really try to create a more fast track process for
0:28:50 these small modular reactors or even micro reactors that use a much safer form of fuel, use much
0:28:57 less nuclear fuel, use a different kind of nuclear fuel that’s not nearly as risk prone as the kind
0:29:01 of nuclear material you’d use them like a weapon, but it’s not nearly as enriched to the same degree.
0:29:03 It’s not even the same material.
0:29:06 And so that process is now getting a lot of steam.
0:29:08 We have an investment in a company called Radiant Nuclear.
0:29:12 They are building a factory that creates what effectively is an SMR.
0:29:14 They would probably call it a micro reactor.
0:29:16 It’s a one megawatt reactor.
0:29:19 It can be put on the back of an 18-wheeler and shipped around.
0:29:23 You can move it to where you need power if there’s been a natural disaster, like a hurricane,
0:29:25 and you need to bring in power overnight.
0:29:30 You could bring in a few trucks with four or eight of these reactors and power up a whole
0:29:31 city after a disaster.
0:29:34 And so that kind of flexibility and power is really compelling.
0:29:36 I think there’s a lot of tailwinds, a lot of good things happening.
0:29:42 One thing to understand is that the DOD spends an incredible amount of money dealing with the
0:29:46 cost and, frankly, the risk to human lives, not just the cost, but the real risk to human lives,
0:29:50 transporting fuel around the world to forward operating bases.
0:29:55 Anytime we do a military exercise, anytime we’re engaged in a conflict, the movement of
0:30:00 fuel factors in as a primary concern and consideration of what we deal with.
0:30:04 And so we’ve read reports that they spend well over $200 a gallon at times, sometimes up to
0:30:09 $400 a gallon for diesel, effectively, to get diesel into the right place at the right time.
0:30:14 And so you can just imagine that having a nuclear reactor you can put on the back of a C-130
0:30:19 and fly around the world to wherever you need power, drop it in the middle of the desert,
0:30:21 turn it on, you have power for five years.
0:30:22 It’s just an incredibly compelling value prop.
0:30:25 There is no question that nuclear needs to be part of the equation.
0:30:30 Not only is that baseload power, but on the SMR micro reactor side, it gives us this incredible
0:30:32 flexibility in grid resilience.
0:30:36 There should not be a single military base in this country that’s not nuclear backed from
0:30:42 a power standpoint, because if the grid goes down, whether it’s from a cyber attack or just
0:30:48 instability or demand issues or cascading failures, you want to be able to fail over to nuclear power
0:30:50 and not worry about the runway lights turning off.
0:30:50 Yeah.
0:30:55 And especially as we start to look at the kind of electrification of our weapon systems,
0:31:02 our military vehicles, our drones, et cetera, like those all need to get charged up somewhere.
0:31:05 And how better to charge them than a nuclear reactor?
0:31:09 Another thing I’ll add to your nuclear comment is I think the advantage of nuclear, and I think
0:31:13 that Radiant has done very well of really leaning in on, is the power density factor.
0:31:18 If you want a reactor that is reliable and power dense, you want it to operate at very high
0:31:19 temperatures.
0:31:22 You want as highly enriched fuel as you possibly can, where it makes sense commercially.
0:31:26 So you want Halo fuel and you want to serve customers that will pay the premium for that.
0:31:29 That’ll be able to buy this reactor that they know is going to work.
0:31:32 And if you’re doing that, you want to have these economies of scale on the manufacturing
0:31:33 side.
0:31:36 You want it to be done out the door and don’t need to like assemble it on site.
0:31:38 You don’t want to have to like have constant maintenance.
0:31:42 And I think the other sort of reactors that we see, maybe on the civilian side, if you build
0:31:46 a reactor in a factory or you build modular components in a factory, but you still need
0:31:51 to do construction work on site, you’re still a construction company, even if the technology
0:31:51 is there.
0:31:54 And I would argue a lot of the existing AP1000 technology is quite good.
0:31:56 And other countries can do it quite cheaply.
0:31:58 China is using a very similar design.
0:32:00 The UAE just built one for incredibly cheap.
0:32:04 And they have very similar nuclear regulation, like in terms of frameworks.
0:32:07 And obviously their regulatory bodies might move faster and things like that, but they’re
0:32:09 not like completely ignorant of some of the concerns.
0:32:12 Well, and maybe this is a much more broad question, but the United States needs to get better
0:32:13 at mega projects.
0:32:16 Things that are a billion dollars, things that are at scale.
0:32:20 And I would argue the NRC is a big component of why it’s expensive, but I think it’s also
0:32:24 the same reason that it takes a billion dollars to build a bike lane in San Francisco is why
0:32:25 we are not able to build stupid power.
0:32:28 Or why we don’t have a high-speed rail in California.
0:32:28 Yep.
0:32:31 We might not have a high-speed rail in California because nobody wants it.
0:32:33 And nobody wants it where they’re building it.
0:32:33 I want it.
0:32:34 I fly to LA all the time.
0:32:35 Sorry.
0:32:36 Nobody wants it where they’re building it.
0:32:37 Sure.
0:32:37 Yeah.
0:32:39 Bakersfield is not a prime destination.
0:32:42 I want to train from San Francisco to LA.
0:32:44 It takes an hour and a half.
0:32:49 We debate a lot internally, like, where does it make sense for VCs and VC capital to plug
0:32:49 in?
0:32:55 And arguably, like, we’re not going to move the needle on, you know, these multi-billion dollar
0:32:56 mega projects in the U.S.
0:33:01 Like, we’re probably not the best people to figure out how to capitalize and build a multi-billion
0:33:04 dollar project in California to generate the power for the grid.
0:33:10 But I do think that there is a role for technology at kind of, like, every single layer and every
0:33:12 single phase of how mega projects get built.
0:33:17 It’s like, how do you use AI to navigate kind of site selection?
0:33:22 How do you use tools to, like, move through the various permitting processes faster?
0:33:27 Like, how do you use AI to help you do extremely complex and interdependent project management
0:33:28 better and more effectively?
0:33:33 So that’s something that you have a project with 4,000 people working on it and everyone
0:33:36 engaging with different suppliers and timelines that are dependent on each other.
0:33:39 Like, how do you get all those things to align better so that you don’t get these 10-year
0:33:44 delays so that projects actually happen on time and on budget and, as a result, attract
0:33:46 private capital backers?
0:33:48 Like, I think that there’s a role of technology here.
0:33:49 You know what that looks like.
0:33:50 TBD.
0:33:53 We’ve seen a lot of companies that maybe five years ago were primarily trying to sell
0:33:57 to utilities and grid operators, which is incredibly painful, incredibly difficult.
0:33:58 Perhaps rightfully so.
0:34:00 Like, they have poles in the ground that are 50 years old.
0:34:03 Why would they trust a two-year-old company to sell them software?
0:34:04 Are they going to be around in 20 years?
0:34:07 And this is a fair question to ask, especially for something as critical as the grid.
0:34:11 But now they’re developing this software and there’s such demand of understanding how grid
0:34:15 operators might think and potentially get there faster or, you know, have different conclusions.
0:34:20 And so now you can go to data centers or people who want to build solar farms or people who
0:34:21 want to build massive, like, battery farms.
0:34:23 And you can sell a very similar software.
0:34:26 Individual people who want to make sure that their power isn’t going to go out and they’re
0:34:29 going to be caught without energy during an important moment in their lives.
0:34:29 Yeah.
0:34:30 And so everyone cares now.
0:34:35 There’s a lot more money who cares about what is the grid actually going to think and where
0:34:35 can I build?
0:34:37 Where is there excess capacity?
0:34:41 Maybe I’m connected to the grid, but I also need some battery and solar backup or like a
0:34:45 radiant microreactor or something like that to be used in certain situations.
0:34:49 It’s a lot more complex, this sort of microgrid setup, but it’s the way we’re headed.
0:34:50 And software is going to be a big piece of that.
0:34:55 I want to hear more about our requests for startups or things that we want to exist that
0:34:55 we haven’t yet discussed.
0:34:59 I mean, put differently, I’m curious where we think there’s most bang for the buck in terms
0:35:03 of the issues that we’ve been talking about in terms of if there was like a regulatory intervention
0:35:05 or some sort of technological unlock.
0:35:06 What comes to mind?
0:35:11 One area where there’s probably a venture scale software company to be built is really around
0:35:13 grid management monitoring.
0:35:19 I think we see this in the IT landscape, we see it in the OT landscape, but we don’t really
0:35:22 see it in the grid where there’s just full, very, very large.
0:35:24 There is no Splunk for the electrical grid.
0:35:27 There’s no power to networks for the electrical grid yet.
0:35:31 There’s a whole bunch of things that mirror the IT and OT landscape, whether it’s around cyber
0:35:34 and monitoring and logging and analytics.
0:35:37 There’s no looker for the electrical grid yet.
0:35:38 There’s just none of these companies exist.
0:35:42 I’m not sure if it’s three separate companies, I’m not sure if it’s one company, but there
0:35:48 is a big company to be built and really managing and monitoring the grid and helping to orchestrate
0:35:53 and even deal with some of the things Ryan spoke around, around demand response, coordinating
0:35:56 that, creating those marketplaces, tracking all those incentives.
0:36:00 So I think when we see a company that we think can really be the breakout company there, we
0:36:00 would lean into it.
0:36:06 I also think around sort of project planning and development, how do you make it faster
0:36:09 and easier to build projects within the current regulatory framework?
0:36:10 How do you do site selection?
0:36:12 How do you navigate permitting?
0:36:14 How do you navigate project management?
0:36:16 How do you navigate your sort of construction supply chain?
0:36:20 We’re starting to see companies pick off pieces of that, but I think broadly speaking, there’s
0:36:24 room for tech and software in that kind of project development space as well.
0:36:29 I think, you know, in a more general sense, like anything that can bring generation capacity
0:36:32 or storage capacity closer to load, I think is going to be very compelling.
0:36:37 And a lot of the times it’s less maybe the technology, novel technology, but it’s a system integration
0:36:38 or it’s an innovative business model.
0:36:42 I think something like radicalizing that I experienced is, and I implore everybody to go home and check
0:36:46 their power bill, they’ll now like often separate like the delivery costs from the actual
0:36:47 generation costs.
0:36:51 So what we’ve seen and we’ve mentioned it, but like the cost to generate electricity, the cost
0:36:55 of like power has dropped immensely, gas, solar, things like that.
0:36:58 But the cost to actually deliver that electricity has increased a ton.
0:37:00 And so in net, it’s sort of not changed.
0:37:01 And I think that’s terrible.
0:37:03 And I think we all agree that’s bad.
0:37:06 And so I think there’s a lot of opportunity of bringing sort of that generation capacity.
0:37:09 In some ways, this is sort of like this more liberalizing force.
0:37:10 It’s like we all should have our own backup.
0:37:11 We all should have our own technology.
0:37:14 I think there’s a lot of really interesting ways to do that and scale it.
0:37:19 And overall, as a grid gets more heterogeneous, all of the seams and intersections between
0:37:24 things, like there’s just so much more opportunity for technology than when you had a single utility
0:37:27 managing a single source of power centrally distributed out broadly.
0:37:30 I’ll throw another one out there as I’m just thinking about this.
0:37:36 One thing that I’ve been noodling around is this idea that all the regulation and permitting
0:37:40 and policy frameworks that we have in this country, you could think of those as part of
0:37:44 the like infrastructure that we all have to like live and work with and interact with.
0:37:49 So companies that really facilitate, and I would say applying AI to navigating the permitting
0:37:50 process.
0:37:51 So nuclear is a good example.
0:37:58 Again, a nuclear reactor application or a fuel transport license or a fuel manufacturing license.
0:38:03 These things have thousands and thousands of pages of regulation and documentation that go
0:38:04 with them.
0:38:08 You make one small change in your application, it has these reverberation effects.
0:38:09 You have to update all your documents elsewhere.
0:38:13 If you’re the regulator trying to go through all these applications, it’s just incredibly
0:38:18 onerous, borderline impossible to imagine that a regulator can even possibly get it right.
0:38:20 You could argue that it’s actually not possible.
0:38:21 They just do a best effort.
0:38:27 But AI could actually help these things, could help the applicants go through the process of
0:38:30 filling out their applications and saying, hey, this is where you should drill down.
0:38:31 This is where you should clarify.
0:38:36 They can look at all previous published applications and say, this is how you need to tailor it.
0:38:41 You can probably make it 85% the same, and then based on your design or your location or
0:38:42 whatever, make some modifications.
0:38:45 And then the regulator can do the same and say, look, here’s an application that came in,
0:38:49 highlight all the areas I need to drill down, or show me the things that are different from
0:38:52 every other nuclear fuel transport application we’ve ever seen.
0:38:54 Are they using the rail infrastructure?
0:38:57 Are they using the national highway infrastructure to move the fuel?
0:39:02 AI can just automate all these things that take armies of consultants months or years to do,
0:39:05 can be brought down into being minutes or hours.
0:39:08 I’m not sure how big of a company, but I think potentially there’s a very large company
0:39:09 to be built there.
0:39:13 If our check sizes were in the billions, not just the millions, which were in Jason Horowitz,
0:39:15 you never know, how would our strategy change?
0:39:17 I think you need even more than just billions.
0:39:18 It’s tens of billions, hundreds of billions.
0:39:20 It’s such a tough question.
0:39:23 I think there’s tons of policy around this as well.
0:39:27 I hesitate to say we should look to how China has built up their grid,
0:39:30 but I think the elephant in the room is in the early 2000s, they were experiencing blackouts.
0:39:32 This was a very common thing.
0:39:33 This was horrific.
0:39:37 But now, think of 4X their grid in the last couple of decades.
0:39:41 And so the way they’ve done this is by basically deploying generation capacity,
0:39:46 building hydro, building massive storage facilities, of course, BYD, CATL, tons of battery production.
0:39:51 They built HVDCs, these large high-voltage transmission lines.
0:39:52 I would do all of that.
0:39:53 I mean, I would look to all of it.
0:39:56 And whether or not it’s a good investment or not, I’d look at a number of factors,
0:40:00 but it is much more of the infrastructure projects, the glue that connects this stuff together.
0:40:05 I think our lens today is looking at these technologies that enable a lot of this more flexible grid.
0:40:08 But I think there’s also going to be these large infrastructure, the webbing in between it.
0:40:12 I think software is a big piece of it that we’re spending a lot of time on looking at.
0:40:14 But I think, how is ERCOT going to be connected to the rest of the grid?
0:40:16 Or how are we going to move this power around?
0:40:19 If it’s really sunny in the Southwest, solar is going to be really cheap.
0:40:22 Is there an efficient way to move that to New York or something like that?
0:40:23 China’s done this effectively.
0:40:26 And I think if you had hundreds of billions of dollars spent or trillions of dollars,
0:40:28 what does the grid look like?
0:40:29 Like, it’s going to be a lot more interconnected.
0:40:33 Maybe another answer to your question or a different answer to your question is,
0:40:37 I think the energy industry is probably medium to long-term,
0:40:42 like one of the most prime spots to deploy physical autonomy.
0:40:50 So when you think about applications of robotics, whether it’s humanoids or more kind of task-specific
0:40:53 robotics, we’re talking about dangerous jobs often.
0:40:57 We’re talking about manufacturing jobs to build up, whether it’s small-scale reactors or batteries
0:40:58 or whatever.
0:41:04 So I don’t know what the shape of the company is here and how reliant it would be on some of
0:41:06 the robot learning work that’s happening.
0:41:10 But I do think that as we scale our energy capacity, there’s going to be a pretty massive
0:41:13 application of industrial robotics to the energy sector.
0:41:19 I made this joke, I think, once to Ryan, or maybe I made it at the American Dynamism Summit,
0:41:24 that we survived the greatest nuclear disaster in U.S. history just recently when we finished
0:41:28 the Vogel 3 and 4 reactors and let all those employees go back to other jobs.
0:41:32 So I think if we were writing a billion-dollar check into power, what we would do is we would
0:41:36 just give jobs to those people and not let them go back to whatever it was they were doing
0:41:38 before they were building nuclear reactors.
0:41:43 And we would just really work to streamline the process to make sure that we go build Vogel
0:41:48 5, 6, 7, 8, 9, 10, and all these different states around the country and just put these
0:41:52 people to work for the next decade-plus building reactors.
0:41:56 And that, to me, was the greatest miss and probably the greatest opportunity.
0:42:01 I don’t think it’s particularly our opportunity, but I do think it’s an opportunity for somebody
0:42:02 to do.
0:42:07 Labor broadly, like, this is a little tangential, but when Microsoft was building their new data
0:42:12 center in Georgia last year, at one point, they had on staff at Microsoft or on contract
0:42:15 more than a third of the electricians in the state of Georgia.
0:42:17 And they basically maxed out.
0:42:20 They hired every single electrician that they possibly could.
0:42:23 So I don’t think it applies to just electricians.
0:42:29 It’s, do you, to your point, like the cement mixers, it’s the mechanical engineers, it’s the nuclear
0:42:30 engineers.
0:42:36 Like, how do we actually train the next generation energy workforce that we’re going to need to
0:42:39 modernize the grid is a big, big challenge.
0:42:43 These are very high-paying jobs where you don’t have to check your email on your phone at 9 p.m.
0:42:45 at night after you go home from work.
0:42:46 They’re high-paying.
0:42:48 It’s good exercise jobs.
0:42:50 And they’re relatively low stress.
0:42:51 These are good jobs for people.
0:42:55 I think one more comment on this, and it’s more of an industrial policy question, is we’re
0:42:59 talking about specific things, but like oftentimes that just moves the bottleneck.
0:43:01 Like we could solve a lot of sort of the grid connection hookup.
0:43:05 We could build a lot of like transmission lines, but then we need more transformers.
0:43:07 And to build more transformers, you need more electric steel.
0:43:10 You can do the same sort of equation for much of the supply chain.
0:43:14 Battery is another good example is then, okay, cool, we’re building cells, but then we also
0:43:17 need active materials and we need to mine and things like that.
0:43:21 And so, you know, it’s sort of a whole effort of examining sort of our infrastructure and
0:43:23 our supply chains, and you need to do all of it.
0:43:25 And I think that’s a complicated question.
0:43:26 That’s an expensive question.
0:43:27 Any last reflections?
0:43:32 I think the last thing I would say is that people underestimate how critical and important
0:43:38 a resilient, reliable, dispatchable electrical grid is to our national security.
0:43:42 You cannot have national defense and national security without reliable electricity.
0:43:44 It’s just not possible.
0:43:48 So all of these things we’re talking about are about the upside, about capitalizing on
0:43:53 AI compute, the switch to electric vehicles and our insatiable thirst for electricity.
0:43:56 But at a fundamental level, there is no safety.
0:43:57 There is no national defense.
0:44:00 There is no national security without a reliable electrical grid.
0:44:06 To reiterate on that, people want reliable, cheap, and clean power in that order.
0:44:09 And I think that’s, I think largely how we should think about our energy policy.
0:44:12 And I think that’s sort of the direction we’re going.
0:44:14 And I think we need to make sure we stay aligned with that.
0:44:16 That’s an exciting note to wrap on.
0:44:17 David, Ryan, Aaron, thanks so much for coming to the podcast.
0:44:18 Thank you.
0:44:19 Thank you.
0:44:23 Thanks for listening to the A16Z podcast.
0:44:28 If you enjoyed the episode, let us know by leaving a review at ratethispodcast.com slash
0:44:29 A16Z.
0:44:32 We’ve got more great conversations coming your way.
0:44:33 See you next time.
0:44:44 We’ll see you next time.
0:44:44 We’ll see you next time.

0:00:02 What is the journey over the next decade?
0:00:06 It’s about the speed at which humans can change their workflows.
0:00:08 Why doesn’t the breakthrough that we just saw get released?
0:00:11 Why doesn’t that permeate every corporation within six months?
0:00:16 It’s so strange to me how many disruptions are happening all at the same time.
0:00:17 Your R&D is changing.
0:00:18 Yeah.
0:00:20 Every part of the stack is changing.
0:00:20 Like everything.
0:00:23 We’re not in like a fear of AI world.
0:00:28 We’re in a, we know this is going to happen and it needs to happen to us faster than it
0:00:31 happens to our competitors, which is a totally different dynamic than we saw with cloud.
0:00:36 What do you think is the best metric for anybody interested in tracking this stuff as far as
0:00:37 like how fast it’s going?
0:00:37 Is it GDP?
0:00:39 Is it margin?
0:00:40 Is it top line?
0:00:42 Is it headcount growth?
0:00:43 Is it all the above?
0:00:47 It’s basically fully assumed that AI is going to take over the enterprise.
0:00:50 How does AI actually change the enterprise?
0:00:54 Not just in theory, but in how software is built, sold, and used?
0:01:00 In today’s episode, A16Z general partner Martin Casado sits down with Aaron Levy, co-founder
0:01:05 and CEO of Box, to explore what it means to be an AI-first company from product strategy
0:01:06 to internal workflows.
0:01:11 They talk about why incumbents may be better positioned than expected, how startups can still
0:01:15 break out, the rise of agents and vibe coding, and what happens when the bottleneck isn’t the
0:01:16 tech, but the org chart.
0:01:22 Aaron also shares how Box is using AI internally today and why he thinks the next generation of
0:01:25 employees may spend more time managing agents than writing code.
0:01:26 Let’s get into it.
0:01:33 As a reminder, the content here is for informational purposes only, should not be taken as legal
0:01:37 business, tax, or investment advice, or be used to evaluate any investment or security,
0:01:42 and is not directed at any investors or potential investors in any A16Z fund.
0:01:46 Please note that A16Z and its affiliates may also maintain investments in the companies
0:01:47 discussed in this podcast.
0:01:54 For more details, including a link to our investments, please see A16Z.com forward slash disclosures.
0:02:01 Aaron, thank you very much for joining us.
0:02:01 Thank you.
0:02:03 Everybody here already knows you.
0:02:05 However, I still think you should intro yourself, just for completeness.
0:02:12 Aaron Levy, CEO, co-founder of Box, and at Box, we help enterprises basically take all of their
0:02:18 unstructured data or enterprise content and turn it into valuable information, and AI is
0:02:20 absolutely this incredible accelerant for that problem.
0:02:22 I just learned that we’re investors, didn’t you?
0:02:24 Well, many years ago.
0:02:24 Many years ago.
0:02:26 So no claims post-IPO.
0:02:31 But actually, Ben Horowitz had this early kind of blog post on basically, I think it was
0:02:32 the title of The Fat Startup.
0:02:32 Yeah.
0:02:33 Yeah, yeah, yeah.
0:02:35 In response to enterprises, the lean startup.
0:02:36 Yeah, that’s right.
0:02:40 And let’s just say we very much took that to heart, and we basically like deployed every
0:02:45 single lesson, which was like the name of that game is you get big fast, you scale aggressively,
0:02:48 and that was a very important period in our company’s journey.
0:02:52 So the notional topic of this is AI in the enterprise.
0:02:56 But I think it’s good to be kind of nuanced about this, because it’s less obvious than
0:03:02 people think, and you’ve been talking a lot about AI on X, but also you’re thinking about
0:03:03 it in the terms of your business.
0:03:07 So let me just kind of set up the first question as follows, which is, AI has historically been
0:03:11 this very B2B enterprise thing, like chatbots or whatever, personalization systems.
0:03:16 But what’s unique about Gen AI is a lot of the use cases are actually like a consumer or
0:03:17 prosumer, right?
0:03:23 Think like creativity or developers, and it actually hasn’t made intros as much.
0:03:24 into the enterprise yet.
0:03:25 It’s just starting now.
0:03:27 So maybe just a couple of questions.
0:03:30 First off, A, does that match with your experience?
0:03:34 And then B, how are you thinking about this transition to the enterprise?
0:03:35 Yeah.
0:03:41 I think if you were to probably like do the idiosyncrasies of AI and then reverse engineer
0:03:45 why that was the journey, basically up until, let’s say, pre-chat to be team moment,
0:03:47 AI was extremely hard to use.
0:03:52 It required in many cases having custom models for basically every problem you tried to solve.
0:03:57 And so there was almost no way that a consumer ecosystem could flourish based on that.
0:03:59 It was just not generalizable enough.
0:04:03 There was really few products other than like maybe Siri, Alexa, et cetera, that you’d interact
0:04:05 with that would even have some sense of AI.
0:04:10 And so enterprises were the early adopters of AI systems to bring workflow automation to
0:04:11 their companies.
0:04:16 Then boom, ChatGPT happens and all of a sudden it’s the exact right form factor for mass adoption.
0:04:17 There’s no startup costs.
0:04:19 It costs two seconds to learn the product.
0:04:20 It’s a chat interface.
0:04:24 So it was like perfectly ripe for just taking off in the consumer space.
0:04:29 And then you have also these incredible conditions set up for mass adoption.
0:04:31 You have billions of people on the internet.
0:04:32 It was set up as a free product.
0:04:36 Again, it kind of solved this sort of latent kind of question mark that everybody had, which
0:04:40 is like, when are we going to see AI work and touch our lives?
0:04:45 And so everything was kind of like the perfect conditions to get mass consumer adoption.
0:04:49 On the enterprise side, you have unfortunately kind of the opposite, right?
0:04:53 You have lots of workflows that have been kind of ingrained for decades and decades.
0:04:59 You have lots of legacy IT systems that have data kind of not set up well to be accessed by AI.
0:05:05 You have a sort of shadow IT problem, which is most corporations don’t want, and users just
0:05:10 injecting text into prompts that might contain information that the AI models could learn off of.
0:05:14 So it’s sort of a difficult environment for that same level of virality.
0:05:20 With the exception of a few of these pro-sumer categories, I have talked to large corporation
0:05:24 CIOs that are seeing people just show up with Windsurf and Cursor and Replit.
0:05:28 And so you’re getting actually this sort of shadow IT version that we saw 15 years ago.
0:05:30 DevTools has always had that.
0:05:31 Yeah, 100% fair.
0:05:31 100%.
0:05:32 So DevTools have had that.
0:05:36 But I think that you’re still seeing that now in the chat to BT kind of leakage into organizations.
0:05:37 Right.
0:05:42 I’m sure their pro-sumer inside of a corporation firewall usage is off the charts, even separate
0:05:43 from the people that pay for it.
0:05:43 Totally.
0:05:48 So now the question, though, is what is the journey over the next decade for the real
0:05:54 change management of deployment of AI systems that drive the more like GDP changing productivity
0:05:55 gains?
0:05:58 And that’s something where I do think we have to be prepared for.
0:05:59 This is many years.
0:06:03 It’s about the speed at which humans can change their workflows as opposed to how kind of quickly
0:06:06 the technology can just sort of evolve in advance.
0:06:11 And so we in Silicon Valley and certainly anybody tuning into this sort of imagines like, well,
0:06:13 why doesn’t the breakthrough that we just saw get released?
0:06:16 Why isn’t that permeate every corporation within six months?
0:06:20 And it’s because like people just have meetings and they have budget processes and they have
0:06:24 to go through a governance council and they have to get compliance on board and they have
0:06:28 to figure out like who has the liability when the thing recommends this stock and then the
0:06:30 financial services provider shares that with a client.
0:06:33 Like that takes years and there’s going to be case law that needs to happen.
0:06:36 And we still have lawsuits that are going on about who owns the IP of this stuff.
0:06:38 So that part is going to take years.
0:06:42 What’s interesting, and I think you’ll especially appreciate this on the cloud side is I remember
0:06:47 when we first were scaling up in the enterprise, let’s say 2007, 2008, 2009, let’s say that three
0:06:53 to five year period, post AWS, post kind of cloud starting its journey, basically to a
0:06:58 T, every conversation you’d have with a CIO or a group of CIOs was basically like, yeah,
0:06:59 that’s nice.
0:07:01 Maybe some little corner of our organization could use this.
0:07:03 We are never going to go fully to the cloud.
0:07:05 They had their arms wrapped around their servers.
0:07:06 I remember.
0:07:06 Yeah.
0:07:09 And basically they did not want to give up the infrastructure.
0:07:11 There was too many questions, too many compliance issues.
0:07:15 There was just existential job questions of, well, what happens when this, you know, gets
0:07:16 delivered as a service?
0:07:17 Here’s a super interesting.
0:07:21 Let’s say we’re now two and a half years into the ChatGPT moment.
0:07:24 That same group of CIO conversations, none of that.
0:07:29 It is basically assumed, it’s basically fully assumed that AI is going to take over the enterprise.
0:07:36 Like the CEO, the CEO, the CIO, the CDO, every job, every org leader is basically like, we know this is going to happen.
0:07:39 This is not like a, oh, we’re trying to kind of push it off.
0:07:42 It is purely a sequence of events.
0:07:43 Who do I deploy?
0:07:44 How do I deploy it?
0:07:45 How do I drive the change management?
0:07:46 Is the model ready?
0:07:55 So what’s really interesting is I think the level of buy-in you have now in the enterprise is like five times greater than we had in the early days of cloud.
0:07:57 And you can even see it.
0:08:04 Like to me, the classic litmus test was, if you remember like 15 years ago, I think Jamie Dimon was probably most famous for saying like, we’re never going to go to the cloud.
0:08:04 Yes.
0:08:07 So like they basically said, J.P. Morgan will never go to the cloud.
0:08:07 Yeah.
0:08:19 Today, that equivalent commentary, whether I don’t have a perfect Jamie Dimon quote, but David Solomon at Goldman Sachs has given this anecdote of they can write now an SEC filing or an S1 for an IPO in like a few minutes.
0:08:21 That used to take a number of analysts a few days.
0:08:28 And so the fact that like those are the anecdotes already coming out of the biggest banks means that we’re not in like a fear of AI world.
0:08:36 We’re in a, we know this is going to happen and it needs to happen to us faster than it happens to our competitors, which is a totally different dynamic than we saw with cloud.
0:08:42 So do you think this has implications for companies today that are building products that are pre-AI products?
0:08:49 So for example, with the cloud wave, you basically had a bunch of cloud native companies that ended up taking over.
0:08:54 So for example, Snowflake is a great example of this, which is like the ones that decided not to go all in and were hybrid.
0:08:57 Like hybrid kind of became known as like means it won’t work.
0:08:59 You know, anything called hybrid hasn’t worked.
0:09:09 So do you think because the buyer and the enterprise is more ready that like companies that are pre-AI have more of an opportunity?
0:09:13 Or do you think that you’re going to see the same thing with a lot of like AI native companies do well?
0:09:16 I’m going to basically give you the non-answer of I think both.
0:09:28 And one benefit that the cloud cohort has or the SaaS kind of posts like us all understanding and agreeing on what SaaS would look like, what we all have is whether we adhered to this perfectly or not is a question.
0:09:31 But we basically all tried to build API first platforms.
0:09:31 Yeah.
0:09:34 Or at least like API kind of like equal platform.
0:09:36 So we have the UI and we have the API.
0:09:42 And if you think about it, like AI and AI agents are like the perfect consumers of an API, right?
0:09:46 And so they basically become these super users within your system on your APIs.
0:10:01 So if I had to just say, okay, I want to deploy agents to go and automate my ServiceNow workflows, I think I’m better off just deploying the ServiceNow agent to go do that than do an entire reinvention of my ITSM system to solve that use case.
0:10:02 And you can just go down the list.
0:10:11 Like Workday, if I want an AI agent to do some kind of HR-related task, I think I’m better off to just do that within Workday than I am building an entire new system.
0:10:14 So you have a bunch of different factors versus the pre-cloud days.
0:10:17 Like pre-cloud to post-cloud was an entire rewriting of your software.
0:10:19 You had to go from single-tenant to multi-tenant.
0:10:21 The scaling of the systems were totally different.
0:10:25 Even the functionality and application logic was different because like it should be real-time.
0:10:26 It should be collaborative.
0:10:30 It shouldn’t be as sort of async and batches as the on-prem systems were.
0:10:35 And so in a cloud world, it is a reinvention of the user experience and what you’re doing in the system.
0:10:36 And we should definitely get to that.
0:10:40 Well, I just want to make sure I tease this out because it’s actually a very interesting point.
0:10:47 So your claim is to go from pre-cloud to post-cloud, like that ripped through the entire stack all the way down to like the infrastructure, for example, like tenancy.
0:10:49 Like you have to rewrite everything.
0:10:55 And then what you’re saying about AI is more of a consumption layer thing, which is you just treat the existing systems as they are.
0:10:56 And then the AI becomes the consumption layer.
0:11:02 Do you think this is like a 1.5 step and like the 2.0 step kind of rips through the entire stack?
0:11:04 Okay, so let’s bookmark that one for one second.
0:11:09 But like if you do pure Clay Christensen sort of approach, you know, sustaining innovation, disruptive innovation.
0:11:14 Disruptive innovation is this thing that looks like so much harder, so different, so less profitable.
0:11:17 Sustaining is like, actually, no, I’d like to build that because it’s incremental.
0:11:19 It’s better for our business overall.
0:11:22 The on-prem guys had a disruptive innovation.
0:11:26 Everything about the business model of SaaS looked different, harder, stranger.
0:11:28 I don’t have the talent.
0:11:32 I’m running a service delivery operation as opposed to I ship you a CD-ROM with my code.
0:11:35 Everything about the finances and pricing model.
0:11:35 Yes, everything.
0:11:36 Everything, the business model, everything.
0:11:41 AI, again, with the bookmark being like the really big disruption that you could contemplate,
0:11:45 right now with AI, everything kind of looks like a sustaining innovation if you’re an incumbent,
0:11:49 which is like, instead of a user pressing the buttons in the application,
0:11:53 let’s have an agent run through the API and operate as if they were that user.
0:11:57 And so all of a sudden, for a lot of SaaS providers, this looks like a TAM expansion
0:12:02 because now, for the first time ever, I can actually deploy my software for use cases
0:12:06 where the customer didn’t have users on the other end before to do those things.
0:12:07 So I think you have a lot of TAM expansion.
0:12:09 Now, the good news for startups.
0:12:11 With one caveat, which maybe we’ve bookmarked and we’re going to get to,
0:12:12 but let me just say the one caveat.
0:12:17 The one caveat is you now have a component that has a very different COGS model
0:12:19 if you’re a software provider.
0:12:19 Yes.
0:12:22 And so like now, it’s almost like when we went from like on-prem to cloud,
0:12:24 we went from perpetual to recurring.
0:12:30 And it feels like with AI, you kind of have to go from recurring to usage-based just because.
0:12:30 Yeah.
0:12:33 So business model will shift for some of the use cases
0:12:37 because even if you look at the cursors, replets, windsrifts of the world,
0:12:40 there does seem to be this baseline seat price.
0:12:43 And then your consumption usage thing is sort of this add-on.
0:12:43 This overage sheet.
0:12:47 And so SaaS providers are kind of well-structured to be able to have that kind of dynamic.
0:12:48 Yeah.
0:12:51 If it was 100% usage and the user seat goes away,
0:12:53 I do agree that you have this, then you have a little bit of a business model crisis.
0:12:57 Oh, so you think, but right now, it’s not clear that that’s going to go all the way over.
0:13:00 Well, until the human literally is not a seat on the system,
0:13:04 I think you don’t remove the end user license as a component.
0:13:04 Okay.
0:13:06 But again, that could be like the much bigger disruption.
0:13:11 Now, just to fully lay out the market dynamics, I think SaaS incumbents,
0:13:14 especially you have a couple other idiosyncrasies right now versus the on-prem days.
0:13:17 Another idiosyncrasy is I would say like on the margin,
0:13:20 you tend to have founders still leading the SaaS companies.
0:13:20 100%.
0:13:22 And so we didn’t really have that in the on-prem world.
0:13:28 And so like Siebel already had three CEOs later and PeopleSoft already had multiple CEOs later.
0:13:30 So it was a different leadership structure in these organizations.
0:13:33 A lot of times you still have the founder around, they’re poking around,
0:13:33 they’re really into AI.
0:13:37 So there can be a more natural pivot of the company from the leadership standpoint.
0:13:39 So a bunch of different factors.
0:13:43 Now, to the benefit of startups, which is why I can hold both of these in my head,
0:13:48 which is I’m very bullish on the SaaS incumbent being the natural place for that AI agent
0:13:49 relative to that category.
0:13:53 I just think we have this incredible expansion of categories for the first time
0:13:55 that we haven’t seen in probably 15 years.
0:14:02 So the SaaS 1.Wave actually expanded the software universe where we had these new categories of
0:14:04 software that we didn’t expect before.
0:14:08 Like nobody would have predicted the confluence and the snowflakes in the pre-on-prem days.
0:14:11 We didn’t have all of these different cuts of how do you work with data?
0:14:12 How do you do this workflow?
0:14:16 Like lines of business didn’t have 15 different applications they got to use.
0:14:17 Post-SaaS, they did.
0:14:23 So for startups in the AI world, the equivalent of that is I think there’s a lot of categories now
0:14:29 where there is no actually software incumbent in that category where AI agents all of a sudden
0:14:31 let you go build software for that category.
0:14:33 Legal, healthcare, education, and so on.
0:14:36 So that’s definitely true on the consumer side, right?
0:14:39 If you look at the top use cases of open AI, it’s almost like the top of the pyramid of needs,
0:14:40 right?
0:14:42 It’s like creativity and fulfillment, et cetera.
0:14:46 And I think like number five is like professional coding, but everything above that is one of these.
0:14:48 So on the consumer, that’s very clear.
0:14:50 Is that clear on the enterprise side?
0:14:50 I absolutely think so.
0:14:57 I think if we did a snapshot 10 years ago of the size of the contract management market or
0:15:00 the legal document market, it’s like sub 2 billion.
0:15:01 I’m making up the numbers.
0:15:02 It could be plus or minus a billion.
0:15:08 Would you agree that in five years from now, the AI agent related spend on legal services
0:15:12 should be in the many, many billions to double digit billions?
0:15:13 Absolutely.
0:15:13 Okay.
0:15:13 No question.
0:15:18 So all of a sudden there’s like not these natural incumbents that were like, oh, we captured all
0:15:19 that market.
0:15:23 AI agents all of a sudden expands the size of the software related spend in that space.
0:15:28 So I can underwrite that for healthcare, legal, consulting services.
0:15:31 I think there’s entire areas of financial services.
0:15:33 Like we always think, oh, finance has been wired up for so many years.
0:15:36 No, banking, consumer banking has been wired up.
0:15:37 Trading has been wired up.
0:15:39 Investment banking never went digital.
0:15:41 Wealth management never went digital.
0:15:46 Like these were not categories where you ever had like major software platforms to help these
0:15:47 entire categories of the economy.
0:15:51 And the reason it was because the work was unstructured.
0:15:56 It’s very ad hoc, very dynamic, lots of unstructured data as opposed to stuff that goes into databases.
0:15:58 All of that is now ripe for AI.
0:16:02 And that will then largely be ripe for many startups because there won’t be a natural incumbent
0:16:03 in those spaces.
0:16:08 I mean, it’s so strange to me how many disruptions are happening all at the same time with AI,
0:16:08 right?
0:16:11 I mean, if you think about like everything you said, which is basically vertical SaaS or vertical
0:16:13 use cases, which a lot of that is actually human budget, right?
0:16:13 Yep.
0:16:14 That’s being disrupted.
0:16:17 There’s a bunch of new use cases that we never really thought about before, which is
0:16:21 like the creativity and I mean, who would have thought that 2D image would be some massive
0:16:21 market?
0:16:22 Yes.
0:16:23 But it’s a massive market, right?
0:16:27 It turns out, you know, I’ve been a programmer for 30 years, right?
0:16:30 And in that time, like software would disrupt other things.
0:16:30 Yeah.
0:16:33 Like we disrupt all of these things, but we never got disrupted.
0:16:34 We’re safe.
0:16:35 We’re screwing you guys.
0:16:38 But clearly now software is being disrupted, right?
0:16:40 For the first time like I’ve ever seen in 30 years.
0:16:46 And so do you think this level of disruption is something that existing companies will not?
0:16:48 Like maybe a more fine point.
0:16:49 You are a business leader right now.
0:16:50 You have to think about product.
0:16:52 You have to think about your organization.
0:16:53 Does it require you to have to think about too much?
0:16:55 Like how do you structure your company as well?
0:16:59 How do you structure your product or do you think this is actually all pretty manageable?
0:17:01 I think it’s your R&D product.
0:17:04 Like literally like I’m putting myself in your shoes, right?
0:17:04 Yeah, yeah, yeah.
0:17:05 Which is like your CEO.
0:17:06 Like your R&D is changing.
0:17:07 Yeah.
0:17:08 You’re like-
0:17:09 Every part of the stack is changing.
0:17:09 Like everything.
0:17:10 Yeah.
0:17:23 I think the reason that I’m probably frankly more distracted by what we’re building that I don’t have enough time to stress out about the actual organizational side because I’m stressed out enough about just literally like the actual pure like delivery of the product.
0:17:27 I think if I had a little bit more time, I’d get more stressed out about all the other change.
0:17:30 We are very much leaning into the idea of being AI first.
0:17:31 We have a twofer on this.
0:17:36 Like one, by being as AI first as possible, we’ll see the use cases that our product should go solve for customers.
0:17:37 So like check that box.
0:17:41 And then second is I’m just a believer of the efficiency productivity gains.
0:17:41 Yeah, yeah, sure.
0:17:44 And I do think it does change basically everything about work.
0:17:46 And there’s lots of these interesting examples of what it means.
0:17:51 So in the future, does the individual contributor basically become a manager of agents?
0:17:52 Yeah, yeah, yeah.
0:17:53 So that’s a totally different job.
0:17:54 Right.
0:17:54 Right?
0:18:03 Like my recent kind of go-to is just thinking about it as a lot of the productivity of your organization was rate limited by literally like how fast can somebody use a computer to do something?
0:18:03 Yeah.
0:18:06 To type an email, to write code, to generate a marketing asset.
0:18:06 Yeah.
0:18:10 When that’s no longer a limiter, how do these jobs begin to change?
0:18:18 And it’s like, okay, your job is now orchestration, integration of work, planning, task management, reviewing, auditing, and that will radically change work.
0:18:33 Interestingly, this probably behooves us to not over-rotate on transforming yet internally for any given company simply because the technology is changing so fast that like you probably wouldn’t want to snap the line right now, run your whole business on this technology.
0:18:35 Because in two years from now, it’s going to happen.
0:18:37 Because in two years from now, it’s going to be so much better.
0:18:45 And so I think progressively figuring out which workflows have high impact upside, getting it rolled out in a decentralized way so people can experiment.
0:18:47 Like I think you want to do a few of those kind of things first.
0:18:50 I mean, I can’t imagine a listener not knowing what Box does.
0:18:57 But just for completing this, maybe can you just talk to us very quickly about what Box does and how you’re thinking about how that dovetails with AI?
0:19:02 Yeah, so we started the company with a really simple premise, make it easy to access and share your files from anywhere.
0:19:06 And we pivoted about two years into the journey to focus on the enterprise market.
0:19:10 And the whole idea was enterprises are awash with all this unstructured data.
0:19:17 So corporate documents, research files, marketing assets, M&A documents, contracts, invoices, all of this.
0:19:22 And as companies move to the cloud and as they move to mobile, they need a way to access that information.
0:19:25 They need a way to collaborate securely on it.
0:19:28 They want to be able to integrate that data across different systems.
0:19:30 So we built a platform to help companies do that.
0:19:34 We have about 120,000 customers, about 65 or so percent of the Fortune 500.
0:19:40 And so what’s incredible right now is we’ve had this ongoing problem since the creation of the company,
0:19:46 which is with structured data, the stuff that goes into your database, you can query it, you can synthesize it,
0:19:49 you can calculate it, you can analyze it, your unstructured data, the stuff that we manage,
0:19:54 you create it, you share it, you look at it, and then you basically kind of get forgotten about.
0:19:57 Like it goes into some folder and you almost never see it again.
0:20:01 And maybe you kind of find it once every five years for some task you’re doing, but that’s about it.
0:20:06 And so most companies are sitting on most of their data being unstructured
0:20:11 and getting the least amount of value from it relative to their other structured data.
0:20:13 AI is basically the unlock.
0:20:17 So AI lets you finally say, okay, we can ask this data questions.
0:20:22 We can structure it so we can look at a contract, pull out the 10 most important fields.
0:20:25 Once we have all that data, we can analyze that information.
0:20:26 We can get insights from it.
0:20:31 And then you can start to do things like workflow automation that was never possible with your unstructured data.
0:20:36 So if I want to move a contract through an automatic process, I can’t do it if I don’t know what’s in the contract.
0:20:40 And the computer previously was not able to know what’s in the contract.
0:20:46 So for us, there’s just a huge unlock of now what you can finally do with your information and your content.
0:20:50 So we’re building an AI platform to handle all of the kind of plumbing user experience
0:20:53 to make then your content AI ready effectively.
0:20:57 I don’t want to be like too bullshitty and provocative, but I have to ask this.
0:20:58 Please.
0:21:00 I’ve been in enterprise software for a very long time.
0:21:05 A lot of the business model is predicated on the fact that building software is hard and takes a long time.
0:21:05 Yeah.
0:21:08 To what extent do you worry about that not being the truth going forward?
0:21:13 Do you think we enter like this time of bespoke software being upon us?
0:21:17 I’m bearish on the extreme version of the essence of that.
0:21:22 So the extreme version of that, if you imagine the polls of this, like the extreme on one poll,
0:21:24 basically all software is prepackaged.
0:21:26 It’s the Ford Model T.
0:21:28 It’s going to work only in one way.
0:21:29 Everybody uses the same thing.
0:21:30 Okay.
0:21:30 That’s not going to happen.
0:21:31 We get that.
0:21:34 The other extreme is like everything is just like homebrew.
0:21:37 You wake up in the morning, you utter something, you get your software for the day.
0:21:38 You get your software for that thing.
0:21:40 And then the next day you do it again and you change it.
0:21:40 Okay.
0:21:45 The downsides of that model of why basically I think it doesn’t work is I think if you
0:21:49 ask like the world population, you probably find that 90 plus percent just don’t care enough.
0:21:54 They just don’t care about the tabs on their software and the modules on their dashboard.
0:21:57 Like it’s like they want someone else to just be like, this is what you should look at in
0:21:58 the morning.
0:21:58 Yeah.
0:22:01 They don’t want to have to even prompt the AI to tell them what to look at.
0:22:01 Yeah.
0:22:05 So given that that’s basically guaranteed to be where 90% of the world, no matter how you
0:22:06 cut anything.
0:22:06 Yeah, that’s a great point.
0:22:11 That means that basically 90% of our software should largely be like, okay, you log into the
0:22:13 HR system and it just looks like an HR system.
0:22:19 And in fact, there’s another interesting dynamic, which is like over many years, our software
0:22:24 and our actual way that we operate companies, there’s this flywheel relationship between them.
0:22:29 And so the way we run our HR department is like not so different than the way Workday wants
0:22:31 us to run our HR department.
0:22:31 Yeah.
0:22:35 And it’s fine because that’s not the area that we’re going to have a lot of upside innovating
0:22:35 on.
0:22:39 And like the way that we do our ticket management from customer tickets is like the way that
0:22:41 Zendesk decided to do ticket management.
0:22:44 And that’s fine because that’s not the core IP of the company.
0:22:47 In a way, it solves an operational problem for you.
0:22:47 Yes.
0:22:48 You don’t have to figure it out.
0:22:48 Right.
0:22:50 And people miss that about software.
0:22:54 I don’t want to have to think about the workflow of an HR payroll process.
0:22:56 I just want the software to do that.
0:22:59 And so that’s what people are buying.
0:23:01 And so nobody wants to customize those things.
0:23:05 Now, again, given that we’re going to be in this world of many different outcomes playing
0:23:11 out, the reason I’m still bullish on Replit and Vibe coding is for a different category,
0:23:15 which is like I’m the IT person and I just have this crazy queue of tasks.
0:23:17 And then someone’s like, can you build a website for this thing?
0:23:21 Can you like code up some inventory random plugin for this product?
0:23:24 It’s like that now becomes 10 times easier.
0:23:27 So the new prototyping, scripting, the long tail of stuff that we want to do.
0:23:31 And that long tail is so long and people never get to any of those things in that long tail.
0:23:36 And so I could underwrite a 10x growth of the amount of custom software that gets written
0:23:41 and the fact that these core systems don’t go away because there’s just actually going
0:23:43 to be way more software in the world that gets created.
0:23:44 Let me pressure test this.
0:23:47 So like, okay, so I can imagine why it would be hard to rebuild Box because what you do
0:23:48 is actually hard.
0:23:49 This is core infrastructure.
0:23:50 You store data like that’s really important.
0:23:52 And so I don’t think you just Vibe code that away.
0:23:56 But from my perspective, a lot of SaaS apps just look like CRUD.
0:24:00 To me, CRUD is, I don’t know what the acronym stands for, but it’s basically you’re reading
0:24:02 and writing data from like a backend.
0:24:08 And so do you think that there is a world where the consumption layer evolves to just using AI
0:24:09 and this class of companies go away?
0:24:13 Or do you actually think, if I heard what you just said, that the durability of these companies
0:24:16 is that it basically teaches you what the workflow is?
0:24:17 Well, I’m still going to say the latter.
0:24:20 Now, I don’t know if you need to bleep it out, but if you want to share a couple examples
0:24:25 of who you put in the not hard CRUD layer, then we can parse that.
0:24:26 But up to you.
0:24:27 The not hard CRUD layer?
0:24:28 Yeah.
0:24:34 I mean, I would say most vertical SaaS companies I see, the technology is trivial.
0:24:35 Yeah.
0:24:36 But the understanding of the domain is not.
0:24:36 No, no, no.
0:24:37 This is what you said before.
0:24:38 This is what I want to present.
0:24:38 That’s the thing.
0:24:40 That’s actually a great insight.
0:24:44 I’ve always underestimated vertical SaaS relative to the outcome.
0:24:44 Yeah.
0:24:47 And 20 years into doing enterprise software, I’m just like, no longer going to underestimate
0:24:48 vertical SaaS.
0:24:49 It’s not about the technology.
0:24:52 It’s the fact that somebody else has figured out the business model that works.
0:24:57 Like they have 10 people from the pharma industry that is like sitting next to the engineer
0:25:00 being like, this is how you should do the clinical trial workflow.
0:25:02 And that becomes so much of the IP.
0:25:07 Now, that translates fine to agents, but I still would then bet on that vertical player
0:25:13 doing that as opposed to somebody prompts their way into ChattoBT to build a FDA compliance
0:25:13 agent.
0:25:19 I would still largely bet on complianceagent.ai to do that over the pure horizontal system
0:25:21 that has no particular domain kind of expertise for that.
0:25:26 And then I think the other thing, I still think that there’s a relationship between some
0:25:31 amount of GUI and the agent and the APIs, because again, like you don’t want to every day
0:25:34 of your life, go to a blank empty screen and say, what’s our revenue today?
0:25:37 You just want a dashboard at some point and it just shows you the revenue.
0:25:37 That’s right, of course.
0:25:39 It’s almost like cast queries in a way.
0:25:40 Like somebody has made the decision.
0:25:43 Yes, this is like a known way to solve this problem in the enterprise.
0:25:48 And so I think that’s why the theory of the full abstraction away from the interface and
0:25:49 it’s all an API call.
0:25:50 I don’t think that happens.
0:25:55 And so ironically, probably what will happen is in a couple of years from now, we will see
0:25:59 agents like rebuild entire webpages and dashboards.
0:26:01 And then we’re going to find ourselves like, wait, why are we having an age?
0:26:06 Why do I have to spend tokens to create a thing that is a config on a dashboard?
0:26:10 And we’ll just be back to where we started for some amount of software, which will mean
0:26:13 that basically like these things are going to live together.
0:26:14 Cool.
0:26:16 Let’s move from software to decision process.
0:26:21 So I won’t say the name of the company, but I just spoke with a very, very legit company,
0:26:22 household name.
0:26:23 It’s a private company, though.
0:26:24 It’s not a public company.
0:26:30 We’re at the board level for every decision they ask the AI for like basically more information
0:26:31 for the decision.
0:26:32 Okay.
0:26:36 And this has actually been great from like discussion fodder to be provocative.
0:26:42 And it also shows how like fundamentally unoriginal the board members are.
0:26:45 Like this founder was telling me, it’s literally better than half of my board members.
0:26:46 Right.
0:26:51 And so like, how much have you thought about bringing AIs in to like help with decision
0:26:52 process?
0:26:52 Yeah.
0:26:57 And by the way, I think the board is like low hanging fruit because boards tend to not have
0:26:58 a lot of context to the business.
0:27:00 And so the incidents are probably less anyways.
0:27:02 But is this something that you’ve thought about?
0:27:04 The board one is an interesting one.
0:27:05 So maybe we can unravel that one.
0:27:10 But like I already use it for, let’s say, our earnings calls where we’ll do a draft of the
0:27:11 initial earnings script.
0:27:16 And then, I mean, again, because BoxAI deals with unstructured data, I just load up the
0:27:21 earnings script and I’ll use a better model and say, give me 10 points that analysts are
0:27:22 going to ask about this.
0:27:23 And like, how would I improve the script?
0:27:25 And it just spits out a bunch of things.
0:27:25 And it’s…
0:27:26 And how good is it at predicting?
0:27:27 Oh, extremely good.
0:27:28 Oh, 100%.
0:27:29 But the thing is, that’s not surprising.
0:27:33 Like it has access to every public earnings call in history.
0:27:34 Yeah, yeah.
0:27:38 And like at the end of the day, analysts can only ask you like, tailwinds, headwinds, who’s
0:27:38 buying what?
0:27:41 It’s not because the analysts are smart or not smart.
0:27:44 It’s just like, those are the things that like you would try and deduce from an earnings
0:27:45 call, buying a stock.
0:27:47 And you wouldn’t have thought of these questions beforehand?
0:27:48 Or is it just like…
0:27:49 I think you’re doing…
0:27:50 On the margin, on the margin.
0:27:51 No, no, sorry.
0:27:56 So what I’m using is then the specific parts of the document that is missing the answers
0:27:57 to those questions.
0:27:59 So I can actually inject the answers into that.
0:28:03 Because like you’re typing out a thing and like, I forgot to give two case studies in
0:28:04 this section or whatever.
0:28:07 It’s a quick way to just do some analysis on something.
0:28:12 But yeah, I mean, so Bezos famously had this memo-oriented, essay-oriented kind of meeting
0:28:12 structure.
0:28:13 And we never did that.
0:28:16 But I was always fascinated by the companies that could do it.
0:28:19 And actually, we’re entering a world where probably you could just pull that off, right?
0:28:23 So imagine if, whether it’s a board meeting or product meeting, you just do a quick, deep
0:28:24 research essay on the topic.
0:28:29 Like, obviously, every meeting, every strategy meeting in history would be better off if
0:28:32 you probably had that as a starting asset to get everybody informed.
0:28:36 I think the argument against that would be, the reason Bezos said it is because it forced
0:28:39 people to think clearly about what they’re doing and writing it down.
0:28:42 So the exercise meant that people walking in the meeting had more context.
0:28:46 This would almost argue that they would have less context because something else did the
0:28:46 thinking.
0:28:47 Well, two things.
0:28:51 It was to make sure that the person doing the thing had the clarity to write it, for sure.
0:28:54 But it was also still to inform everybody else that didn’t do that work.
0:28:57 And so it certainly would have helped everybody else in the room.
0:28:59 And I’m not 100%.
0:29:03 I mean, we should do a full longitudinal analysis of like the people that wrote the essay.
0:29:04 Did they actually have the better products?
0:29:06 Or like, I mean, there’s some Amazon products I don’t like.
0:29:09 And so they obviously wrote an essay also for those.
0:29:12 So I don’t know the hit rate ultimately on the essay specifically as much as the idea of
0:29:14 like write down a strategy, think it through.
0:29:18 And so why not have an agent do 90% of the heavy lifting?
0:29:24 So a lot of my workflows are like, if I have a topic where like maybe the direct change
0:29:29 of my workflow on this front is the kind of thing that three years ago, I might sort
0:29:32 of lob over to the chief of staff and say, hey, can you like go research like the pricing
0:29:35 strategy of this ecosystem or something?
0:29:37 That’s just a deep research query now.
0:29:39 And then I’ll wake up and it all look at this thing.
0:29:44 But what that does is because now I’m not having to calculate that person’s time, their
0:29:45 tasks, their trade-offs.
0:29:51 I just do it for the most random things, which means like I’m expanding and exploring
0:29:54 way more spaces mentally than I would have before.
0:29:55 And these are the kind of parts.
0:29:59 And this is equally why I’m like actually more optimistic on the jobs front, because what we
0:30:04 do too many times with an AI is we like look at today’s way of working and we’re just like
0:30:06 AI will come in and take 30% of that.
0:30:06 And it’s like, no, no, no.
0:30:08 We’ll just do totally different things with AI.
0:30:12 I wouldn’t have researched that thing before when it was people required to research it
0:30:15 because that would have been an inane task to send to somebody.
0:30:15 Yeah, yeah, yeah.
0:30:17 One thing.
0:30:22 So when we run the numbers and by run the numbers, I mean look through how AI companies are doing,
0:30:23 where does the value accrue?
0:30:26 There’s basically one takeaway.
0:30:30 And that is like these markets are very large and growing very fast.
0:30:32 And value is kind of accruing at every layer.
0:30:35 Everything from like literally chips up to apps.
0:30:40 And so like the only real sin is zero-sum thinking to be like, oh, like the models are
0:30:43 not going to be defensible or whatever your zero-sum thinking is that just hasn’t proven
0:30:44 out.
0:30:47 Now, this has still largely been a consumer phenomenon.
0:30:48 So what I’ve been thinking about, and I don’t have an answer.
0:30:53 I’d love to hear your thought is, is when it comes to enterprise budgets, like you can’t
0:30:55 just create budget out of thin air.
0:30:57 So like you actually do have a limited resource.
0:31:02 And so as budgets get reallocated, to what extent do you think this is like zero-sum,
0:31:06 like the old budget gets robbed versus like budget accretive?
0:31:07 Or like, how do you think about that?
0:31:10 Because again, like where we’ve come from, that has not been an issue.
0:31:12 I think in the enterprise, it probably will be.
0:31:14 So it does have to come from somewhere.
0:31:14 It’s fully logical.
0:31:15 A couple of things.
0:31:16 Yeah.
0:31:20 A large number for startups can also be a very small number for a large corporation.
0:31:21 Yeah.
0:31:23 So you have that dynamic playing out.
0:31:28 I’ll make up random stats, but you could probably take a meaningful engineering team and
0:31:32 probably for the price of five of those engineers or 10 of those engineers, you could
0:31:34 probably pay for cursor licenses for the entire engineering team.
0:31:38 But this would argue that it’s actually coming out of headcount.
0:31:41 So here’s where the asterisk is.
0:31:43 There’s an infinite set of ways.
0:31:46 This is why like you can never take a point in time snapshot on these kinds of things.
0:31:46 Yeah, totally.
0:31:48 There’s an infinite set of ways that this actually plays out.
0:31:49 Yeah.
0:31:56 Next year’s planning process, maybe in a perfectly like parallel universe, the salary increase
0:31:59 that year would have been 3.5% for employees.
0:32:02 And this coming year, it’s 3% because we’re going to take 0.5% and we’re going to deploy
0:32:04 AI for the company.
0:32:09 Or maybe next year, we would have added 50 engineers, but we’re going to add 25 and then pay for AI.
0:32:10 But guess what?
0:32:14 The year after, we’re going to have engineering productivity gains.
0:32:16 So it increases because it’s still a competitive environment.
0:32:21 We then now add engineers the year later because we’re getting higher productivity gains.
0:32:21 Yeah.
0:32:25 I think that most companies of any reasonable scale post 100 employees, let’s say, have
0:32:31 enough sort of dynamism in the financial model within a one to two year period where it doesn’t
0:32:33 look like what the economists would think it looks like.
0:32:34 Can I just spit this back?
0:32:37 Because I think this is actually a very good point that’s buried in there.
0:32:42 I just want to make sure I’m following along, which is the software license cost to a startup
0:32:46 relative to like a large people organization is relatively small.
0:32:50 It’s just a couple of headcount, which if you just look like normal performance management,
0:32:55 normal attrition, normal like variability, and even like hiring timelines is kind of in
0:32:56 the noise.
0:32:59 And so you already have an annual budgeting cycle to fix that up.
0:33:03 And so like basically within the noise, even of just like headcount planning, all of this
0:33:05 could work out without some massive disruption.
0:33:06 And I think that’s such a cool point.
0:33:10 And there could be an upper limit of this point, but let’s say the going rate in Silicon Valley
0:33:15 of a new engineer coming out of college, let’s just say it’s somewhere between 125 and 200.
0:33:16 Okay.
0:33:17 I’m just making up.
0:33:17 Okay.
0:33:17 Yeah.
0:33:22 Let’s say your most aggressive cursor usage or something is like a thousand bucks a year,
0:33:22 2000 bucks a year.
0:33:25 So you’re at like 1% salary maybe.
0:33:26 Here’s the question.
0:33:28 Again, do this like crazy apples to apples thing.
0:33:33 If you went and recruited from Stanford right now and you said, okay, you Stanford grad have a
0:33:33 choice.
0:33:41 You can work at this company and get paid 125K with no AI, or you can get paid 123K with
0:33:42 full access to AI.
0:33:43 Which one are you going to do?
0:33:45 They would do the 123 all day long.
0:33:48 But even that, yeah, I mean like your argument is, which makes a lot of sense to me.
0:33:50 It’s kind of on the margin when it comes to like 20.
0:33:55 But just as a way of exploring like why these things are not the high order bit of the cost
0:33:55 increase on budgets.
0:33:56 Oh, I love that.
0:33:57 That’s great.
0:34:02 And I did one kind of late night sort of like modeling once, but I’m afraid to say all the
0:34:03 numbers here because I think they’re just going to be so wrong.
0:34:07 But I think it’s something on the order like five or six trillion in knowledge worker headcount
0:34:08 spend in the U.S.
0:34:08 Yeah.
0:34:11 Everybody says for developers, let’s say 40 million.
0:34:12 Let’s just say it’s 30 million.
0:34:12 Yeah.
0:34:14 Let’s say that the average is 100K or you’re at three trillion.
0:34:16 Man, these are just massive, massive numbers.
0:34:17 So it’s many trillion.
0:34:17 Yeah.
0:34:19 So you have many trillions of dollars.
0:34:23 If you take a couple percent of that or five percent of that, you’re already doubling the
0:34:24 entire sort of U.S.
0:34:25 enterprise software spend.
0:34:26 Yeah.
0:34:28 So you can just make it work within.
0:34:31 And this is why I don’t think that people will not make cuts because they have to pay for
0:34:31 AI.
0:34:33 They might make cuts for other reasons.
0:34:33 Sure.
0:34:37 But even in those cases, I think you’ll often have it be for myopic reasons temporarily.
0:34:38 Yeah.
0:34:42 And there’s enough flexibility to basically consume this and then actually like recap on
0:34:43 the productivity game.
0:34:43 Yeah.
0:34:43 I think that’s great.
0:34:49 I try and parse everything you say through the lens of like, where are you landing on AI
0:34:50 coding?
0:34:53 And you seem to have a very pragmatic view of where things actually are at.
0:34:54 Yeah.
0:34:55 Where are you landing right now?
0:34:57 Well, it’s been an evolving.
0:35:02 So I would say in the entire AI thing, the biggest surprise to me is how effective it is
0:35:02 at code.
0:35:08 And so my sense is, so I’m just going to say a couple of, I think, facts, and then we can
0:35:10 kind of back out what this means in aggregate.
0:35:16 Because I think one fact is, the reality is, I do think that AI helps better developers more
0:35:17 than not better developers.
0:35:22 And the reason is you just have to deal with and be able to like know what to ask for and
0:35:22 know how to deal with the outcome.
0:35:23 So I think that’s one.
0:35:27 Someone said it, I thought, beautifully, which was, this is a very good developer.
0:35:28 And this was on X.
0:35:29 I forgot who it was, but I thought it capsulated.
0:35:35 He’s like, you know, 90% of what I know, the value of it has gone to zero, but 10% has
0:35:37 tripled more than 10X or whatever it is.
0:35:38 They’ve got 100X.
0:35:40 And I think that’s exactly right.
0:35:46 I do think that for a lot of rote use cases, the AI can do it and it doesn’t need to be
0:35:47 double checked, right?
0:35:50 So there’s a lot, to your point, like things like prototyping, things like scripting.
0:35:55 And so I do think if you look at usage of like open AI, if you actually look at code
0:35:59 usage, it’s like the primary use is actually professional developers, which means it’s part
0:36:00 of a developer workflow.
0:36:07 And then probably the most controversial stance I have is, and this is probably like sunk cost
0:36:10 fallacy because I’ve been a programmer for, I mean, like my PhD is in computer science,
0:36:12 like, you know, so maybe this is sunk cost fallacy.
0:36:17 But I just don’t see a world where you get rid of formal programming languages just because
0:36:20 they arose out of natural languages for a reason.
0:36:24 Like we started with English and then we made programming languages so that we could formally
0:36:25 describe stuff.
0:36:27 And so it’d be kind of a regression to go back.
0:36:30 So I still think we’ll use languages and maybe they’ll change, maybe more like a scripting
0:36:35 language, but I think like the existing tool set will evolve, but it’ll still be a professional
0:36:35 developer.
0:36:38 Like I think we’ll still have developers, still have developer tools.
0:36:38 So that’s kind of where I am.
0:36:39 I would love to hear where you’re at.
0:36:42 If I’m fully, I’m on the exact same page.
0:36:46 The fun thing to me is how coding is just at the tip of the kind of iceberg.
0:36:51 It’s the best thing to first sort of experience agentic automation, but I think you’ll see
0:36:52 this in basically every other space.
0:36:59 But what’s so fun is just in a one-year shift, let’s say, of like the nature of the relationship
0:37:00 with the AI.
0:37:04 So if you think about the GitHub co-pilot moment was like, oh, this thing is incredible.
0:37:07 It’s going to type ahead and predict what I’m typing.
0:37:12 And then you’re basically using it to work 20 or 30% faster and which parts of it do you
0:37:13 take on or not.
0:37:19 And now the relationship is like totally different within, again, a year or two period where you’re
0:37:23 using cursor, Windsor for whatever, and the agent is generating this chunk of an output
0:37:25 and then you’re just reviewing it.
0:37:30 But what’s incredible is like none of your expertise is any less valuable in that review.
0:37:34 In fact, it’s probably even more important than ever before because in some cases, like
0:37:39 it’s just going to be like wrong 3% of the time and then you review it, but then you’re
0:37:41 literally doing 3x the amount of output.
0:37:47 And the nature of how that changes both programming, but just like, why not have that for basically
0:37:51 everything is sort of this new way that both software should work and then actually we will
0:37:56 work is like, you know, the big joke a year after Chat to BT is like, okay, this thing generates
0:37:59 a legal case and it’s like wrong 10% of the time.
0:38:00 And it’s like, well, actually, hold on.
0:38:04 If you think about what the new paradigm of work looks like, and it’s like such a weird inversion
0:38:07 of it used to be the AI was fixing your errors.
0:38:09 That’s what we thought the AI was going to be.
0:38:10 And it’s just like a total flip.
0:38:12 It’s like the human’s job is to fix the AI errors.
0:38:14 And that’s the new way that we are going to work.
0:38:15 Right.
0:38:16 So this begs a very obvious question.
0:38:17 I’m going to work up to the question.
0:38:22 So there is a great paper in NSDI from an MIT team, which basically says you can optimize
0:38:24 a running system with agents.
0:38:29 And the way they did it is they basically have a teacher agent and then like more junior
0:38:32 agents and then the more junior agents would go try a bunch of stuff.
0:38:37 And of course, they had much more knowledge of the literature than any single human being.
0:38:39 So they try all of different things to try it.
0:38:41 And then the one at the senior agent would say, oh, this is good.
0:38:42 This isn’t good.
0:38:44 And then once it optimized the system, they would do it.
0:38:48 And then, you know, like the human being is then kind of helping the teacher agent decide
0:38:52 what are the parameters, what is good, what is not good and provide high level direction.
0:38:52 Right.
0:38:58 And so you’re already starting to see cases where human beings are running multiple agents
0:39:02 and even that already is starting to have some kind of bifurcation, which one way to think
0:39:08 about it is in any R&D organization, of course, people start as like ICs, but then they very
0:39:11 quickly get interns and go into management.
0:39:12 And so maybe we’re just skipping that step.
0:39:17 So the obvious question is what happens to entry level engineers?
0:39:22 Like does this change how people get introduced to computer science, for example?
0:39:26 The cool thing is probably more people will even now get introduced to computer science
0:39:27 because you’ll be able to…
0:39:28 Anybody can learn.
0:39:29 Anybody can learn it.
0:39:33 And, you know, it’s been 25 years for me, but like in the early days of programming basic
0:39:37 applications or putting other websites, it was just extremely frustrating that you would
0:39:40 spend days and days being like, why does that thing not work?
0:39:40 Yeah, yeah, yeah.
0:39:43 And like I have very few resources of figuring out why the thing didn’t work.
0:39:47 It would have been a hundred times easier if I could have had an agent write the thing.
0:39:48 I would have learned 10 times faster.
0:39:49 Yeah, yeah.
0:39:53 Honestly, what you did is you was like, well, not 25 years ago, but 10 years ago, you go
0:39:53 to Stack Overflow.
0:39:55 And so it was like the slow version.
0:40:00 Yeah, but so think about how many people missed the window pre-Stack Overflow that got sort
0:40:03 of pushed out of the ecosystem because they’re just like, this is too frustrating.
0:40:07 And so you’re going to have a way bigger funnel at the top of people now learning programming
0:40:08 and computer science.
0:40:11 I think a similar percentage of people will fall out.
0:40:14 So it’s not like, again, you’re going to get a 10x increase in programmers because you
0:40:17 still have to enjoy it and you have to like solving problems and whatnot.
0:40:21 It’s going to change the nature of the incoming class of engineers that you hire.
0:40:25 They will literally not be able to code without AI assisting them.
0:40:30 And it’s not 100% obvious that’s a bad thing because assuming you have internet and the site
0:40:32 stays up, like we should have access to the agents.
0:40:37 So I think it’s mostly just like we have to adapt how we think about the role of an engineer
0:40:40 and what these tools are giving us in terms of the productivity gains.
0:40:45 Actually, I meet with a lot of larger, not tech-oriented companies as customers.
0:40:50 And generally, the thing I’m recommending is hire a bunch of these people because they’re
0:40:55 going to flip your company on its head of how much faster the organization can run.
0:40:56 So I do understand.
0:40:59 I want to be sympathetic to the job market for anybody coming out of college because I don’t
0:41:00 think it’s easy right now.
0:41:02 And it probably hasn’t been easy in a number of years.
0:41:10 But if you are graduating, the thing I would be selling any corporation some way or another
0:41:15 is that if you are AI native right now coming out of college, the amount you can teach a company
0:41:16 is unbelievable.
0:41:20 And then conversely, if you’re a company, you should be actually like prioritizing this talent
0:41:26 that is just like, why does it take you guys two weeks to research a market to enter?
0:41:29 I can do that in deep research and get an answer to you in 30 minutes.
0:41:32 They will be able to show companies way faster ways of working.
0:41:37 Do you think there’s any stumbling into problems this way, which is like you kind of adopt too quickly.
0:41:42 It’s like you get into a morass you can’t get out of or you think at this point it’s pretty clear
0:41:44 this stuff can be practically consumed.
0:41:46 What would the morass be that you’d get into?
0:41:50 You hire a bunch of vibe cutters and then they create something that nobody can maintain
0:41:51 and it’s really slow.
0:41:51 Oh, yeah, yeah, totally.
0:41:54 Which, by the way, I will say I have seen this.
0:41:54 Yeah, yeah, yeah.
0:41:56 Okay, you could easily overdo this whole thing.
0:42:01 So I think as with anything, like deploying these strategies in moderation
0:42:05 while we’re all collectively still getting the technology to work better and better
0:42:08 is super important and understanding the consequences of these systems.
0:42:12 So, yes, this is not like a moment to just have your whole company vibe code.
0:42:15 I will say one of my favorite things that I’m witnessing in the whole coding thing,
0:42:18 I don’t know, the point of this talk is kind of the AI and the enterprise generally,
0:42:20 but like the coding thing is just so salient,
0:42:23 is that a lot of these OG programmers that I’ve known for a long time
0:42:26 that are off-creating companies or CEOs of public companies like yourself
0:42:27 are all back to programming.
0:42:27 Yeah.
0:42:30 And then you talk to them, you know, many of them, like I code, you know,
0:42:33 most nights with Cursor just because it’s really enjoyable.
0:42:37 And the reason I didn’t code before is because I just couldn’t keep up with the fucking frameworks.
0:42:39 I’m like, dude, I don’t know how to install the fucking thing.
0:42:41 And what is this Python environment stuff?
0:42:45 And like, you’re literally learning bad design choices that somebody else just made up.
0:42:47 Like, they’re not fundamental to the laws of the universe.
0:42:49 They don’t make you any smarter.
0:42:51 It’s just like waste of brain space.
0:42:55 And so in some way, the AI just gets rid of this kind of crufty stuff
0:42:57 that you probably shouldn’t be wasting brain space on anyway.
0:43:01 The amount of frustration I have when I look through, let’s say, our product roadmap.
0:43:01 Yeah.
0:43:04 Let’s say pre-AI, although this still obviously happens
0:43:05 because we haven’t fully transformed everything about how we work.
0:43:08 But pre-AI, when you would see things like,
0:43:12 we have to upgrade the Python library in this particular product.
0:43:15 And it’s like three engineer, two quarters.
0:43:16 No, exactly.
0:43:20 And like, at the end of that project, zero customers will notice that we did something.
0:43:24 We resolved some fringe vulnerability that is not going to even happen.
0:43:26 But you have to do it because there’s some compliance thing
0:43:30 where you have to make sure you’re on the latest version, which is super important.
0:43:32 But like, the thing is never going to happen.
0:43:37 And all of a sudden, like, you are wasting hundreds of thousands of dollars of engineering time.
0:43:41 And the fact that like, that’s now like a codex task is just unbelievable.
0:43:45 And the amount of just now things that you can actually relieve your team to go and work on is incredible.
0:43:51 And the other big, like, boon for the economy, and this is again where the economists just totally miss this stuff,
0:43:56 is think about every small business on the planet, of which there’s millions, tens of millions, whatever,
0:44:01 that for the first time ever in history, they have access to resources that are somewhat approximate
0:44:03 to the resources of a large company.
0:44:06 Like, they can do any marketing campaign.
0:44:08 Did you see the NBA finals video from Kalshi?
0:44:09 No.
0:44:10 The VO3 video?
0:44:11 Oh, yeah, yeah, yeah, yeah, yeah.
0:44:15 Like, you can now put together an otherwise million dollar marketing video.
0:44:18 For a couple hundred bucks of tokens.
0:44:21 And that being applied to every domain in every service area,
0:44:24 I can run a campaign that translates into every language.
0:44:29 I can have this long tail of bugs that I never got around to automatically get solved.
0:44:34 I can have the analysis of a top tier consulting firm done for my particular business.
0:44:43 So for the people or companies that are resourceful and are creative and imaginative, the access to resources right now is just truly unprecedented.
0:44:48 What do you think is the best metric for anybody interested in tracking this stuff as far as, like, how fast it’s going?
0:44:49 Is it GDP?
0:44:51 Is it margin?
0:44:52 Is it top line?
0:44:53 Is it headcount growth?
0:44:54 Is it all the above?
0:44:55 Like, how do you measure it?
0:44:56 Yeah.
0:45:02 I mean, for us, so internally first, and then maybe we’ll spitball some macro solutions to this.
0:45:08 Internally, we’ve actually explicitly taken the stance that we want to use AI to increase the capacity and capability of the company.
0:45:10 So just do more.
0:45:12 For anything that you track, just make sure it happens fast.
0:45:13 Just do more.
0:45:14 Just, like, do more or do faster.
0:45:16 In a given time period, yeah.
0:45:21 And so that somewhat relieves the pressure from people that, like, this is about cost cutting.
0:45:21 Yeah, yeah, yeah.
0:45:23 It’s just like, no, no, just, like, do more right now.
0:45:24 Let’s figure out what works.
0:45:25 Some things won’t work.
0:45:26 We want experimentation.
0:45:28 So just use AI to do more.
0:45:29 Okay, so that’s us.
0:45:35 The way you should measure that then in a couple years from now is either the growth rate of the company should be faster.
0:45:36 Yeah.
0:45:45 Or the amount of things that we’re collectively doing should be more, and the only reason that wouldn’t show up in growth rate is that every other company also does more, and so that gets competed away.
0:45:45 Yeah.
0:45:50 Which is, like, also a very viable outcome, is this is just the new standard of running a business.
0:45:51 But there’s no shift in the equilibrium.
0:45:52 Right.
0:45:52 There’s no shift in the equilibrium.
0:45:54 You just have to do it.
0:46:00 And then the ultimate product of all of that is some other kind of metric of satisfaction of, like, our products get better.
0:46:02 It could be, like, consumer price index or something.
0:46:04 Yeah, but, like, did the iPhone show up in GDP?
0:46:07 I don’t know, but my life is better with the iPhone pre than without the iPhone.
0:46:08 I’m pretty sure it did.
0:46:09 Okay, yeah, fine.
0:46:15 So, but, like, it would ultimately then show up in, like, new cures to diseases, better health care.
0:46:20 I don’t know that the dollars would move around all that differently as much as just, like, life expectancy should go up.
0:46:22 Like, cost of housing should go down.
0:46:30 Like, weird metrics that productivity gains will then drive that the economists wouldn’t naturally associate to, like, enterprise software and AI.
0:46:36 By the way, this is where I am, which is, like, clearly there’s a disruption because marginal costs are going down on a bunch of stuff.
0:46:36 Yeah.
0:46:38 Like, writing code and language reasoning and whatever.
0:46:43 And, like, some companies will take advantage of that, but I don’t think, like, the fundamental equilibrium changes.
0:46:46 I think, to your point, I think we just do more tech, products get better faster.
0:46:47 Yeah.
0:46:51 We saw problems that we haven’t before, but, like, it’s not asymmetric that we’ve seen in other companies.
0:46:59 The way I kind of think about it is, you know, if we go back to, like, 1985 and we just looked at how everybody works, I think we would just be totally stunned by how slow everything is.
0:47:05 And just, like, how long did it take you to research a thing or analyze a market or create a campaign or whatever?
0:47:05 Yeah.
0:47:10 But, like, it just has now been baked into our human productivity that we just do all those things really fast now.
0:47:10 Yeah.
0:47:16 And so, in 10 years from now, when we all have AI agents running around, we will just look back to today and be, like, how did we function?
0:47:22 Like, you spent two weeks to decide, like, the message for the marketing campaign?
0:47:24 Like, how is that possible?
0:47:27 Like, what we do now is we run 50 experiments with AI agents.
0:47:28 They all come back with versions.
0:47:32 We look at them all together, and then we make a decision in an hour, and we move on.
0:47:34 That’s obviously, like, how work works.
0:47:36 And, like, that’s what we will be saying 10 years from now.
0:47:38 Do you think we’ll ever saturate the consumer?
0:47:44 I caveat this by saying this comes up every one of these inflection points, and so I just wanted to ask it again for the umpteenth time.
0:47:46 I think I’ll say yes, just because at some point, maybe.
0:47:50 But, like, my list of purely consumer demands has not gone down.
0:47:55 Like, healthcare is, like, a totally unmet need that I have.
0:48:01 I do not like to go to doctors or dentists or anybody because of just how hard it is to get scheduled.
0:48:03 I mean, buying a fucking car, man.
0:48:05 Like, there are so many things that, like, just need to be sold.
0:48:06 The cost of housing.
0:48:07 We clearly don’t have enough houses.
0:48:10 Like, where will AI, you know, drive that?
0:48:12 Okay, so, you know, maybe robotics would be then the play there.
0:48:17 But, like, I don’t think we’re anywhere close to consumer satisfaction or satisfying all needs of consumers.
0:48:25 Well, actually, I meant more, will things change so fast that, like, it saturates the ability to adopt new things?
0:48:27 I do think that that is certainly possible.
0:48:34 I think I track sort of, let’s say, my parents as a decent kind of proxy or even just, like, college friends that aren’t particularly in tech.
0:48:34 Yeah.
0:48:38 And they’re still, like, in their ChatGPT phase of adoption.
0:48:39 And they haven’t moved on from that.
0:48:40 They haven’t made a VO video yet.
0:48:45 They’re just, like, using ChatGPT to ask questions about life experiences they have.
0:48:50 And so, maybe, ironically, like, one of the problems was ChatGPT was so good.
0:48:58 If you, like, you know, imagine what people thought AI should be able to do for them, it already met, like, 80% of, like, where they would have sort of projected it.
0:48:58 Yeah, yeah.
0:49:01 But when we know, actually, no, it can probably still do 10 to 20x more.
0:49:02 Yeah.
0:49:05 But their needs are going to be satisfied for some time on those core use cases.
0:49:06 Yeah.
0:49:10 So, I think for, like, the most basic consumer query type things.
0:49:19 But then this is the opportunity for startups, which is, like, now AI will show up in sort of ways that maybe the person isn’t even, like, in the market for an AI thing.
0:49:21 They just want a better version of that category.
0:49:24 I was going to say, this could just, like, simply be another market constraint.
0:49:27 As soon as it saturates, you just make the product better given, like, the existing.
0:49:35 Then, if I could just get better healthcare, but I don’t need to think about that as an AI problem or not an AI problem, but AI will be behind the scenes delivering that.
0:49:35 Yeah.
0:49:37 Then I don’t think you’re saturated anytime soon.
0:49:39 Yeah, it’s just the consumption capacity just becomes another market constraint.
0:49:41 But there’s a ton of other ways that you can improve things.
0:49:42 100%.
0:49:42 That’s great.
0:49:42 Good.
0:49:44 I love that you’re so optimistic about it.
0:49:47 I am, I think, 98th percentile optimistic.
0:49:47 Same.
0:49:48 Good.
0:49:48 All right.
0:49:54 So, I think we’ve had a fairly pragmatic conversation about the current impacts and the near-term impacts.
0:50:00 If you do a longer view, can you dare to guess what things look like in five to ten years?
0:50:04 So, Sam Altman and Jack Altman had a podcast recently.
0:50:05 Yeah, yeah, yeah.
0:50:12 And I’m going to paraphrase probably in some wrong way, but they were going back and forth about how, like, we just got what we would have predicted as AGI five years ago.
0:50:14 And it’s just like, we use it.
0:50:16 And it’s, like, it’s now just built in.
0:50:17 The most anticlimactic.
0:50:18 Yeah, it’s anticlimactic.
0:50:19 Anti-anticlimactic.
0:50:24 And I think that’s my instinct for a lot of this is five years, ten years, whatever your number is.
0:50:34 And this is why I’m so optimistic on just society and jobs and all this stuff is I don’t think it’s the Terminator kind of crazy outcome scenario of we automate away everything.
0:50:48 I think the human capacity for wanting to solve new problems, for creating new products, for serving customers in new ways, for delivering better healthcare, to try and do scientific discovery, like, all of this stuff is just built in us.
0:50:48 Yeah.
0:50:49 And it will continue.
0:50:55 And AI is this kind of up-leveling of the tools that we use to do all those things.
0:50:59 And so I think the way we work will be totally different in five years or ten years.
0:50:59 Totally.
0:51:05 But you’re already seeing enough of probably what it will look like that I think it’s an extrapolation of that.
0:51:15 It’s when you want the marketing campaign done, you have a set of agents that go and create the assets and choose the markets and figure out the ad plan.
0:51:19 And then you have a few people review it and debate it and say, okay, let’s go in this direction instead.
0:51:21 And then you deploy it and you’re on to the next thing.
0:51:25 And so each company, their units of output grow.
0:51:29 As a result of that growth, we’re all still in competitive spaces.
0:51:33 So some of it gets competed out and others will keep growing faster than they would have before.
0:51:35 So they’ll hire more people and you’ll have new types of jobs.
0:51:38 Like we’ll have jobs for people just to manage agents.
0:51:40 And like you’ll have operations teams.
0:51:43 You know, Adam D’Angelo had this cool role that just kind of got announced.
0:51:44 Yeah, that was really cool.
0:51:49 Yeah, the role is to work with Adam at Quora and figure out which workflows can be automated with AI.
0:51:51 I think you’ll have a lot of those kind of functions.
0:51:56 But I think one of the exciting things about at least being in Silicon Valley or anybody kind of tuning in, being in this ecosystem,
0:52:00 is like we’re seeing the change happen faster here.
0:52:05 And it’s going to be five or ten years of this rolling out to the rest of the economy.
0:52:11 And so I think we will spend the next five years making the technology actually deliver on the things that we’re all collectively talking about
0:52:18 to make it more and more robust and the accuracy goes up and the costs go down and the workflows it can tie into are better.
0:52:20 And we will be working on that for quite some time.
0:52:29 And you think ultimately this leads to the biggest peace dividend of better products for users, better user experience?
0:52:31 Yeah, I think the software gets better.
0:52:33 Our healthcare gets better.
0:52:34 The life sciences discoveries increase.
0:52:36 I think it’s all a society net positive.
0:52:37 I love it.
0:52:42 Thanks for listening to the A16Z podcast.
0:52:48 If you enjoyed the episode, let us know by leaving a review at ratethispodcast.com slash A16Z.
0:52:50 We’ve got more great conversations coming your way.
0:52:51 See you next time.

0:00:03 Today, we’re doing something a little different.
0:00:05 We’re dropping an episode from Impact Theory,
0:00:07 a show hosted by Tom Bilyeu,
0:00:09 featuring a conversation with Ben Horowitz,
0:00:11 co-founder of Andreessen Horowitz.
0:00:13 Ben rarely does interviews like this,
0:00:16 and in this one, he goes deep on AI, on power,
0:00:18 on the future of work, and what it really means
0:00:20 to be a human in a world of intelligent machines.
0:00:24 He breaks down why AI is not life, not consciousness,
0:00:25 but something else entirely.
0:00:28 Why blockchain is critical to preserving trust and truth
0:00:29 in the age of deepfakes.
0:00:32 Why distribution may matter more than code,
0:00:34 and why history tells us this isn’t the end of jobs,
0:00:36 but the beginning of something new.
0:00:37 Let’s get into it.
0:00:42 This information is for educational purposes only
0:00:45 and is not a recommendation to buy, hold, or sell
0:00:47 any investment or financial product.
0:00:49 This podcast has been produced by a third party
0:00:51 and may include paid promotional advertisements,
0:00:53 other company references, and individuals
0:00:54 unaffiliated with A16Z.
0:00:57 Such advertisements, companies, and individuals
0:01:00 are not endorsed by AH Capital Management LLC,
0:01:02 A16Z, or any of its affiliates.
0:01:04 Information is from sources deemed reliable
0:01:05 on the date of publication,
0:01:08 but A16Z does not guarantee its accuracy.
0:01:14 Revolutions don’t always come with banners and protests.
0:01:18 Sometimes, the only shots fired are snippets of code.
0:01:20 This is one of those moments.
0:01:24 AI is the most disruptive force in history,
0:01:26 and it’s no longer a distant possibility.
0:01:28 It is here right now,
0:01:31 and it’s already changing the foundations of power
0:01:32 and the economy.
0:01:34 Few people have been as influential
0:01:36 in shaping the direction of AI
0:01:38 than mega-investor Ben Horowitz.
0:01:40 A pioneer in Silicon Valley,
0:01:42 he spent decades at the center
0:01:44 of every major technological disruption,
0:01:47 including standing up to the Biden administration’s
0:01:49 attempts to limit and control AI.
0:01:50 In today’s episode,
0:01:53 he lays out where AI is really taking us,
0:01:55 the forces that will define the next decade,
0:01:57 and how to position yourself
0:01:59 before it’s too late.
0:02:04 You are in an area making investments,
0:02:05 thinking about some of the things
0:02:07 that I think are the most consequential
0:02:08 in the world today
0:02:09 as it relates to innovation.
0:02:10 But along those lines,
0:02:13 you and Marc Andreessen are all in on AI,
0:02:15 but how do we make sure
0:02:16 that it benefits everyone
0:02:18 instead of making humans obsolete?
0:02:19 To begin with,
0:02:21 we have to just realize,
0:02:22 you know, what AI is
0:02:23 because I think that
0:02:24 because we called it
0:02:25 artificial intelligence,
0:02:26 you know,
0:02:29 our whole industry of technology
0:02:30 has a naming problem
0:02:31 in that we started,
0:02:32 you know,
0:02:34 by calling computer science,
0:02:35 computer science,
0:02:37 which everybody thought,
0:02:39 oh, that’s just like computers.
0:02:42 It’s like the science of a machine
0:02:44 as opposed to information theory
0:02:45 and what it really was.
0:02:47 And then in Web3 world,
0:02:48 which you’re familiar with,
0:02:49 we call it cryptocurrency,
0:02:51 which to normal people
0:02:52 means secret money.
0:02:54 But that’s not what it does.
0:02:55 That’s a good point.
0:02:56 And then I think
0:02:57 with artificial intelligence,
0:03:01 I think that’s also like a bad name
0:03:02 in a lot of ways
0:03:03 and that, you know,
0:03:04 look, the people
0:03:05 who work on in the field
0:03:08 call what they’re building models.
0:03:09 And I think that’s
0:03:11 a really accurate description
0:03:12 in the sense that
0:03:13 what we’re doing
0:03:15 is we’re kind of modeling
0:03:17 something we’ve always done,
0:03:18 which is we’re trying
0:03:19 to model the world
0:03:22 in a way that enables us
0:03:23 to predict it.
0:03:24 And then, you know,
0:03:25 we’ve built much more
0:03:26 sophisticated models
0:03:27 with this technology
0:03:29 than we could.
0:03:30 in the old days,
0:03:31 we had E equals MC squared,
0:03:33 which was like amazing,
0:03:35 but a relatively simple model.
0:03:38 Now we have models
0:03:39 with what,
0:03:41 like 600 billion variables
0:03:42 and this kind of thing.
0:03:43 And we can model like,
0:03:45 what’s the next word
0:03:46 that I should say
0:03:48 and that kind of thing.
0:03:51 So that’s amazing and powerful.
0:03:52 But I would say like,
0:03:53 we need to distinguish
0:03:54 the fact that
0:03:56 it’s a model
0:03:59 that is directed by us
0:04:00 to tell us things
0:04:01 about the world
0:04:02 and do things
0:04:03 on our behalf.
0:04:05 But it’s not a,
0:04:06 it’s not life.
0:04:08 It doesn’t have a free will
0:04:09 and these kinds of things.
0:04:11 So I think default,
0:04:13 you know,
0:04:13 we are the master
0:04:16 and it is the servant
0:04:18 as opposed to vice versa.
0:04:20 The question though
0:04:21 that you’re getting at,
0:04:21 which is, okay,
0:04:24 how do we not get obsoleted?
0:04:26 Like, why do we need us
0:04:27 if we’ve got these things
0:04:28 that can do all the jobs
0:04:29 that we currently do?
0:04:31 And I think, you know,
0:04:33 we’ve gone through that
0:04:34 in the past
0:04:36 and it’s been interesting, right?
0:04:39 So in I think 1750,
0:04:41 over 90% of the jobs
0:04:42 in the country
0:04:43 were agricultural.
0:04:45 And, you know,
0:04:46 there was a huge fight
0:04:47 and a group called
0:04:47 the Luddites
0:04:48 that fought the plow
0:04:49 and, you know,
0:04:50 some of these
0:04:53 newfangled inventions
0:04:54 that eventually,
0:04:55 by the way,
0:04:57 eliminated 97%
0:04:58 of the jobs
0:04:59 that were there.
0:05:01 But I think that
0:05:03 most people would say,
0:05:03 gee,
0:05:05 the life I have now
0:05:06 is better than
0:05:07 the life that I would have
0:05:08 had on the farm
0:05:10 where all I did
0:05:11 was farm
0:05:12 and nothing else
0:05:13 in life.
0:05:15 But, you know,
0:05:15 like,
0:05:16 and if you want to farm
0:05:16 still,
0:05:17 you can.
0:05:18 that is an option.
0:05:20 But most people
0:05:21 don’t take that option.
0:05:23 So the jobs
0:05:24 that we have now
0:05:25 will,
0:05:25 you know,
0:05:26 a lot of them
0:05:27 will go away
0:05:28 and will have,
0:05:29 but will likely
0:05:30 have new jobs.
0:05:30 I mean,
0:05:31 humans are pretty good
0:05:33 at figuring out
0:05:34 new things to do
0:05:36 and new things
0:05:36 to pursue
0:05:37 and so forth.
0:05:40 And, you know,
0:05:40 like including,
0:05:41 like,
0:05:42 going to Mars
0:05:42 and that kind of thing,
0:05:44 which obviously
0:05:45 isn’t a thing today
0:05:46 but could very well
0:05:47 be a thing tomorrow.
0:05:50 So I think that,
0:05:50 you know,
0:05:51 we have to stay creative
0:05:53 and keep dreaming
0:05:54 about, like,
0:05:55 a better future
0:05:56 and how to kind of
0:05:57 improve things for people.
0:05:57 But I think that,
0:05:58 you know,
0:05:58 particularly
0:06:01 for kind of the people
0:06:01 in the world
0:06:03 that are on the struggle bus
0:06:04 who are living
0:06:05 on a dollar a day
0:06:07 or, you know,
0:06:07 kind of subject
0:06:08 to all kinds of diseases
0:06:09 and so forth,
0:06:10 life is going to get,
0:06:10 like,
0:06:11 radically better for them.
0:06:13 When I look at the plow example
0:06:13 and the Luddites
0:06:14 fighting against it,
0:06:15 I think you’ll see
0:06:16 the same thing with AI.
0:06:17 You’re going to get people
0:06:19 that just completely reject it,
0:06:19 refuse to engage
0:06:21 in anything for sure.
0:06:24 But when I look at AI,
0:06:26 what I worry about
0:06:26 is that there will be
0:06:28 no refuge to go to,
0:06:30 meaning if you realize,
0:06:30 oh,
0:06:31 I can’t plow as well
0:06:33 as a plow
0:06:34 or a tractor
0:06:35 or now a combine,
0:06:36 there are still
0:06:37 a lot of other things
0:06:38 that technology
0:06:39 can’t do better than me.
0:06:40 Do you think
0:06:41 there’s an upper bound
0:06:42 to artificial intelligence,
0:06:43 bad name or not,
0:06:45 or do you think
0:06:46 that it keeps going
0:06:47 and it literally
0:06:49 becomes better than us
0:06:50 once it’s embodied
0:06:50 in robotics
0:06:52 at everything?
0:06:54 We are kind of limited
0:06:55 by the new ideas
0:06:56 that we have.
0:06:58 And artificial intelligence
0:06:58 is really,
0:06:59 by the way,
0:07:00 artificial human intelligence,
0:07:01 meaning,
0:07:03 right,
0:07:05 humans looked at the world,
0:07:06 humans figured out
0:07:07 what it was,
0:07:08 you know,
0:07:09 described it,
0:07:09 came up with these
0:07:11 concepts like trees
0:07:11 and,
0:07:12 you know,
0:07:14 air and all this stuff
0:07:16 that it’s not necessarily
0:07:16 real,
0:07:19 just how we decided
0:07:20 to structure the world.
0:07:21 And AI has learned
0:07:22 our structure.
0:07:23 Like,
0:07:24 they’ve learned language,
0:07:25 human language,
0:07:26 which is a version
0:07:27 of the universe
0:07:27 that is not
0:07:29 an accurate version
0:07:29 of the universe.
0:07:30 It’s just our version
0:07:31 of the universe.
0:07:32 But you’re going to
0:07:33 have to go deep on that.
0:07:34 I know my audience
0:07:35 is going to be like,
0:07:36 air seems pretty real
0:07:37 when you’re underwater.
0:07:38 What do you mean
0:07:39 that trees and air
0:07:40 are not necessarily real?
0:07:41 It’s,
0:07:41 look,
0:07:42 it’s a construction
0:07:43 that we made.
0:07:43 You know,
0:07:44 we decided
0:07:46 it is literally
0:07:48 the way humans
0:07:48 have interpreted
0:07:49 the world
0:07:50 in order for humans
0:07:51 to navigate it.
0:07:52 And,
0:07:53 you know,
0:07:55 as is language,
0:07:55 right?
0:07:56 Language isn’t
0:07:57 the universe
0:07:58 as it is.
0:07:59 Like,
0:08:00 if completely objective,
0:08:02 if you had an objective
0:08:02 look at,
0:08:03 you know,
0:08:05 the atoms and so forth
0:08:06 and how they were arranged
0:08:07 and whatnot,
0:08:09 you probably,
0:08:10 you know,
0:08:11 those descriptions
0:08:12 are lacking
0:08:13 in a lot of ways.
0:08:13 They’re not
0:08:14 completely accurate.
0:08:16 They don’t,
0:08:17 you know,
0:08:17 they certainly don’t
0:08:18 kind of predict
0:08:19 everything about
0:08:19 how the world works.
0:08:21 And so,
0:08:23 what machines have learned
0:08:23 or like what
0:08:25 artificial intelligence is
0:08:26 is an understanding
0:08:27 of our knowledge,
0:08:29 of the human knowledge.
0:08:31 so it’s taken
0:08:31 in our knowledge
0:08:32 of the universe
0:08:34 and then it is
0:08:36 kind of,
0:08:36 can refine that
0:08:37 and it can work on that
0:08:39 and it can derive things
0:08:40 from our knowledge,
0:08:42 our axiom set.
0:08:44 But it isn’t actually
0:08:45 observing the world
0:08:46 at this point
0:08:47 and figuring out
0:08:47 new stuff.
0:08:48 So,
0:08:48 you know,
0:08:49 at the very least,
0:08:50 you know,
0:08:52 humans still have to
0:08:55 discover new principles
0:08:56 of the universe
0:08:57 or kind of
0:08:57 interpret it
0:08:58 in a different way
0:08:59 or the machines
0:09:00 have to somehow
0:09:01 observe directly
0:09:02 the world
0:09:02 which they’re not
0:09:03 yet doing.
0:09:06 And so,
0:09:06 you know,
0:09:07 that’s a pretty big role
0:09:08 I would say
0:09:09 but then,
0:09:09 you know,
0:09:10 in addition,
0:09:11 you know,
0:09:11 we direct the world.
0:09:12 I think like Star Trek
0:09:13 is actually a pretty
0:09:14 good metaphor for that.
0:09:15 Like the Star Trek
0:09:16 computer was pretty
0:09:16 badass
0:09:18 but,
0:09:19 you know,
0:09:20 the people in Star Trek
0:09:21 were still like
0:09:22 flying around the universe
0:09:23 discovering new things
0:09:23 about it.
0:09:24 You know,
0:09:24 there was still
0:09:25 much to do
0:09:27 and I think that
0:09:27 it’s always
0:09:29 a little
0:09:31 kind of difficult
0:09:32 to figure out
0:09:33 what the new jobs
0:09:34 that get created
0:09:34 are.
0:09:35 And we’ve had
0:09:37 intelligence for a while,
0:09:37 right?
0:09:38 Like we’ve had
0:09:39 machines that could
0:09:40 do math way better
0:09:40 than us
0:09:42 and,
0:09:43 you know,
0:09:43 I mean,
0:09:44 I can remember
0:09:46 when I was in
0:09:47 junior high school
0:09:49 or junior high
0:09:49 put on a play
0:09:50 about like
0:09:51 how bad it was
0:09:51 that there were
0:09:52 calculators
0:09:52 because nobody
0:09:53 would know
0:09:53 how to do
0:09:54 arithmetic
0:09:54 and then
0:09:56 all the calculators
0:09:56 would break
0:09:57 and then we’d
0:09:57 be stuck.
0:09:58 We’d be trying
0:09:59 to like fly around
0:09:59 the universe
0:10:00 and rockets
0:10:01 but we wouldn’t
0:10:02 be able to do math
0:10:03 and the calculators
0:10:03 would be broken
0:10:04 and we’d be screwed.
0:10:06 So there is always
0:10:07 that fear
0:10:09 and,
0:10:09 you know,
0:10:10 we’ve had computers
0:10:11 that can play games
0:10:12 better than us.
0:10:12 We currently have
0:10:13 computers that can
0:10:14 drive better than us
0:10:15 and so forth.
0:10:16 So we have a lot
0:10:16 of intelligence
0:10:17 out there
0:10:19 but it hasn’t,
0:10:21 you know,
0:10:22 created like
0:10:23 this super dystopia,
0:10:24 you know,
0:10:25 in any degree.
0:10:25 It’s actually
0:10:26 made things better,
0:10:27 you know,
0:10:28 everywhere it’s appeared.
0:10:29 So I would expect
0:10:30 that to continue.
0:10:31 What do you think
0:10:32 though is the limiting
0:10:32 function?
0:10:34 So when I look
0:10:34 at AI,
0:10:36 I always say
0:10:37 unless we run
0:10:38 into an upper bound
0:10:39 where the computation
0:10:40 just can’t allow
0:10:41 the intelligence
0:10:42 to keep progressing,
0:10:45 it seems like
0:10:45 it will become
0:10:46 not only
0:10:47 generalized human
0:10:47 intelligence
0:10:48 and thusly be able
0:10:48 to do everything
0:10:49 that we can do,
0:10:50 it will become
0:10:52 embodied as robotics
0:10:53 and if,
0:10:54 I ran the math
0:10:54 on this once,
0:10:56 Einstein is roughly
0:10:58 2.4 times smarter
0:10:58 than someone
0:11:00 who is definitionally
0:11:00 a moron
0:11:02 and the gap
0:11:03 just between
0:11:04 those two
0:11:06 is so dramatic
0:11:07 the army won’t
0:11:08 even draft
0:11:09 somebody
0:11:10 that is a moron
0:11:11 at, you know,
0:11:11 whatever,
0:11:12 81 IQ
0:11:13 or whatever it is.
0:11:14 because it’s,
0:11:14 they create
0:11:15 more problems
0:11:16 than they solve
0:11:17 even just by being,
0:11:18 you know,
0:11:19 bullet fodder.
0:11:21 So do you think
0:11:22 there is something
0:11:23 that’s going to cause
0:11:25 that upper bound
0:11:27 or you have a belief
0:11:29 about the nature
0:11:30 of intelligence
0:11:31 that will keep
0:11:33 AI subservient
0:11:34 to us?
0:11:35 The smartest people
0:11:36 don’t rule the world,
0:11:37 you know,
0:11:38 Einstein wasn’t in charge
0:11:42 and, you know,
0:11:42 many of us
0:11:43 are like ruled
0:11:45 by our cats
0:11:47 and so,
0:11:48 like,
0:11:49 power and intelligence
0:11:50 don’t necessarily
0:11:51 go together,
0:11:51 particularly when
0:11:52 the intelligence
0:11:53 has no free will
0:11:54 or has no
0:11:55 desire to do
0:11:56 free will.
0:11:56 It doesn’t have will.
0:11:58 You know,
0:11:59 it is kind of a model
0:12:00 that’s computing things.
0:12:02 I think also,
0:12:02 you know,
0:12:03 the whole general
0:12:04 intelligence thing
0:12:06 is interesting
0:12:07 in that,
0:12:10 Waymo’s got a super
0:12:10 smart AI
0:12:11 that can drive a car
0:12:14 but that AI
0:12:14 that drives a car
0:12:16 doesn’t know English
0:12:18 and, you know,
0:12:19 isn’t, you know,
0:12:20 particularly good
0:12:21 at other tasks,
0:12:22 you know,
0:12:23 currently
0:12:24 and then the
0:12:26 chat GPT
0:12:27 can’t drive a car.
0:12:30 And so that’s,
0:12:31 you know,
0:12:33 how much things
0:12:35 generalize particularly
0:12:36 and if you look
0:12:36 at, well,
0:12:37 why is that?
0:12:38 A lot of it
0:12:39 actually has to do
0:12:40 with the long tail
0:12:41 of human behavior
0:12:43 where humans,
0:12:45 you know,
0:12:46 the distribution
0:12:47 of human behavior
0:12:47 is,
0:12:49 it’s fractal,
0:12:50 it’s mantle
0:12:51 broadion or whatever,
0:12:53 it’s not evenly
0:12:54 distributed at all
0:12:56 and so,
0:12:58 you know,
0:12:58 an AI
0:13:00 that kind of
0:13:01 captures all that
0:13:02 turns out to be,
0:13:04 you know,
0:13:05 we’re not so much
0:13:06 on the track,
0:13:07 we’re more on the track
0:13:08 for kind of
0:13:10 really great reasoning
0:13:12 over kind of
0:13:12 a set of axioms
0:13:14 that we came up with,
0:13:15 you know,
0:13:15 in say math
0:13:16 or physics
0:13:18 but not so much
0:13:21 you know,
0:13:21 kind of general
0:13:23 human intelligence
0:13:24 which is,
0:13:25 you know,
0:13:26 being able to
0:13:28 navigate other humans
0:13:30 and the world
0:13:30 in a way
0:13:32 that is productive
0:13:33 for us
0:13:33 is kind of a,
0:13:34 it’s a little bit
0:13:35 of a different
0:13:36 dimension of things.
0:13:37 You know,
0:13:37 yeah,
0:13:38 you can compare
0:13:40 the math capabilities
0:13:41 or the go
0:13:43 playing capabilities
0:13:43 or the driving
0:13:44 capabilities
0:13:46 or the IQ test
0:13:47 capabilities
0:13:48 of a computer
0:13:49 but that,
0:13:51 that’s not really
0:13:52 a human.
0:13:53 I think a human
0:13:54 is kind of different
0:13:55 in a fairly
0:13:56 fundamental way.
0:13:58 So what we end up
0:13:58 doing,
0:13:59 I think is going
0:14:00 to be different
0:14:00 than what we’re
0:14:01 doing today
0:14:02 just like what
0:14:02 we’re doing today
0:14:03 is very different
0:14:04 than what we did
0:14:04 a hundred years ago.
0:14:06 But,
0:14:07 you know,
0:14:08 the not having
0:14:09 a need for us,
0:14:10 I think that,
0:14:10 you know,
0:14:12 these AIs
0:14:13 are tools for us
0:14:15 to basically
0:14:16 navigate the world
0:14:17 and help us
0:14:18 solve problems
0:14:19 and do things
0:14:20 like,
0:14:21 you know,
0:14:22 everything from
0:14:24 prevent pandemics
0:14:25 to deal with
0:14:27 climate change
0:14:28 to that sort of thing,
0:14:29 to not kill each other
0:14:30 driving cars,
0:14:32 which we do a lot of.
0:14:34 You know,
0:14:35 hopefully it doesn’t,
0:14:36 you know,
0:14:38 create more wars,
0:14:38 hopefully it creates
0:14:39 less wars,
0:14:40 but we’ll see.
0:14:41 What I know
0:14:43 about the human brain
0:14:44 maybe tricking me
0:14:46 into painting a vision
0:14:46 of the future
0:14:47 that isn’t going
0:14:47 to come true,
0:14:49 let me put words
0:14:49 in your mouth
0:14:50 and you tell me
0:14:52 if they fit appropriately.
0:14:53 What I hear you saying
0:14:54 is something akin
0:14:55 to the way
0:14:56 that we’re approaching
0:14:57 artificial intelligence
0:14:57 right now,
0:14:58 let’s round it
0:15:00 to large language models,
0:15:02 that is going
0:15:04 to hit an upper bound
0:15:05 where it’s not able
0:15:07 to have insights
0:15:07 that a human
0:15:08 won’t already have,
0:15:10 that they are trapped
0:15:11 inside of the box
0:15:11 that we have created,
0:15:12 what you’re calling
0:15:13 the axioms
0:15:14 by which we navigate
0:15:14 the world.
0:15:15 They get trapped
0:15:16 inside that box
0:15:18 and thusly will never
0:15:19 be able to look
0:15:19 at the world
0:15:20 and go,
0:15:21 I’m not going
0:15:21 to predict
0:15:22 the next frame,
0:15:22 I’m going to render
0:15:23 the next frame
0:15:24 based on what I know
0:15:25 about physics.
0:15:27 And so water reacts
0:15:28 this way
0:15:29 in an earthbound
0:15:29 gravity system
0:15:30 and so it’s going
0:15:31 to splash like this
0:15:32 and it understands
0:15:32 liquid dynamics,
0:15:33 et cetera,
0:15:33 et cetera.
0:15:36 So is that accurate?
0:15:37 Are you saying
0:15:37 that it is trapped
0:15:38 inside of our box
0:15:39 and will never have?
0:15:40 It hasn’t demonstrated
0:15:41 that capability yet
0:15:43 so it hasn’t
0:15:44 walked up to a rock
0:15:45 and said this is a rock,
0:15:46 right?
0:15:47 We labeled it a rock
0:15:49 because that’s our structure.
0:15:50 But a rock
0:15:51 isn’t probably
0:15:54 a more intelligent being
0:15:55 would have called it
0:15:55 something else
0:15:57 or maybe the rock
0:15:58 is irrelevant
0:16:00 to how you actually
0:16:01 can navigate
0:16:02 the world safely.
0:16:05 and kind of figuring
0:16:06 those things out
0:16:07 or kind of adopting
0:16:08 to them
0:16:08 is just not something
0:16:11 that, you know,
0:16:12 it’s trained on our
0:16:14 rendition of the universe
0:16:16 in our kind of
0:16:19 literally like the way
0:16:20 we have described it
0:16:22 using language
0:16:22 that we invented.
0:16:26 and so it is
0:16:29 constrained a bit
0:16:30 to that in nature
0:16:31 currently.
0:16:33 You know,
0:16:34 that doesn’t mean
0:16:34 it’s not like
0:16:36 a massively useful tool
0:16:37 and can do things
0:16:38 and by the way
0:16:39 can derive new rules
0:16:40 from the old rules
0:16:41 that we’ve given it
0:16:42 for sure.
0:16:46 but, you know,
0:16:46 we’ll,
0:16:49 like I think
0:16:49 it’s a bit of a jump
0:16:50 to go,
0:16:51 you know,
0:16:52 it’s going to replace
0:16:53 us entirely
0:16:56 when the whole
0:16:57 discovery process
0:16:58 is something that we do
0:16:59 that it doesn’t do yet.
0:17:01 Okay.
0:17:02 The way that the human
0:17:03 mind is architected
0:17:04 is you have
0:17:05 competing regions
0:17:06 of the brain
0:17:06 like if you cut
0:17:07 the corpus callosum
0:17:08 the part that connects
0:17:09 the left and the right
0:17:09 hemisphere
0:17:10 you can get
0:17:12 two distinct personalities
0:17:13 one that is atheist
0:17:14 for instance
0:17:15 and one that
0:17:17 believes deeply in God
0:17:17 and they’ll argue
0:17:18 back and forth.
0:17:18 I mean,
0:17:19 this is in the same
0:17:20 human brain.
0:17:22 So that tells me
0:17:23 that what you have
0:17:24 is basically
0:17:25 regions of the brain
0:17:26 that get good
0:17:26 at a thing
0:17:28 and then they end up
0:17:28 coming together
0:17:29 to collaborate
0:17:30 and that is
0:17:31 sort of human intelligence
0:17:31 and I’ve heard you
0:17:32 talk about
0:17:32 there’s something
0:17:34 like 200 computers
0:17:36 inside of a single car
0:17:38 so if we already know
0:17:38 that you can
0:17:39 daisy chain
0:17:40 all of these
0:17:40 like it’s
0:17:41 it’s a very deep
0:17:42 knowledge about one thing
0:17:44 but as you daisy chain
0:17:45 then the intelligence
0:17:47 gets what I’ll call
0:17:48 more generalized
0:17:49 you don’t see that
0:17:50 as a flywheel
0:17:51 that is going to
0:17:52 keep going.
0:17:53 You know,
0:17:54 what we can compute
0:17:55 will get
0:17:57 better and better
0:17:57 and better
0:18:00 but having said that
0:18:00 you know,
0:18:01 that doesn’t
0:18:02 say that
0:18:05 like humans
0:18:06 one,
0:18:07 you know,
0:18:08 humans
0:18:09 built the machines
0:18:10 plug them in
0:18:11 give them the batteries
0:18:12 all these kinds
0:18:13 of things
0:18:15 and
0:18:17 you know,
0:18:18 and they’ve been
0:18:19 created to
0:18:20 fulfill our purposes
0:18:21 so
0:18:22 you know,
0:18:23 what it means
0:18:23 to be a human
0:18:24 will probably
0:18:25 will change
0:18:26 like it has been
0:18:26 changing
0:18:28 and kind of
0:18:29 how humans
0:18:29 live their life
0:18:30 will change
0:18:31 but humans
0:18:32 still find things
0:18:32 to do
0:18:32 I mean,
0:18:33 like it’s kind
0:18:33 of like
0:18:34 you know,
0:18:35 like a cheetah
0:18:35 has been able
0:18:36 to run faster
0:18:36 than a human
0:18:37 forever
0:18:38 but we
0:18:38 never watch
0:18:39 cheetahs race
0:18:40 we only watch
0:18:40 humans race
0:18:41 each other
0:18:43 you know,
0:18:44 computers have
0:18:44 played chess
0:18:44 better than
0:18:45 humans for a
0:18:45 long time
0:18:46 but nobody
0:18:47 watches computers
0:18:47 play chess
0:18:48 anymore
0:18:48 they watch
0:18:49 humans play
0:18:49 humans
0:18:49 and chess
0:18:50 is more
0:18:50 popular than
0:18:51 it’s ever
0:18:51 been
0:18:52 and so I
0:18:53 think we
0:18:53 have like
0:18:53 a keen
0:18:54 interest
0:18:54 in each
0:18:54 other
0:18:55 and
0:18:57 how that’s
0:18:57 going to
0:18:57 work
0:18:57 and these
0:18:58 will be
0:18:58 kind of
0:18:58 tools
0:18:59 to enhance
0:19:00 that whole
0:19:00 experience
0:19:01 for us
0:19:01 but I
0:19:01 think
0:19:02 it’s
0:19:03 you know
0:19:06 like a
0:19:07 world of
0:19:07 just
0:19:08 machines
0:19:09 seems
0:19:10 like
0:19:10 that
0:19:10 seems
0:19:11 like
0:19:11 really
0:19:11 unlikely
0:19:12 so
0:19:13 you’ve
0:19:13 got
0:19:13 people
0:19:14 like
0:19:15 Elon Musk
0:19:16 Sam
0:19:16 Altman
0:19:17 who have
0:19:17 both
0:19:18 expressed
0:19:18 deep
0:19:19 concerns
0:19:19 about
0:19:21 how
0:19:22 AI
0:19:24 may
0:19:24 in fact
0:19:25 make us
0:19:25 obsolete
0:19:25 Elon
0:19:26 has likened
0:19:27 he’s
0:19:27 certainly
0:19:27 become
0:19:28 fatalistic
0:19:29 but he
0:19:30 gave a
0:19:31 rant
0:19:31 that I
0:19:31 absolutely
0:19:32 love
0:19:32 that is
0:19:33 AI
0:19:33 is a
0:19:33 demon
0:19:34 summoning
0:19:34 circle
0:19:35 and
0:19:35 you’re
0:19:35 calling
0:19:36 forward
0:19:36 this demon
0:19:36 that you
0:19:37 were just
0:19:37 convinced
0:19:38 you’re going
0:19:38 to be
0:19:38 able to
0:19:38 control
0:19:39 and he
0:19:39 certainly
0:19:40 is not
0:19:40 so sure
0:19:40 and at
0:19:41 one point
0:19:41 and again
0:19:42 I’m fully
0:19:42 aware
0:19:43 that he’s
0:19:43 on his
0:19:44 fatalist
0:19:44 arc
0:19:44 and he’s
0:19:45 just
0:19:45 moving
0:19:45 forward
0:19:45 and he’s
0:19:46 building
0:19:46 as fast
0:19:46 as he
0:19:46 can
0:19:48 it’s
0:19:48 interesting
0:19:49 that both
0:19:50 of them
0:19:50 despite
0:19:51 saying
0:19:51 these
0:19:52 things
0:19:52 are
0:19:52 building
0:19:53 AI
0:19:53 as
0:19:54 fast
0:19:54 as
0:19:54 like
0:19:55 they’re
0:19:55 literally
0:19:56 in a race
0:19:56 with each
0:19:56 other
0:19:57 to see
0:19:57 who can
0:19:57 build it
0:19:58 faster
0:20:01 that they’re
0:20:01 warning
0:20:02 about
0:20:03 what do
0:20:03 you take
0:20:04 away
0:20:04 from that
0:20:04 is it
0:20:04 just
0:20:05 regulatory
0:20:05 capture
0:20:06 on both
0:20:06 of their
0:20:07 parts
0:20:07 is it
0:20:08 is Elon
0:20:09 being
0:20:09 sincere
0:20:10 not that I
0:20:10 need you
0:20:10 to mind
0:20:10 read him
0:20:11 but like
0:20:11 what do
0:20:12 you take
0:20:12 away
0:20:13 in
0:20:14 the fact
0:20:14 that
0:20:14 they’ve
0:20:15 both
0:20:15 warned
0:20:15 against
0:20:16 it
0:20:16 and
0:20:16 they’re
0:20:16 both
0:20:17 deploying
0:20:18 it
0:20:18 as
0:20:18 fast
0:20:18 as
0:20:18 they
0:20:18 can
0:20:19 yeah
0:20:20 it seems
0:20:20 fairly
0:20:21 contradictory
0:20:23 look I
0:20:24 think there
0:20:24 is
0:20:25 like I
0:20:26 won’t
0:20:26 question
0:20:27 either
0:20:27 their
0:20:28 sincerity
0:20:28 at some
0:20:28 degree
0:20:29 but I
0:20:29 do
0:20:29 think
0:20:29 there
0:20:30 are
0:20:32 many
0:20:32 reasons
0:20:33 to warn
0:20:33 about
0:20:33 it
0:20:33 but like
0:20:34 I also
0:20:34 think
0:20:35 that
0:20:36 you know
0:20:37 any
0:20:37 kind
0:20:38 of new
0:20:38 super
0:20:38 powerful
0:20:39 technology
0:20:40 you know
0:20:41 in a way
0:20:41 they’re right
0:20:42 to kind
0:20:42 of warn
0:20:42 about
0:20:42 like
0:20:43 okay
0:20:44 this
0:20:45 thing
0:20:46 if we
0:20:48 you know
0:20:48 if we
0:20:48 don’t
0:20:49 think
0:20:49 about
0:20:49 some
0:20:49 of
0:20:49 the
0:20:50 implications
0:20:50 of
0:20:50 it
0:20:51 could
0:20:51 get
0:20:51 dangerous
0:20:52 and I
0:20:52 think
0:20:52 that’s
0:20:52 a good
0:20:53 thing
0:20:53 like
0:20:53 every
0:20:53 technology
0:20:54 we’ve
0:20:54 ever
0:20:54 had
0:20:54 from
0:20:55 fire
0:20:55 to
0:20:57 you know
0:20:58 from
0:20:58 fire
0:20:59 to
0:20:59 automobiles
0:20:59 to
0:21:00 nuclear
0:21:00 to
0:21:01 AI
0:21:02 has
0:21:02 got
0:21:03 the
0:21:03 internet
0:21:05 has
0:21:05 got
0:21:05 downsides
0:21:06 to
0:21:06 it
0:21:06 they
0:21:06 all
0:21:06 have
0:21:07 downsides
0:21:08 and
0:21:08 the
0:21:09 more
0:21:09 powerful
0:21:09 the
0:21:10 more
0:21:10 kind
0:21:10 of
0:21:12 you know
0:21:12 kind
0:21:12 of
0:21:12 intriguing
0:21:13 the
0:21:13 downside
0:21:13 and
0:21:13 you know
0:21:14 maybe
0:21:14 like
0:21:15 you know
0:21:15 without
0:21:16 the
0:21:16 internet
0:21:16 we probably
0:21:17 would have
0:21:17 never gotten
0:21:18 to AI
0:21:20 and so
0:21:22 maybe that was the downside of the internet
0:21:24 that it led to AI or something like that you could argue
0:21:26 but I think
0:21:27 generally
0:21:33 we would take every technology we’ve invented and keep it because you know net net they’ve
0:21:38 been positive for humanity and for the world and and that’s generally and that’s
0:21:40 why I think they’re building it so fast because I think they know that
0:21:47 all right so anybody with a 17 year old right now is thinking oh my how where do I point
0:21:53 my kid what do I tell them to go study that’s future proof what can we learn about the way you guys are investing at
0:22:01 Andreessen Horowitz that would give somebody an inclination of what you think a 17 year old should be focused on now
0:22:10 yeah you know it’s really interesting I think one of the things what we’re saying in the kind of smartest young people that come out is
0:22:18 they spend a lot of time with AI learning everything they possibly can so I think you want to get very good at like high
0:22:19 curiosity
0:22:21 and
0:22:23 and then learning you know you have
0:22:26 available to you all of human knowledge
0:22:28 in something that will talk to you
0:22:33 and that’s you know that’s an incredible opportunity and I think that
0:22:36 anything you want to do in the world to make the world better
0:22:40 you now have the tools as an individual to do that in a way that
0:22:44 you know if you look at kind of what Thomas Edison had to do
0:22:47 in creating GE and like what that took and
0:22:51 and so forth you know it was a way higher bar to have an impact
0:22:53 whereas now I think
0:22:55 you know you can
0:22:57 very quickly
0:23:02 you know build something or do something that you know just pick a problem
0:23:07 it’s not like you know sometimes in this AI conversation the thing that we ignore is like
0:23:10 well what are the problems that we have in the world
0:23:10 well they
0:23:13 well we still have
0:23:14 cancer and diabetes
0:23:16 and sickle cell
0:23:17 and
0:23:18 and every disease
0:23:19 and we still have
0:23:20 the threat of pandemics
0:23:21 and we still have
0:23:22 climate change
0:23:23 and we still have
0:23:25 you know lots of people who are
0:23:26 starving to death
0:23:27 and we still have malaria
0:23:27 and
0:23:30 and so like pick a problem you want to solve
0:23:32 and now
0:23:34 you know
0:23:36 you have a huge helping hand in doing that
0:23:38 that nobody in the history of
0:23:40 the planet has ever had before
0:23:40 so
0:23:42 I think there’s
0:23:44 really great opportunities along those lines
0:23:47 so that
0:23:47 that would be my
0:23:48 you know
0:23:50 my
0:23:51 best advice
0:23:53 I think is to
0:23:55 to get really good with that
0:23:55 and
0:23:58 and look I think a lot of the things that we’ve learned
0:24:01 or that have been valuable skills traditionally
0:24:02 are going to change
0:24:03 so you really
0:24:04 you know again
0:24:06 want to be able to learn how to do anything
0:24:08 and
0:24:09 and I think that’s probably
0:24:11 going to be key
0:24:13 when I look at
0:24:13 the
0:24:14 things you were just talking about
0:24:15 that feels right
0:24:17 for people that have
0:24:17 the
0:24:18 the inclination
0:24:19 that have the
0:24:20 cognitive horsepower
0:24:22 to go and say
0:24:23 okay I’m going to leverage AI
0:24:25 to extend my capabilities
0:24:26 to tackle the biggest problems in the world
0:24:28 certainly right now in this moment
0:24:28 that
0:24:29 that is the
0:24:30 thrilling reality
0:24:31 that people should focus on
0:24:32 but then
0:24:34 I contrast that with
0:24:36 the deaths of despair
0:24:37 a young
0:24:38 among
0:24:39 largely
0:24:40 young men
0:24:43 we have this problem
0:24:43 in
0:24:44 call it
0:24:45 middle America
0:24:46 where
0:24:47 manufacturing jobs
0:24:48 have gone away
0:24:48 so for that
0:24:49 normal
0:24:50 just sort of everyday person
0:24:51 I want to have a trade
0:24:52 I want to go out into the world
0:24:53 and get something done
0:24:55 is AI going to be
0:24:56 useful to them
0:24:58 or are they going to
0:24:59 get replaced by
0:24:59 robotics
0:25:01 the truth of it is
0:25:02 is there’s only one
0:25:03 robot supply chain in the world
0:25:04 and that’s in China
0:25:05 and
0:25:07 you know
0:25:07 so
0:25:07 like
0:25:08 we all need a robot
0:25:09 supply chain
0:25:11 we need to
0:25:11 manufacture that
0:25:13 so I think there’s going to be
0:25:13 like a real
0:25:16 manufacturing opportunity
0:25:18 coming up
0:25:19 and it’ll be
0:25:20 a different kind of
0:25:20 manufacturing
0:25:22 certainly more will be
0:25:23 automated and so forth
0:25:24 but there will be a lot of
0:25:26 things to learn
0:25:26 in that field
0:25:28 that I think will be
0:25:29 you know
0:25:30 super interesting
0:25:31 and
0:25:33 and you know
0:25:34 likely very very good job
0:25:35 so I
0:25:36 ironically
0:25:37 I would say
0:25:38 like going into
0:25:39 manufacturing now
0:25:40 as a young man
0:25:41 and trying to
0:25:42 you know
0:25:43 kind of figure out
0:25:44 what that is
0:25:45 and get engaged in it
0:25:46 will probably lead to
0:25:47 you know
0:25:48 quite a
0:25:49 good career
0:25:50 you know
0:25:50 maybe in
0:25:52 creating
0:25:53 factories
0:25:53 have become
0:25:55 like insanely valuable
0:25:55 and
0:25:55 and
0:25:56 and kind of
0:25:57 and strategic
0:25:58 to the national
0:25:59 interests as well
0:26:00 that makes sense
0:26:01 so again
0:26:02 at the level of
0:26:03 the guy smart enough
0:26:04 to build the facility
0:26:04 yes
0:26:05 and I recently saw
0:26:06 a video of the
0:26:07 grocery store
0:26:08 of the future
0:26:08 where it is
0:26:10 a huge grid
0:26:11 inside of a
0:26:12 giant facility
0:26:13 and there’s just
0:26:14 like these
0:26:15 bots that
0:26:15 look kind of like
0:26:16 small shopping carts
0:26:17 and they’re just
0:26:18 grid patterning
0:26:19 across all the items
0:26:20 snatching up
0:26:20 whatever you order
0:26:22 so you order online
0:26:22 these things
0:26:23 grab all that stuff
0:26:24 and then they send
0:26:25 it off to you
0:26:26 so for the person
0:26:27 that’s savvy enough
0:26:28 to build that facility
0:26:29 yes
0:26:30 tremendous
0:26:31 but
0:26:33 what I think
0:26:33 I hear you saying
0:26:34 and correct me
0:26:34 if I’m wrong
0:26:35 is that
0:26:36 okay there are
0:26:37 two opportunities here
0:26:37 the opportunity
0:26:38 one is
0:26:39 if you’re the kind
0:26:39 of person
0:26:40 that can leverage
0:26:40 AI to build
0:26:41 that facility
0:26:42 massive opportunity
0:26:44 if you’re the kind
0:26:44 of person
0:26:44 that would traditionally
0:26:45 work at that
0:26:45 factory
0:26:46 something new
0:26:47 is coming
0:26:47 we know that
0:26:48 because looking
0:26:48 back at history
0:26:49 all these technologies
0:26:50 unleash things
0:26:51 we can’t yet see
0:26:52 and so I have faith
0:26:53 in the
0:26:54 we can’t yet see it
0:26:54 but it is coming
0:26:55 yeah
0:26:56 no for sure
0:26:56 look I mean
0:26:58 you know what
0:26:58 like the biggest
0:27:00 in-demand job
0:27:00 in the world
0:27:01 is right now
0:27:02 data labelers
0:27:04 and
0:27:05 like data labeling
0:27:06 wasn’t a job
0:27:08 not long ago
0:27:09 but if you talk
0:27:10 I’ve even heard of this
0:27:11 what is data labeling
0:27:12 yeah
0:27:13 so it’s what
0:27:13 Alex said
0:27:14 it’s what scale
0:27:15 AI does
0:27:16 you know
0:27:16 they pay
0:27:18 armies and armies
0:27:19 and armies of people
0:27:20 to label data
0:27:21 so say hey
0:27:23 this is a plant
0:27:24 or this is
0:27:24 you know
0:27:24 a fig
0:27:25 or whatever it is
0:27:26 for the AI
0:27:28 to then understand it
0:27:29 and then you know
0:27:30 now with the
0:27:32 you know
0:27:33 with the kind of
0:27:34 reinforcement learning
0:27:35 coming back
0:27:36 into play
0:27:37 you know
0:27:38 labeling
0:27:38 you know
0:27:39 that kind of
0:27:40 supervised learning
0:27:41 is still like
0:27:42 very very very important
0:27:44 and I think that
0:27:46 you know
0:27:47 right now
0:27:48 like he’s got
0:27:50 unlimited hiring demand
0:27:51 which is
0:27:51 you know
0:27:52 ironic
0:27:53 for scale AI
0:27:54 to have
0:27:55 unlimited need
0:27:56 for humans
0:27:57 and I think
0:27:58 you know
0:27:59 in manufacturing
0:28:00 there are going to be
0:28:01 jobs like that
0:28:02 and there will be
0:28:02 the kind of physical
0:28:05 well when you go
0:28:06 and you go
0:28:07 into these robot
0:28:08 like the software
0:28:09 companies that are
0:28:10 doing robotics
0:28:12 they have people
0:28:14 managing the robots
0:28:15 right
0:28:15 like they’re training
0:28:16 the robots
0:28:17 humans train robots
0:28:19 to do all kinds
0:28:20 of things
0:28:21 and it turns out
0:28:21 that like
0:28:23 folding clothes
0:28:25 doesn’t necessarily
0:28:27 generalize
0:28:28 to making eggs
0:28:30 they’re like
0:28:31 super different
0:28:32 for robots
0:28:34 and so you need
0:28:35 you know
0:28:36 these robots
0:28:36 trained in all
0:28:37 these kinds
0:28:37 of fields
0:28:38 and so forth
0:28:38 so I think
0:28:38 there’s
0:28:39 you know
0:28:40 there’s a whole
0:28:40 new class
0:28:41 of jobs
0:28:42 that are
0:28:42 a little bit
0:28:43 hard to
0:28:44 anticipate
0:28:46 you know
0:28:47 in advance
0:28:48 but I think
0:28:49 at least
0:28:49 for the next
0:28:50 10 years
0:28:50 I think
0:28:51 the number
0:28:52 of new
0:28:52 jobs
0:28:53 related
0:28:54 to making
0:28:54 these machines
0:28:55 smarter
0:28:56 is going
0:28:57 to increase
0:28:57 a lot
0:28:59 and then
0:29:00 after that
0:29:00 you know
0:29:00 like
0:29:02 I think
0:29:02 there will
0:29:03 be
0:29:04 there just
0:29:05 tend to be
0:29:05 like
0:29:06 throughout
0:29:07 history
0:29:08 so many
0:29:09 needs
0:29:09 for new
0:29:09 things
0:29:10 that we
0:29:10 never
0:29:11 anticipated
0:29:11 like well
0:29:12 I mean
0:29:12 you know
0:29:12 one of my
0:29:13 favorite
0:29:13 examples
0:29:13 is
0:29:14 okay
0:29:15 computers
0:29:17 are going
0:29:18 to kill
0:29:20 the typesetting
0:29:20 business
0:29:21 and they did
0:29:22 everybody knew
0:29:22 that
0:29:23 like that
0:29:23 was coming
0:29:24 nobody
0:29:26 nobody
0:29:26 said oh
0:29:26 and then
0:29:27 there’s
0:29:27 going to be
0:29:28 5 million
0:29:28 graphic design
0:29:29 jobs
0:29:29 that come
0:29:30 out of
0:29:30 the PC
0:29:31 like
0:29:32 nobody
0:29:33 not a
0:29:33 person
0:29:34 predicted
0:29:34 that
0:29:35 so
0:29:35 it’s
0:29:35 really
0:29:36 easy
0:29:36 to figure
0:29:36 out
0:29:36 which
0:29:37 jobs
0:29:37 are
0:29:37 going
0:29:37 to go
0:29:37 away
0:29:38 it’s
0:29:38 much
0:29:39 more
0:29:39 difficult
0:29:39 to
0:29:39 kind
0:29:40 of
0:29:40 figure
0:29:40 out
0:29:40 which
0:29:41 jobs
0:29:41 are
0:29:41 going
0:29:41 to
0:29:41 come
0:29:42 but
0:29:43 like
0:29:43 if
0:29:43 you
0:29:43 look
0:29:43 at
0:29:43 the
0:29:44 history
0:29:44 of
0:29:44 automation
0:29:46 which
0:29:46 is
0:29:46 kind
0:29:47 of
0:29:47 automated
0:29:47 away
0:29:48 everything
0:29:48 we
0:29:48 did
0:29:48 100
0:29:49 years
0:29:49 ago
0:29:50 there’s
0:29:54 and so
0:29:54 you go
0:29:56 okay
0:29:57 and then
0:29:57 you know
0:29:57 like
0:29:58 some of
0:29:58 the
0:29:58 employment
0:29:59 will be
0:30:00 much
0:30:00 more
0:30:01 I think
0:30:01 enjoyable
0:30:02 than the
0:30:02 old
0:30:02 employment
0:30:03 as well
0:30:04 as it
0:30:04 has been
0:30:05 you know
0:30:05 over time
0:30:06 and you
0:30:07 always talk
0:30:07 about
0:30:07 manufacturing
0:30:08 jobs
0:30:08 going
0:30:08 away
0:30:08 but
0:30:09 the
0:30:09 manufacturing
0:30:09 jobs
0:30:10 that
0:30:10 have
0:30:10 gone
0:30:10 away
0:30:10 have
0:30:11 been
0:30:11 the
0:30:11 most
0:30:11 mind
0:30:12 numbing
0:30:12 so
0:30:13 I
0:30:13 think
0:30:13 things
0:30:14 evolve
0:30:14 in
0:30:14 very
0:30:15 very
0:30:15 unpredictable
0:30:16 ways
0:30:17 and
0:30:18 you know
0:30:18 like
0:30:18 I
0:30:18 think
0:30:19 the
0:30:19 hope
0:30:19 is
0:30:19 that
0:30:20 you
0:30:20 know
0:30:20 the
0:30:20 world
0:30:21 just
0:30:21 gets
0:30:22 much
0:30:22 better
0:30:23 but
0:30:23 I’m
0:30:23 not
0:30:24 I’m
0:30:24 not
0:30:25 so
0:30:26 worried
0:30:26 about
0:30:27 kind
0:30:27 of
0:30:28 anticipating
0:30:28 all the
0:30:28 horror
0:30:29 that’s
0:30:29 going to
0:30:29 come
0:30:29 I mean
0:30:29 I think
0:30:30 the main
0:30:30 reason
0:30:30 we’re
0:30:31 making
0:30:31 these
0:30:31 things
0:30:31 is
0:30:32 you
0:30:32 know
0:30:33 the
0:30:33 ways
0:30:34 that
0:30:34 they’re
0:30:34 making
0:30:34 life
0:30:35 better
0:30:36 and
0:30:36 you know
0:30:37 just like
0:30:37 we finally
0:30:38 figured out
0:30:38 a way
0:30:39 for
0:30:39 everybody
0:30:40 like
0:30:40 we
0:30:40 already
0:30:41 have
0:30:42 in
0:30:42 our
0:30:43 hands
0:30:44 everybody
0:30:44 can
0:30:44 get
0:30:45 a
0:30:45 great
0:30:45 education
0:30:46 like
0:30:47 that
0:30:47 whole
0:30:49 inequality
0:30:50 of
0:30:50 access
0:30:50 to
0:30:51 education
0:30:51 is
0:30:52 like
0:30:52 literally
0:30:53 gone
0:30:53 right
0:30:53 now
0:30:54 which
0:30:55 is
0:30:56 pretty
0:30:56 amazing
0:30:57 I mean
0:30:57 it’s
0:30:57 certainly
0:30:58 huge
0:30:59 yeah
0:30:59 nothing
0:30:59 that I
0:31:00 ever
0:31:00 thought
0:31:00 I’d
0:31:00 see
0:31:01 so
0:31:02 hopefully
0:31:04 things
0:31:04 go well
0:31:05 the great
0:31:05 irony
0:31:05 it’s
0:31:06 so
0:31:06 crazy
0:31:07 I
0:31:07 don’t
0:31:07 think
0:31:08 anybody
0:31:08 anybody
0:31:09 saw that
0:31:09 coming
0:31:09 it was
0:31:09 always
0:31:10 going
0:31:10 to
0:31:10 be
0:31:10 it’s
0:31:10 going
0:31:10 to
0:31:10 go
0:31:10 for
0:31:10 the
0:31:11 drivers
0:31:11 it’s
0:31:11 going
0:31:11 to
0:31:12 go
0:31:12 for
0:31:12 all
0:31:12 those
0:31:12 hard
0:31:13 difficult
0:31:13 repetitive
0:31:14 tasks
0:31:14 yeah
0:31:14 it’s
0:31:15 been
0:31:15 very
0:31:16 fascinating
0:31:16 to
0:31:16 see
0:31:16 what
0:31:17 actually
0:31:17 is
0:31:17 in
0:31:18 danger
0:31:18 like
0:31:19 super
0:31:19 creative
0:31:20 jobs
0:31:20 very
0:31:21 much
0:31:21 in
0:31:21 danger
0:31:23 but
0:31:23 yeah
0:31:23 as
0:31:23 you
0:31:23 get
0:31:24 down
0:31:24 I
0:31:24 mean
0:31:24 look
0:31:24 I
0:31:25 think
0:31:25 I
0:31:25 believe
0:31:26 way
0:31:26 more
0:31:26 strongly
0:31:27 than
0:31:27 you
0:31:27 do
0:31:27 that
0:31:28 robots
0:31:28 are
0:31:28 just
0:31:28 going
0:31:28 to
0:31:28 get
0:31:29 better
0:31:29 and
0:31:29 better
0:31:29 and
0:31:29 better
0:31:30 and
0:31:30 better
0:31:31 but
0:31:32 that
0:31:32 could
0:31:32 be
0:31:32 that
0:31:33 I’m
0:31:33 not
0:31:33 as
0:31:33 close
0:31:33 to
0:31:33 the
0:31:34 problem
0:31:34 as
0:31:34 you
0:31:34 are
0:31:35 speaking
0:31:35 of
0:31:35 which
0:31:36 how
0:31:36 the
0:31:37 insights
0:31:37 that
0:31:37 you’ve
0:31:38 had
0:31:38 into
0:31:39 AI
0:31:39 how
0:31:39 are
0:31:39 they
0:31:40 informing
0:31:40 the
0:31:41 investments
0:31:41 that
0:31:41 you
0:31:41 guys
0:31:42 make
0:31:42 the
0:31:43 theory
0:31:43 is
0:31:43 there’s
0:31:44 this
0:31:44 one
0:31:44 like
0:31:44 super
0:31:45 intelligent
0:31:45 big
0:31:46 brain
0:31:46 that’s
0:31:46 going to
0:31:46 do
0:31:47 everything
0:31:47 the
0:31:48 reality
0:31:48 on the
0:31:48 ground
0:31:49 is
0:31:49 even
0:31:50 with
0:31:50 the
0:31:50 state
0:31:50 of
0:31:50 the
0:31:50 art
0:31:51 models
0:31:51 they’re
0:31:51 all
0:31:52 kind
0:31:52 of
0:31:52 good
0:31:53 at
0:31:53 slightly
0:31:53 different
0:31:54 things
0:31:54 right
0:31:54 like
0:31:54 you know
0:31:55 Anthropic
0:31:55 is
0:31:55 like
0:31:55 really
0:31:56 good
0:31:56 at
0:31:56 code
0:31:57 and
0:31:58 Grok
0:31:58 is
0:31:58 really
0:31:59 good
0:31:59 at
0:31:59 like
0:31:59 real
0:31:59 time
0:32:00 data
0:32:00 because
0:32:00 they’ve
0:32:00 got
0:32:00 the
0:32:01 Twitter
0:32:01 stuff
0:32:01 and
0:32:02 then
0:32:02 you
0:32:02 know
0:32:03 OpenAI
0:32:03 has
0:32:04 gotten
0:32:04 like
0:32:04 very
0:32:04 very
0:32:05 good
0:32:05 at
0:32:05 reasoning
0:32:07 so
0:32:09 with
0:32:09 all
0:32:09 of
0:32:10 them
0:32:10 are
0:32:10 doing
0:32:11 AGI
0:32:11 but
0:32:12 then
0:32:12 they’re
0:32:12 all
0:32:12 good
0:32:12 at
0:32:13 different
0:32:13 stuff
0:32:13 which
0:32:14 is
0:32:14 you know
0:32:15 from
0:32:15 an
0:32:15 investing
0:32:16 standpoint
0:32:16 it’s
0:32:16 very
0:32:16 good
0:32:17 to
0:32:17 know
0:32:17 that
0:32:19 because
0:32:20 if
0:32:20 something
0:32:20 like
0:32:20 that’s
0:32:21 not
0:32:21 winner
0:32:21 take
0:32:21 all
0:32:22 that’s
0:32:22 very
0:32:23 that
0:32:23 becomes
0:32:23 like
0:32:24 super
0:32:24 interesting
0:32:25 it
0:32:25 also
0:32:26 is
0:32:26 interesting
0:32:27 for
0:32:28 what
0:32:28 it
0:32:28 means
0:32:28 at
0:32:28 the
0:32:29 application
0:32:29 layer
0:32:29 because
0:32:30 if
0:32:30 the
0:32:32 infrastructure
0:32:32 products
0:32:33 aren’t
0:32:33 winner
0:32:34 take
0:32:34 all
0:32:34 and
0:32:34 then
0:32:35 the
0:32:35 other
0:32:35 thing
0:32:35 about
0:32:35 the
0:32:36 infrastructure
0:32:36 products
0:32:37 that’s
0:32:37 interesting
0:32:37 is
0:32:38 that
0:32:38 they’re
0:32:39 not
0:32:39 particularly
0:32:40 sticky
0:32:41 in the
0:32:42 way
0:32:42 that
0:32:42 kind
0:32:43 of
0:32:44 Microsoft
0:32:44 Windows
0:32:45 was
0:32:45 very
0:32:45 sticky
0:32:46 right
0:32:46 it was
0:32:46 sticky
0:32:47 you build
0:32:47 an application
0:32:48 on Windows
0:32:49 it doesn’t run
0:32:49 on other
0:32:50 stuff
0:32:50 you’ve got
0:32:51 to do a lot
0:32:51 of work
0:32:52 to move
0:32:52 it to
0:32:52 something
0:32:52 else
0:32:53 so you
0:32:53 get this
0:32:53 network
0:32:55 effect
0:32:55 with
0:32:56 developers
0:32:57 then you
0:32:57 go
0:32:58 okay
0:32:58 well
0:32:59 how does
0:32:59 that work
0:32:59 with
0:33:00 state-of-the-art
0:33:00 models
0:33:01 well people
0:33:01 build
0:33:01 applications
0:33:02 on these
0:33:02 things
0:33:03 but
0:33:03 guess
0:33:04 what
0:33:04 like
0:33:05 to move
0:33:05 your
0:33:05 application
0:33:05 to
0:33:06 DeepSeq
0:33:06 you didn’t
0:33:07 have to
0:33:08 they just
0:33:09 literally
0:33:09 took the
0:33:09 open
0:33:11 Python
0:33:11 API
0:33:12 and like
0:33:12 it runs
0:33:13 on DeepSeq
0:33:13 now
0:33:16 so
0:33:17 you know
0:33:17 that
0:33:17 that kind
0:33:18 of thing
0:33:19 really impacts
0:33:20 you know
0:33:20 how you
0:33:20 think about
0:33:21 investing
0:33:21 and like
0:33:22 what is
0:33:23 the value
0:33:23 of having
0:33:24 lead in
0:33:24 application
0:33:25 and then
0:33:25 you know
0:33:26 where is
0:33:26 the mode
0:33:26 going to
0:33:27 come from
0:33:28 and of
0:33:28 course
0:33:28 AI is
0:33:29 also
0:33:29 getting
0:33:29 like
0:33:30 the one
0:33:30 thing
0:33:30 it is
0:33:30 getting
0:33:31 amazingly
0:33:32 good at
0:33:32 is
0:33:32 writing
0:33:33 code
0:33:34 and
0:33:34 so
0:33:35 then
0:33:35 you know
0:33:35 how much
0:33:36 of a
0:33:36 lead
0:33:36 do you
0:33:37 have
0:33:38 in
0:33:38 the
0:33:38 code
0:33:39 itself
0:33:39 versus
0:33:41 you know
0:33:42 kind of
0:33:42 the other
0:33:42 traditional
0:33:43 things
0:33:43 you know
0:33:43 when I
0:33:44 started
0:33:44 in the
0:33:44 industry
0:33:45 the
0:33:46 sales
0:33:46 people
0:33:46 were
0:33:46 in
0:33:47 charge
0:33:48 they
0:33:48 were
0:33:48 kind
0:33:49 of
0:33:49 like
0:33:49 the
0:33:49 big
0:33:50 there’s
0:33:50 a
0:33:50 great
0:33:52 TV
0:33:52 show
0:33:52 called
0:33:53 Halt
0:33:53 and Catch
0:33:53 Fire
0:33:54 and
0:33:55 if you
0:33:55 watch it
0:33:56 like the
0:33:56 thing
0:33:57 that’s
0:33:57 really
0:33:58 stunning
0:33:58 if
0:33:58 you’re
0:33:59 you know
0:33:59 kind
0:33:59 of
0:34:00 coming
0:34:00 from
0:34:01 the
0:34:02 2010s
0:34:03 2020s
0:34:04 world
0:34:04 is
0:34:05 why
0:34:05 are
0:34:05 the
0:34:05 sales
0:34:06 people
0:34:06 so
0:34:06 powerful
0:34:08 but
0:34:09 they
0:34:09 were
0:34:09 the
0:34:09 most
0:34:09 powerful
0:34:10 in
0:34:10 those
0:34:10 days
0:34:11 and
0:34:11 it
0:34:11 was
0:34:12 because
0:34:12 you know
0:34:12 distribution
0:34:13 was the
0:34:14 most
0:34:14 difficult
0:34:14 thing
0:34:16 and
0:34:16 I
0:34:17 you know
0:34:17 I think
0:34:18 distribution
0:34:18 is going
0:34:19 to get
0:34:19 in
0:34:19 very
0:34:19 very
0:34:20 important
0:34:20 again
0:34:23 because
0:34:24 maintaining
0:34:24 a
0:34:25 technological
0:34:25 lead
0:34:25 is
0:34:26 a lot
0:34:26 harder
0:34:27 you know
0:34:28 when the
0:34:28 machine
0:34:29 is writing
0:34:29 the code
0:34:30 and writing
0:34:30 it very
0:34:30 fast
0:34:32 although
0:34:32 it’s not
0:34:33 all the way
0:34:34 where it
0:34:34 can build
0:34:35 like super
0:34:35 complex
0:34:36 systems
0:34:37 but there’s
0:34:37 you know
0:34:38 a bunch
0:34:39 of things
0:34:40 out now
0:34:40 you know
0:34:40 Replet’s
0:34:41 got a
0:34:41 great
0:34:42 product
0:34:42 for it
0:34:43 there’s
0:34:43 a company
0:34:43 called
0:34:44 Lovable
0:34:44 that’s got
0:34:44 one out
0:34:45 in Sweden
0:34:46 that just
0:34:46 builds you
0:34:47 an app
0:34:47 like if
0:34:47 you need
0:34:48 an app
0:34:48 for something
0:34:49 just say
0:34:49 build me
0:34:50 this app
0:34:51 and there
0:34:51 it is
0:34:52 yeah
0:34:53 another thing
0:34:53 in cursor
0:34:54 that you
0:34:54 can select
0:34:56 what model
0:34:56 you want
0:34:57 to use
0:34:58 for whatever
0:34:58 thing that
0:34:58 you’re about
0:34:59 to generate
0:34:59 so the
0:35:00 ability to
0:35:00 go oh
0:35:01 I want
0:35:01 10 of
0:35:01 these
0:35:02 things
0:35:02 I’m gonna
0:35:02 use this
0:35:03 one for
0:35:03 this kind
0:35:03 of code
0:35:04 this one
0:35:04 for that
0:35:04 kind
0:35:04 of code
0:35:05 it’s
0:35:06 really
0:35:06 fascinating
0:35:08 now the
0:35:08 Biden
0:35:08 administration
0:35:09 was super
0:35:09 hostile
0:35:10 towards tech
0:35:12 when you
0:35:13 look at
0:35:14 what’s going
0:35:14 on now
0:35:15 with the
0:35:15 changes
0:35:16 in
0:35:16 regulatory
0:35:18 what do
0:35:19 you think
0:35:19 about the
0:35:20 race
0:35:21 between
0:35:21 us
0:35:21 and
0:35:21 China
0:35:22 were we
0:35:22 headed
0:35:23 down a
0:35:23 dark
0:35:23 path
0:35:24 where
0:35:24 if
0:35:25 that
0:35:26 administration
0:35:26 had
0:35:27 stayed
0:35:27 with that
0:35:27 like we’re
0:35:28 going to have
0:35:28 one or two
0:35:29 companies
0:35:29 we’re going to
0:35:30 control them
0:35:30 that’s going
0:35:30 to be that
0:35:32 is it possible
0:35:32 we could have
0:35:33 lost that race
0:35:34 is that race
0:35:35 a figment
0:35:35 of my
0:35:36 imagination
0:35:36 is that
0:35:36 real
0:35:38 I think
0:35:38 there’s
0:35:40 multiple layers
0:35:41 to the
0:35:43 AI race
0:35:43 with China
0:35:44 and then
0:35:45 you know
0:35:45 the Biden
0:35:46 administration
0:35:47 was
0:35:48 kind of
0:35:48 hostile
0:35:49 in many
0:35:50 ways
0:35:50 but all
0:35:51 for kind
0:35:51 of a
0:35:51 central
0:35:52 reason
0:35:52 I think
0:35:54 so
0:35:55 you know
0:35:56 in AI
0:35:56 in particular
0:35:58 you know
0:35:58 when we
0:35:59 met
0:36:01 and I
0:36:01 should be
0:36:02 very specific
0:36:03 we did
0:36:03 meet with
0:36:04 Jake Sullivan
0:36:04 but he was
0:36:05 very good
0:36:05 about it
0:36:05 we met
0:36:06 with Gina
0:36:06 Raimondo
0:36:07 she was
0:36:07 very good
0:36:08 about it
0:36:09 but we
0:36:10 met with
0:36:10 the kind
0:36:11 of White
0:36:11 House
0:36:13 and
0:36:13 their
0:36:14 you know
0:36:15 their position
0:36:15 was
0:36:16 super
0:36:18 kind of
0:36:18 I would
0:36:18 say
0:36:19 ill-informed
0:36:20 so
0:36:21 they
0:36:21 basically
0:36:22 were
0:36:23 they
0:36:24 walked in
0:36:24 with this
0:36:25 idea
0:36:26 that like
0:36:26 we’ve got
0:36:27 a three-year
0:36:28 lead on
0:36:28 China
0:36:30 and
0:36:30 we have
0:36:31 to protect
0:36:31 that lead
0:36:33 and
0:36:34 there’s
0:36:34 no
0:36:36 and
0:36:36 therefore
0:36:37 we need
0:36:37 to shut
0:36:38 down
0:36:38 open
0:36:38 source
0:36:39 and
0:36:39 that
0:36:39 doesn’t
0:36:40 matter
0:36:41 to
0:36:41 you
0:36:42 guys
0:36:42 and
0:36:42 startups
0:36:42 because
0:36:43 startups
0:36:44 can’t
0:36:44 participate
0:36:45 in AI
0:36:45 anyway
0:36:47 because
0:36:47 they don’t
0:36:48 have enough
0:36:48 money
0:36:50 and
0:36:50 the only
0:36:51 companies
0:36:51 that are
0:36:52 going to
0:36:52 do
0:36:52 AI
0:36:53 are
0:36:54 going to
0:36:54 be
0:36:54 kind
0:36:54 of
0:36:55 ironically
0:36:56 the
0:36:57 two
0:36:57 startups
0:36:58 Anthropic
0:36:58 and Open
0:36:58 AI
0:36:59 that are
0:36:59 out
0:37:00 and then
0:37:00 the big
0:37:00 companies
0:37:00 Google
0:37:01 and so
0:37:01 forth
0:37:01 and
0:37:02 Microsoft
0:37:03 and so
0:37:03 we can
0:37:04 put a
0:37:05 huge
0:37:05 regulatory
0:37:06 barrier
0:37:06 on them
0:37:06 because
0:37:07 they have
0:37:07 the money
0:37:07 and the
0:37:08 people
0:37:08 to deal
0:37:08 with it
0:37:09 and then
0:37:10 that’ll
0:37:10 be
0:37:11 and you
0:37:11 know
0:37:12 in their
0:37:12 minds
0:37:12 I think
0:37:13 they actually
0:37:13 believe
0:37:14 that
0:37:15 that would
0:37:15 be how
0:37:16 we would
0:37:16 win
0:37:18 but of course
0:37:18 you know
0:37:19 in retrospect
0:37:20 that makes
0:37:20 no sense
0:37:21 and it
0:37:21 kind of
0:37:23 it damages
0:37:23 you know
0:37:23 if you look
0:37:24 at China
0:37:24 and what
0:37:24 China’s
0:37:25 great at
0:37:25 then this
0:37:26 goes to the
0:37:26 next thing
0:37:27 so there’s
0:37:27 how good
0:37:28 is your
0:37:28 AI
0:37:28 and then
0:37:30 how well
0:37:30 is it
0:37:31 integrated
0:37:31 into
0:37:33 your
0:37:33 military
0:37:34 and the way
0:37:34 the government
0:37:34 works
0:37:35 and so
0:37:35 forth
0:37:36 and I
0:37:36 think
0:37:36 that
0:37:37 China
0:37:37 being a
0:37:38 top-down
0:37:39 society
0:37:40 their strength
0:37:40 is
0:37:42 you know
0:37:43 that whatever
0:37:44 AI they have
0:37:44 they’re going
0:37:45 to integrate
0:37:45 into it’s
0:37:46 already all the
0:37:46 companies are
0:37:47 highly integrated
0:37:47 into the
0:37:48 government
0:37:49 so you know
0:37:50 they’re going
0:37:50 to be able
0:37:50 to deploy
0:37:51 that and
0:37:52 and we’re
0:37:52 going to
0:37:52 see it
0:37:53 in action
0:37:55 with their
0:37:55 military
0:37:56 very fast
0:37:57 I think
0:37:57 that the
0:37:58 advantage
0:37:58 of the
0:37:59 US
0:37:59 is like
0:38:00 we’re not
0:38:01 a top-down
0:38:01 society
0:38:02 we’re like
0:38:02 a wild
0:38:03 messy
0:38:03 society
0:38:04 but it
0:38:05 means that
0:38:05 all of our
0:38:06 smart people
0:38:07 can participate
0:38:08 in the field
0:38:10 and look
0:38:11 there’s more
0:38:11 to AI
0:38:12 than just
0:38:12 the big
0:38:12 models
0:38:13 as you said
0:38:13 like
0:38:14 you know
0:38:15 how important
0:38:15 is cursor
0:38:16 it’s really
0:38:17 important
0:38:18 if you’re
0:38:18 building stuff
0:38:19 so like
0:38:20 oh you want
0:38:20 to go
0:38:21 build
0:38:23 you know
0:38:24 the next
0:38:25 whatever
0:38:26 thing that
0:38:27 the CIA
0:38:27 needs
0:38:27 or the
0:38:28 NSA
0:38:28 needs
0:38:28 or this
0:38:29 and that
0:38:30 like
0:38:30 you’re
0:38:30 building
0:38:30 that
0:38:30 with
0:38:31 cursor
0:38:31 you’re
0:38:31 using
0:38:31 a state
0:38:32 of the
0:38:32 art
0:38:32 model
0:38:33 but like
0:38:33 if you
0:38:33 had
0:38:34 eliminated
0:38:34 if you
0:38:34 know
0:38:35 if
0:38:35 the
0:38:35 Biden
0:38:36 White
0:38:36 House
0:38:36 had
0:38:36 gotten
0:38:36 their
0:38:36 way
0:38:37 they’d
0:38:37 eliminate
0:38:37 things
0:38:38 like
0:38:38 cursor
0:38:38 they’d
0:38:38 eliminate
0:38:39 startups
0:38:39 being
0:38:40 able
0:38:40 to do
0:38:41 anything
0:38:41 in AI
0:38:43 and so
0:38:44 the advantages
0:38:45 that we have
0:38:45 that we
0:38:45 don’t just
0:38:46 have a model
0:38:47 we’ve got
0:38:47 all this
0:38:47 other stuff
0:38:48 that goes
0:38:48 with it
0:38:50 and then
0:38:50 we have
0:38:51 new ideas
0:38:51 on models
0:38:53 with new
0:38:53 algorithms
0:38:53 and this
0:38:54 and that
0:38:55 and that’s
0:38:55 what the
0:38:55 U.S.
0:38:56 is great
0:38:56 at
0:38:57 and I
0:38:57 think
0:38:57 what China
0:38:58 is great
0:38:58 at
0:38:58 is
0:39:01 by the
0:39:01 way
0:39:01 they’re
0:39:01 very good
0:39:01 at math
0:39:02 people
0:39:02 people
0:39:03 are good
0:39:03 at math
0:39:03 and AI
0:39:04 is math
0:39:04 so
0:39:04 their
0:39:06 models
0:39:06 are good
0:39:07 they also
0:39:07 have
0:39:08 a data
0:39:08 advantage
0:39:09 on us
0:39:10 where they
0:39:10 have access
0:39:10 to the
0:39:11 Chinese
0:39:11 internet
0:39:12 they have
0:39:12 access
0:39:12 to
0:39:13 copywritten
0:39:13 material
0:39:14 which
0:39:15 they do
0:39:15 not
0:39:16 have
0:39:16 the
0:39:17 same
0:39:17 difference
0:39:18 for it
0:39:18 that we
0:39:18 do
0:39:18 in the
0:39:19 U.S.
0:39:19 and so
0:39:19 they’re
0:39:19 able
0:39:20 to
0:39:20 kind
0:39:21 of
0:39:21 get
0:39:21 to
0:39:22 you know
0:39:23 if you
0:39:23 use
0:39:23 DeepSeq
0:39:24 you go
0:39:24 wow
0:39:24 DeepSeq
0:39:25 really is
0:39:25 a great
0:39:25 writer
0:39:26 compared
0:39:27 to a lot
0:39:27 of the
0:39:28 U.S.
0:39:28 models
0:39:28 why
0:39:28 is
0:39:29 that
0:39:29 well
0:39:29 they
0:39:30 train
0:39:30 on
0:39:30 bigger
0:39:30 data
0:39:31 set
0:39:31 than
0:39:31 we
0:39:31 do
0:39:32 and
0:39:33 that’s
0:39:33 amazing
0:39:34 so
0:39:35 it
0:39:35 really
0:39:36 you know
0:39:37 I think
0:39:37 what we
0:39:38 want
0:39:38 is
0:39:38 we
0:39:39 want
0:39:39 to
0:39:39 have
0:39:40 kind
0:39:40 of
0:39:40 world
0:39:41 class
0:39:41 first
0:39:41 class
0:39:41 AI
0:39:42 in
0:39:42 the
0:39:42 U.S.
0:39:42 and I
0:39:43 think
0:39:43 of it
0:39:43 less
0:39:43 as
0:39:44 you know
0:39:45 is it
0:39:45 ahead
0:39:45 of
0:39:45 China
0:39:46 is it
0:39:46 slightly
0:39:46 ahead
0:39:46 of
0:39:46 China
0:39:47 I
0:39:47 think
0:39:47 that
0:39:47 model
0:39:50 you
0:39:52 know
0:39:52 what
0:39:52 we’ve
0:39:52 seen
0:39:53 with our
0:39:53 own
0:39:53 state
0:39:53 of the
0:39:53 art
0:39:54 models
0:39:54 are
0:39:55 very
0:39:55 shallow
0:39:57 and I
0:39:57 think
0:39:57 that’ll
0:39:58 continue
0:39:58 as long
0:39:58 as we’re
0:39:59 able
0:39:59 and allowed
0:40:00 to build
0:40:00 AI
0:40:00 and then
0:40:01 economically
0:40:02 what you’d
0:40:02 like
0:40:02 is you’d
0:40:03 like
0:40:04 you know
0:40:04 to have
0:40:05 a vibrant
0:40:06 AI
0:40:06 ecosystem
0:40:07 coming out
0:40:07 of the
0:40:07 U.S.
0:40:08 so other
0:40:08 countries
0:40:09 you know
0:40:10 who aren’t
0:40:11 state-of-the-art
0:40:12 with this
0:40:12 stuff
0:40:13 adopt our
0:40:14 technology
0:40:14 and you know
0:40:15 we continue
0:40:16 to be strong
0:40:16 economically
0:40:18 as opposed
0:40:18 to everything
0:40:19 goes to
0:40:19 China
0:40:19 and that
0:40:20 was like
0:40:20 a big
0:40:21 big risk
0:40:21 with the
0:40:21 Biden
0:40:22 administration
0:40:22 I think
0:40:24 and you
0:40:24 know
0:40:25 which was
0:40:26 you know
0:40:26 what they
0:40:27 were doing
0:40:27 on AI
0:40:28 was
0:40:29 you know
0:40:30 tough
0:40:30 I would
0:40:30 say
0:40:30 what they
0:40:31 were doing
0:40:31 on kind
0:40:32 of fintech
0:40:32 and crypto
0:40:32 was even
0:40:33 tougher
0:40:35 in that
0:40:35 they were
0:40:36 just trying
0:40:36 to get
0:40:36 rid of the
0:40:37 industry
0:40:37 in its
0:40:38 entirety
0:40:38 you know
0:40:38 with AI
0:40:39 they were
0:40:39 trying to
0:40:42 I would
0:40:42 say they
0:40:43 were extremely
0:40:43 arrogant
0:40:44 in their
0:40:46 in what
0:40:46 they thought
0:40:47 their ability
0:40:47 was to
0:40:47 predict
0:40:48 the future
0:40:50 you know
0:40:50 Mark and I
0:40:51 were in
0:40:51 there
0:40:51 you know
0:40:52 like our
0:40:52 job is
0:40:52 to
0:40:53 predict
0:40:53 it
0:40:53 like this
0:40:53 is our
0:40:54 job to
0:40:54 invest in
0:40:55 the future
0:40:55 to predict
0:40:56 the future
0:40:57 and they
0:40:57 were saying
0:40:58 things that
0:40:59 like were
0:41:01 so arrogant
0:41:01 that we
0:41:02 would never
0:41:02 even think
0:41:03 to say
0:41:03 them even
0:41:03 if we
0:41:03 thought
0:41:03 them
0:41:04 because
0:41:04 we’re
0:41:04 like
0:41:05 we know
0:41:06 that we
0:41:06 don’t know
0:41:07 the future
0:41:07 like that
0:41:09 you know
0:41:09 it’s just
0:41:10 unknowable
0:41:11 there’s too
0:41:11 many moving
0:41:11 parts
0:41:12 and these
0:41:12 things are
0:41:12 really
0:41:13 complicated
0:41:15 all right
0:41:16 well speaking
0:41:16 of the future
0:41:18 fully accepting
0:41:18 that it is
0:41:19 very opaque
0:41:20 and very
0:41:20 difficult to
0:41:21 see
0:41:21 what would
0:41:21 you say
0:41:22 is the
0:41:22 most
0:41:23 controversial
0:41:23 view that
0:41:24 you hold
0:41:24 about the
0:41:25 future
0:41:26 if we
0:41:27 don’t
0:41:28 get to
0:41:29 world class
0:41:30 in crypto
0:41:30 we’re gonna
0:41:32 be you know
0:41:32 AI really has
0:41:34 the potential
0:41:34 to wreck
0:41:35 society
0:41:37 and what I
0:41:37 mean by that
0:41:38 is if you
0:41:39 think about
0:41:40 what is
0:41:42 obviously
0:41:43 clearly gonna
0:41:43 happen in an
0:41:44 AI world
0:41:44 is one
0:41:46 we’re not
0:41:46 gonna be able
0:41:46 to tell the
0:41:47 difference between
0:41:47 a human and
0:41:48 a robot
0:41:49 two
0:41:49 we’re not
0:41:50 gonna know
0:41:50 what’s
0:41:50 real or
0:41:51 fake
0:41:53 three
0:41:54 the level
0:41:55 of security
0:41:56 attacks
0:41:57 on big
0:41:58 central data
0:41:58 repositories
0:41:59 is gonna
0:41:59 get so
0:42:00 good
0:42:01 that
0:42:02 everybody’s
0:42:02 data is
0:42:03 going to
0:42:04 be out
0:42:04 there
0:42:05 and you
0:42:05 know
0:42:06 there is
0:42:06 no safe
0:42:07 haven for
0:42:07 a consumer
0:42:09 and then
0:42:10 finally
0:42:10 you know
0:42:10 for these
0:42:11 agents
0:42:12 and these
0:42:12 bots
0:42:13 actually be
0:42:13 useful
0:42:14 they
0:42:14 actually
0:42:15 need to
0:42:15 be able
0:42:15 to use
0:42:16 money
0:42:17 and pay
0:42:17 for stuff
0:42:18 and get
0:42:18 paid for
0:42:19 stuff
0:42:19 like so
0:42:22 and if you
0:42:22 think about
0:42:22 all those
0:42:23 problems
0:42:23 those are
0:42:24 problems
0:42:25 that are
0:42:25 by far
0:42:26 best solved
0:42:27 by kind
0:42:27 of blockchain
0:42:28 technology
0:42:30 so one
0:42:32 we absolutely
0:42:33 need
0:42:34 a public
0:42:34 key
0:42:34 infrastructure
0:42:36 such that
0:42:37 every citizen
0:42:38 has their
0:42:39 own wallet
0:42:39 with their
0:42:40 own data
0:42:40 with their
0:42:41 own
0:42:41 information
0:42:43 and if
0:42:44 you need
0:42:44 to get
0:42:45 credit
0:42:46 or prove
0:42:46 you’re a
0:42:46 citizen
0:42:47 or whatever
0:42:48 you can do
0:42:49 that with
0:42:49 the zero
0:42:49 knowledge
0:42:49 proof
0:42:50 you don’t
0:42:50 have to
0:42:50 hand over
0:42:51 your social
0:42:51 security
0:42:52 numbers
0:42:52 your bank
0:42:53 account
0:42:53 information
0:42:54 all this
0:42:55 kind of
0:42:55 thing
0:42:56 because
0:42:58 the AI
0:42:58 will get
0:42:59 it
0:43:00 so you
0:43:01 really need
0:43:01 your own
0:43:02 keys
0:43:02 and your
0:43:02 own
0:43:02 data
0:43:03 and there
0:43:03 can’t be
0:43:04 these
0:43:04 gigantic
0:43:06 massive
0:43:06 honeypots
0:43:07 of information
0:43:08 that people
0:43:08 can go
0:43:09 after
0:43:10 I think
0:43:10 that with
0:43:11 deep
0:43:11 fakes
0:43:12 if you
0:43:12 think
0:43:12 about
0:43:13 okay
0:43:14 we’re
0:43:14 going to
0:43:15 have to
0:43:15 be able
0:43:16 to
0:43:16 whitelist
0:43:17 things
0:43:17 we’re
0:43:17 going to
0:43:17 have to
0:43:17 be able
0:43:18 to say
0:43:18 what’s
0:43:18 real
0:43:19 but who
0:43:20 keeps
0:43:20 track
0:43:20 of what’s
0:43:21 true
0:43:21 then
0:43:21 is it
0:43:21 the
0:43:22 government
0:43:23 you know
0:43:24 please
0:43:24 Jesus
0:43:24 everybody
0:43:25 trust
0:43:25 everybody
0:43:25 trust
0:43:26 Trump
0:43:26 now
0:43:27 you know
0:43:27 everybody
0:43:27 trusted
0:43:28 Biden
0:43:28 is it
0:43:29 going to
0:43:29 be
0:43:29 Google
0:43:29 we
0:43:30 trust
0:43:30 those
0:43:30 guys
0:43:31 or is
0:43:31 it
0:43:31 going to
0:43:32 be
0:43:32 the
0:43:32 game
0:43:33 theoretic
0:43:33 mathematical
0:43:34 properties
0:43:34 of the
0:43:35 blockchain
0:43:35 that
0:43:36 can hold
0:43:36 that
0:43:39 and so
0:43:39 I think
0:43:39 that
0:43:40 you know
0:43:40 it’s
0:43:41 essential
0:43:42 that we
0:43:43 regenerate
0:43:43 our
0:43:44 kind of
0:43:45 blockchain
0:43:45 crypto
0:43:46 development
0:43:46 in the
0:43:46 US
0:43:47 and we
0:43:47 get
0:43:47 very
0:43:48 serious
0:43:48 about
0:43:48 it
0:43:49 and
0:43:49 you know
0:43:49 like
0:43:50 if the
0:43:50 government
0:43:50 were to
0:43:51 do
0:43:51 something
0:43:51 I think
0:43:51 it should
0:43:52 be to
0:43:52 start to
0:43:53 require
0:43:54 these
0:43:55 information
0:43:56 distribution
0:43:56 networks
0:43:57 these social
0:43:57 networks
0:43:58 to have
0:43:59 a way
0:43:59 to
0:44:00 you know
0:44:01 verifiably
0:44:01 prove
0:44:02 you’re human
0:44:02 you know
0:44:03 prove where
0:44:04 a piece
0:44:04 of data
0:44:05 came from
0:44:05 and so
0:44:05 forth
0:44:06 and I
0:44:07 think
0:44:07 that
0:44:08 you know
0:44:09 we have
0:44:09 to
0:44:09 you know
0:44:10 have
0:44:10 banks
0:44:12 start
0:44:12 accepting
0:44:13 zero
0:44:13 knowledge
0:44:14 proofs
0:44:15 and you
0:44:15 know
0:44:16 and that
0:44:16 be just
0:44:17 the way
0:44:17 the world
0:44:17 works
0:44:17 we need
0:44:19 a network
0:44:20 architecture
0:44:20 that is
0:44:21 up to
0:44:22 the challenge
0:44:22 of you
0:44:23 know
0:44:23 these
0:44:23 super
0:44:24 intelligent
0:44:24 agents
0:44:25 that are
0:44:25 running
0:44:25 around
0:44:26 we were
0:44:27 talking
0:44:27 before we
0:44:27 started
0:44:28 rolling
0:44:28 that you
0:44:28 guys
0:44:28 have an
0:44:29 office
0:44:29 in DC
0:44:30 and part
0:44:30 of what
0:44:30 you do
0:44:32 is advise
0:44:32 on that
0:44:33 like what
0:44:33 what does
0:44:34 the infrastructure
0:44:35 changes
0:44:35 what do
0:44:35 they need
0:44:36 to look
0:44:36 like
0:44:37 what are
0:44:37 a small
0:44:38 handful
0:44:38 of things
0:44:39 that you
0:44:39 guys are
0:44:40 really
0:44:40 pushing
0:44:41 to see
0:44:41 the government
0:44:42 adopt
0:44:42 to modernize
0:44:43 the way
0:44:43 that
0:44:44 the whole
0:44:44 bureaucracy
0:44:45 works
0:44:46 yeah so
0:44:47 there’s a few
0:44:48 things you know
0:44:48 and one of the
0:44:49 things is because
0:44:50 you know
0:44:51 blockchain
0:44:52 technology
0:44:53 involves money
0:44:55 we do need
0:44:55 right it’s not
0:44:56 like we don’t
0:44:56 need any
0:44:57 regulation
0:44:58 we do need
0:44:58 regulation
0:45:00 and there are
0:45:00 kind of very
0:45:01 specific things
0:45:02 that that
0:45:03 we’re working
0:45:04 with the
0:45:04 administration
0:45:05 to make
0:45:05 sure
0:45:06 are done
0:45:07 in a way
0:45:08 that kind
0:45:09 of creates
0:45:09 a great
0:45:10 environment
0:45:10 for everybody
0:45:11 what do
0:45:12 you guys
0:45:12 hoping
0:45:13 will get
0:45:13 blocked
0:45:13 out
0:45:14 for instance
0:45:14 is that
0:45:14 what you’re
0:45:14 about to
0:45:15 cover
0:45:16 yeah I mean
0:45:17 so like
0:45:18 one of the
0:45:20 first thing
0:45:20 you need
0:45:21 is you know
0:45:22 we do need
0:45:23 electronic money
0:45:23 you know
0:45:24 in the
0:45:26 form of
0:45:26 stable coin
0:45:27 so actual
0:45:27 currency
0:45:29 and
0:45:30 but we need
0:45:31 that to not
0:45:32 like it’s very
0:45:32 bad if one of
0:45:33 those collapses
0:45:36 because then
0:45:36 like the
0:45:37 whole trust
0:45:37 and the
0:45:37 system
0:45:38 breaks down
0:45:38 and so
0:45:38 forth
0:45:39 well why
0:45:39 do we need
0:45:40 this kind
0:45:40 of money
0:45:41 this kind
0:45:42 of internet
0:45:42 native money
0:45:44 well
0:45:45 I’ll give you
0:45:45 an example
0:45:46 so we have
0:45:46 a company
0:45:48 called
0:45:48 Daylight Energy
0:45:50 and what
0:45:50 they do
0:45:50 is
0:45:52 so we’re
0:45:52 running
0:45:52 we’re going
0:45:53 to run
0:45:53 into a big
0:45:54 energy problem
0:45:55 with AI
0:45:55 that I think
0:45:56 most people
0:45:57 probably listening
0:45:58 to this know
0:45:58 about where
0:45:59 AI consumes
0:46:00 a massive
0:46:00 amount of
0:46:01 energy
0:46:01 you know
0:46:02 much more
0:46:02 than Bitcoin
0:46:02 ever did
0:46:03 by the way
0:46:04 which everybody
0:46:04 was all up
0:46:05 in arms about
0:46:07 and you know
0:46:07 so much so
0:46:08 that like
0:46:08 you can’t
0:46:09 really even
0:46:09 get it out
0:46:09 of the power
0:46:10 grid
0:46:10 and I think
0:46:11 Trump has
0:46:11 been smart
0:46:12 about this
0:46:12 saying hey
0:46:12 you probably
0:46:13 need to build
0:46:15 a kind
0:46:17 of power
0:46:17 next to your
0:46:18 data center
0:46:19 because we
0:46:19 can’t be
0:46:20 giving it
0:46:20 to you
0:46:20 from the
0:46:21 central tank
0:46:22 but beyond
0:46:22 that I think
0:46:23 that you know
0:46:23 kind of
0:46:24 individuals
0:46:25 you know
0:46:26 now have
0:46:27 Tesla
0:46:28 kind of
0:46:28 solar panels
0:46:29 and power
0:46:29 walls
0:46:30 and these
0:46:30 kinds of
0:46:31 things
0:46:32 and when
0:46:32 you have
0:46:32 one of
0:46:33 those
0:46:33 you
0:46:33 sometimes
0:46:34 have
0:46:34 more
0:46:34 energy
0:46:35 than you
0:46:35 need
0:46:36 and sometimes
0:46:36 have
0:46:36 less
0:46:37 and wouldn’t
0:46:37 it be
0:46:38 great
0:46:39 if you
0:46:39 could
0:46:40 you know
0:46:40 if there
0:46:41 was a
0:46:41 nice
0:46:42 system
0:46:43 that figured
0:46:43 out who
0:46:44 needed
0:46:44 energy
0:46:45 and who
0:46:46 had
0:46:46 energy
0:46:47 and you
0:46:47 could just
0:46:47 trade
0:46:48 and there
0:46:49 was some
0:46:50 kind of
0:46:50 contract
0:46:51 that said
0:46:51 okay this
0:46:51 is what
0:46:52 you pay
0:46:52 during peak
0:46:53 this is what
0:46:53 you pay
0:46:53 at different
0:46:54 periods
0:46:55 and that
0:46:56 contract
0:46:56 probably
0:46:57 best done
0:46:57 in the
0:46:57 form of
0:46:58 a smart
0:46:58 contract
0:46:59 but a
0:46:59 power wall
0:47:00 is not
0:47:00 a human
0:47:01 so it
0:47:01 doesn’t have
0:47:01 a credit
0:47:02 card
0:47:02 it can’t
0:47:02 get a credit
0:47:03 card
0:47:03 doesn’t have
0:47:03 a bank
0:47:04 account
0:47:04 doesn’t have
0:47:04 a social
0:47:05 security number
0:47:07 but it can
0:47:07 trade crypto
0:47:08 it can trade
0:47:08 stable coins
0:47:11 and so we
0:47:11 need that
0:47:11 kind of
0:47:12 currency
0:47:13 to kind
0:47:14 of facilitate
0:47:16 all these
0:47:16 kind of
0:47:17 automated
0:47:17 agreements
0:47:18 and automated
0:47:19 transfer
0:47:19 of
0:47:22 kind of
0:47:23 wealth
0:47:24 between
0:47:25 entities
0:47:26 in order to
0:47:26 kind of
0:47:27 solve these
0:47:28 big problems
0:47:28 that we have
0:47:29 like energy
0:47:31 and so we
0:47:31 need a
0:47:32 stable coin
0:47:32 bill that
0:47:33 kind of
0:47:33 says okay
0:47:34 look
0:47:35 we need
0:47:36 these
0:47:36 currencies
0:47:37 to be
0:47:37 backed
0:47:38 one for
0:47:39 one with
0:47:39 US dollars
0:47:40 or whatever
0:47:41 it is
0:47:42 so that
0:47:43 you know
0:47:43 we can
0:47:44 have a
0:47:46 system that
0:47:46 works and
0:47:46 is trusted
0:47:47 now there’s
0:47:48 this really
0:47:49 interesting side
0:47:49 benefit to
0:47:50 that which
0:47:51 is if
0:47:51 you look
0:47:52 at treasury
0:47:52 auctions
0:47:53 lately
0:47:55 the demand
0:47:56 for dollars
0:47:56 is not
0:47:57 good
0:47:58 you know
0:47:58 and a lot
0:47:59 of that
0:47:59 is the
0:48:00 two biggest
0:48:01 kind of
0:48:02 lenders
0:48:02 to the
0:48:03 US have
0:48:04 been China
0:48:04 and Japan
0:48:06 and you
0:48:06 know China
0:48:06 has backed
0:48:07 off a lot
0:48:08 and Japan
0:48:09 has backed
0:48:10 off somewhat
0:48:11 and so
0:48:11 the demand
0:48:12 for dollars
0:48:12 has gone
0:48:12 down
0:48:13 we’ve done
0:48:13 things to
0:48:14 also dampen
0:48:15 demand like
0:48:16 you know
0:48:16 when we
0:48:17 sanctioned
0:48:18 Russia
0:48:18 and we
0:48:19 seized the
0:48:19 assets of
0:48:20 the Russian
0:48:20 central bank
0:48:21 you know
0:48:21 there were
0:48:22 other countries
0:48:22 that had
0:48:23 you know
0:48:23 other entities
0:48:24 that had
0:48:24 money there
0:48:26 and their
0:48:26 money got
0:48:26 frozen
0:48:27 and they
0:48:27 couldn’t
0:48:28 access it
0:48:28 and so
0:48:28 that makes
0:48:29 people more
0:48:30 wary of holding
0:48:30 everything in
0:48:31 dollars
0:48:32 so we’ve
0:48:33 done a lot
0:48:34 to dampen
0:48:34 that which
0:48:34 of course
0:48:35 is you
0:48:35 know
0:48:36 fueled
0:48:36 inflation
0:48:37 in the
0:48:38 same way
0:48:38 that increasing
0:48:39 supply
0:48:40 fuels
0:48:40 inflation
0:48:41 killing
0:48:41 demand
0:48:42 fuels
0:48:43 inflation
0:48:44 so here
0:48:44 we would
0:48:44 have this
0:48:45 new
0:48:46 major source
0:48:46 of demand
0:48:47 for dollars
0:48:48 and then the
0:48:49 dollars would be
0:48:49 much more useful
0:48:50 because you can
0:48:50 use them
0:48:51 online as well
0:48:53 and machines
0:48:53 can use them
0:48:54 and so forth
0:48:54 so we
0:48:55 really
0:48:55 and sorry
0:48:56 really fast
0:48:56 for people
0:48:57 that are
0:48:57 trying to
0:48:58 track that
0:48:58 the reason
0:48:59 that that
0:49:01 would increase
0:49:01 the demand
0:49:01 for dollars
0:49:02 is that they
0:49:02 would the
0:49:03 stable coin
0:49:03 would be
0:49:04 backed one
0:49:04 for one
0:49:05 with debt
0:49:06 is that the
0:49:06 idea
0:49:07 well yeah
0:49:07 with
0:49:09 where you
0:49:09 would basically
0:49:10 have
0:49:10 you would
0:49:11 have to
0:49:11 have a
0:49:12 dollar
0:49:12 right
0:49:15 for
0:49:16 every
0:49:16 whole
0:49:17 treasuries
0:49:17 stable coin
0:49:17 yeah you
0:49:18 basically
0:49:18 hold
0:49:18 treasuries
0:49:19 so that
0:49:20 if somebody
0:49:20 wanted to
0:49:21 redeem their
0:49:21 stable coins
0:49:22 they could
0:49:23 and then that
0:49:23 way
0:49:24 you know
0:49:24 kind of
0:49:25 the equivalent
0:49:26 of the
0:49:26 gold standard
0:49:27 in the old
0:49:27 days
0:49:27 you know
0:49:28 when dollars
0:49:28 were trying
0:49:29 to get
0:49:29 credible
0:49:30 you know
0:49:31 we would
0:49:31 need
0:49:31 like
0:49:33 dollars
0:49:34 to be the
0:49:34 gold standard
0:49:35 for the
0:49:35 stable coin
0:49:37 you know
0:49:38 and probably
0:49:38 we should
0:49:39 never back
0:49:40 off of that
0:49:40 maybe we
0:49:40 should never
0:49:41 backed off
0:49:41 of gold
0:49:42 but
0:49:43 you know
0:49:43 it’s easier
0:49:44 when it’s
0:49:44 dollars
0:49:45 because we
0:49:45 did kind of
0:49:46 start to run
0:49:46 out of gold
0:49:47 a bit
0:49:49 so
0:49:50 yeah so
0:49:51 that’s
0:49:51 you know
0:49:51 one thing
0:49:53 then
0:49:54 secondly
0:49:55 there’s
0:49:56 a bill
0:49:57 that went
0:49:58 through the
0:49:58 house
0:49:58 known as
0:49:59 the market
0:49:59 structure
0:50:00 bill
0:50:00 it was
0:50:01 technically
0:50:01 called
0:50:02 fit 21
0:50:03 that’s a
0:50:04 very important
0:50:04 you know
0:50:05 whether it’s
0:50:05 exactly that
0:50:06 or some
0:50:06 form of
0:50:07 that
0:50:08 because
0:50:09 you know
0:50:09 when you
0:50:10 talk about
0:50:11 tokens
0:50:12 which are
0:50:13 this kind
0:50:13 of instrument
0:50:14 that’s very
0:50:15 very important
0:50:16 in blockchain
0:50:16 world
0:50:17 because it’s
0:50:17 the way
0:50:19 that this
0:50:20 amazing
0:50:21 kind of
0:50:21 network
0:50:22 of computers
0:50:23 gets paid
0:50:23 for it
0:50:23 so
0:50:24 you know
0:50:25 who
0:50:26 who pays
0:50:26 the people
0:50:26 for running
0:50:28 the computers
0:50:28 well that’s
0:50:29 paid in the
0:50:29 form of
0:50:30 these tokens
0:50:31 but these
0:50:31 tokens
0:50:33 which can be
0:50:34 created on
0:50:34 blockchain
0:50:36 have
0:50:38 they can be
0:50:39 many things
0:50:41 so you can
0:50:41 create a token
0:50:42 that’s a
0:50:42 collectible
0:50:43 you know
0:50:44 you can
0:50:45 create a
0:50:45 token
0:50:47 that is a
0:50:48 digital property
0:50:48 right
0:50:49 you know
0:50:49 that links
0:50:51 to you know
0:50:51 some piece
0:50:52 of real
0:50:52 estate
0:50:53 or a piece
0:50:53 of art
0:50:53 or so
0:50:54 forth
0:50:55 a token
0:50:55 can be
0:50:57 a you know
0:50:58 pokemon card
0:50:58 a token
0:50:58 could be
0:50:59 a coupon
0:51:00 a token
0:51:00 could be
0:51:01 a security
0:51:02 that represents
0:51:02 a stock
0:51:04 it could be
0:51:04 you know
0:51:05 a dollar
0:51:06 so which
0:51:07 one is it
0:51:08 is a very
0:51:09 kind of
0:51:09 important set
0:51:10 of rules
0:51:11 that doesn’t
0:51:11 exist
0:51:12 and this is
0:51:12 one of the
0:51:14 most insidious
0:51:14 thing that
0:51:15 the Biden
0:51:15 administration
0:51:16 did was
0:51:17 basically say
0:51:18 well everything
0:51:19 is a security
0:51:19 everything’s
0:51:20 a stock
0:51:22 or you know
0:51:22 like some
0:51:23 thing with
0:51:23 asymmetric
0:51:24 information
0:51:25 which kind
0:51:26 of basically
0:51:28 undermines
0:51:28 the whole
0:51:30 power of the
0:51:30 technology
0:51:31 and so it
0:51:32 was basically
0:51:33 a scheme
0:51:34 for them
0:51:36 to kind
0:51:37 of get rid
0:51:37 of the industry
0:51:38 but it was
0:51:39 very very
0:51:40 dark
0:51:43 cynical
0:51:44 way of
0:51:46 legislating
0:51:47 things and
0:51:47 they would
0:51:47 make these
0:51:48 fake claims
0:51:49 about scams
0:51:50 and so forth
0:51:51 but the
0:51:52 market structure
0:51:52 bill is
0:51:53 very very
0:51:53 important in
0:51:54 that way
0:51:54 and by the
0:51:55 way also
0:51:55 you know
0:51:56 another thing
0:51:57 that was in
0:51:57 the original
0:51:58 market structure
0:51:59 bill which
0:52:00 is important
0:52:00 is look
0:52:01 there are
0:52:03 also scams
0:52:04 there are
0:52:04 you know
0:52:05 and we call
0:52:05 it the casino
0:52:07 but you know
0:52:08 like I can
0:52:09 create some
0:52:10 coin like the
0:52:11 Hak Tua girl
0:52:11 did right
0:52:12 like and
0:52:13 she creates
0:52:14 a coin
0:52:15 she kind
0:52:16 of lies
0:52:17 about
0:52:19 you know
0:52:20 her holdings
0:52:20 and says
0:52:21 she’s going
0:52:21 to hold
0:52:21 them but
0:52:22 then sells
0:52:23 them you
0:52:23 know
0:52:24 after people
0:52:25 buy it
0:52:25 in a short
0:52:26 time period
0:52:26 and so forth
0:52:27 and there’s
0:52:28 no part of
0:52:28 the problem
0:52:28 is there’s
0:52:29 no kind
0:52:30 of rules
0:52:30 around that
0:52:31 but in
0:52:31 the kind
0:52:32 of bill
0:52:33 that passed
0:52:33 the house
0:52:34 it said
0:52:34 like you
0:52:35 can create
0:52:35 a token
0:52:37 but if
0:52:37 you hold
0:52:37 it you
0:52:38 can’t trade
0:52:38 it for
0:52:39 four years
0:52:40 that kind
0:52:41 of takes
0:52:41 a lot
0:52:42 of the
0:52:42 ability
0:52:43 to scam
0:52:44 out of
0:52:44 it
0:52:45 and kind
0:52:45 of forces
0:52:46 people to
0:52:46 do things
0:52:47 that are
0:52:47 their real
0:52:48 utilities
0:52:48 or if
0:52:48 it’s a
0:52:49 collectible
0:52:50 you know
0:52:50 if it is
0:52:51 the Hak Tua
0:52:51 collectible
0:52:53 you know
0:52:53 it’s got to
0:52:54 be a real
0:52:54 collectible
0:52:55 where you
0:52:56 don’t just
0:52:56 you know
0:52:57 rug
0:52:58 the users
0:52:59 of it
0:52:59 right away
0:53:00 and so
0:53:01 that you
0:53:01 know
0:53:01 that’s
0:53:02 right away
0:53:04 it’s okay
0:53:04 to do it
0:53:04 later
0:53:04 but
0:53:05 well but
0:53:05 you know
0:53:06 like in
0:53:06 four years
0:53:07 it is
0:53:07 what it
0:53:07 is
0:53:08 right
0:53:08 yeah
0:53:09 no no
0:53:09 I’m just
0:53:09 giving you
0:53:10 a hard
0:53:10 time
0:53:10 I just
0:53:10 know how
0:53:11 that’s
0:53:11 going to
0:53:11 sound
0:53:11 to
0:53:11 people
0:53:12 yeah
0:53:13 yeah
0:53:14 thank you
0:53:15 but you
0:53:15 know
0:53:15 so these
0:53:16 kinds of
0:53:16 things I
0:53:17 think are
0:53:18 are going
0:53:18 to be
0:53:18 really
0:53:18 important
0:53:19 to
0:53:19 making
0:53:19 the
0:53:20 whole
0:53:20 industry
0:53:20 work
0:53:21 and so
0:53:21 we’re
0:53:21 working
0:53:22 you know
0:53:23 on that
0:53:23 you know
0:53:24 trying to
0:53:24 make it
0:53:25 safe for
0:53:25 everybody
0:53:26 but as
0:53:27 I said
0:53:27 it’s just
0:53:28 such a
0:53:28 critical
0:53:29 technology
0:53:34 yes it’s
0:53:34 just going
0:53:34 to be
0:53:35 like a
0:53:35 very
0:53:36 kind of
0:53:37 problematic
0:53:39 you know
0:53:40 it’s going
0:53:40 to be
0:53:40 it’s
0:53:41 cyberpunk
0:53:41 it’s
0:53:44 it’s
0:53:44 a
0:53:44 it’s
0:53:44 a
0:53:45 not
0:53:46 yeah
0:53:46 it’s
0:53:46 a
0:53:47 high
0:53:47 technology
0:53:49 difficult
0:53:50 society
0:53:51 yeah
0:53:51 yeah
0:53:52 it
0:53:53 it was
0:53:54 shocking
0:53:54 to me
0:53:55 the level
0:53:55 of
0:53:56 backlash
0:53:57 that
0:53:57 the
0:53:57 blockchain
0:53:58 web
0:53:58 3
0:53:59 community
0:54:00 got
0:54:01 what
0:54:02 do you
0:54:02 think
0:54:03 drives
0:54:03 that
0:54:03 is
0:54:03 it
0:54:04 just
0:54:04 the
0:54:05 perception
0:54:05 that
0:54:05 it
0:54:06 was
0:54:06 only
0:54:06 scams
0:54:06 and
0:54:07 there’s
0:54:07 nothing
0:54:07 real
0:54:08 like
0:54:08 what
0:54:08 was
0:54:08 that
0:54:08 all
0:54:09 about
0:54:09 so
0:54:09 there
0:54:10 was
0:54:10 multiple
0:54:11 factors
0:54:11 so
0:54:11 the
0:54:11 first
0:54:11 one
0:54:12 is
0:54:12 the
0:54:12 one
0:54:12 that
0:54:12 hits
0:54:13 all
0:54:13 new
0:54:13 technology
0:54:14 where
0:54:15 oh
0:54:16 it’s
0:54:16 a
0:54:16 toy
0:54:18 it
0:54:18 doesn’t
0:54:18 do
0:54:18 anything
0:54:19 new
0:54:19 like
0:54:20 the
0:54:20 old
0:54:20 way
0:54:20 of
0:54:20 doing
0:54:21 things
0:54:21 is
0:54:21 better
0:54:22 and
0:54:24 you
0:54:24 know
0:54:24 we
0:54:24 saw
0:54:24 that
0:54:25 with
0:54:25 social
0:54:25 networking
0:54:26 we
0:54:26 actually
0:54:27 saw
0:54:27 that
0:54:27 with
0:54:27 the
0:54:27 internet
0:54:28 I
0:54:28 think
0:54:29 Paul
0:54:29 Krugman
0:54:29 famously
0:54:30 said
0:54:30 never
0:54:31 have
0:54:31 more
0:54:32 economic
0:54:32 impact
0:54:32 on a
0:54:33 fax
0:54:33 machine
0:54:33 and so
0:54:33 forth
0:54:34 so
0:54:34 that’s
0:54:34 just
0:54:34 kind of
0:54:34 a
0:54:35 normal
0:54:35 thing
0:54:35 that
0:54:35 happens
0:54:36 with
0:54:36 new
0:54:36 technologies
0:54:37 as they
0:54:37 start
0:54:37 out
0:54:57 and
0:54:57 so
0:54:58 you
0:54:58 know
0:54:58 if
0:54:58 you
0:54:59 look
0:54:59 at
0:55:00 even
0:55:00 like
0:55:01 the
0:55:01 iPhone
0:55:02 it
0:55:02 was
0:55:02 a
0:55:03 bad
0:55:03 phone
0:55:04 it
0:55:05 had
0:55:05 a
0:55:05 horrible
0:55:06 keyboard
0:55:06 you
0:55:07 know
0:55:07 if
0:55:07 you
0:55:07 compared
0:55:07 it
0:55:07 to
0:55:08 anything
0:55:08 it
0:55:08 wasn’t
0:55:08 very
0:55:09 powerful
0:55:09 it
0:55:09 had
0:55:10 a
0:55:10 little
0:55:10 itty
0:55:10 bitty
0:55:11 screen
0:55:12 but
0:55:13 it
0:55:13 had
0:55:13 a
0:55:13 feature
0:55:14 that
0:55:15 was
0:55:15 pretty
0:55:15 awesome
0:55:16 which
0:55:16 you
0:55:16 could
0:55:16 put
0:55:16 in
0:55:16 your
0:55:17 pocket
0:55:18 and
0:55:18 it
0:55:18 had
0:55:18 like
0:55:18 a
0:55:19 GPS
0:55:19 in
0:55:19 it
0:55:19 and
0:55:19 a
0:55:19 camera
0:55:20 in
0:55:20 it
0:55:21 and
0:55:21 so
0:55:21 now
0:55:21 you
0:55:21 could
0:55:21 build
0:55:22 Instagram
0:55:22 you
0:55:22 could
0:55:22 build
0:55:23 Uber
0:55:24 which
0:55:24 you
0:55:24 could
0:55:24 not
0:55:24 build
0:55:24 with
0:55:25 the
0:55:25 PC
0:55:25 and
0:55:25 you
0:55:25 still
0:55:26 can’t
0:55:26 build
0:55:26 with
0:55:26 the
0:55:26 PC
0:55:28 and
0:55:28 so
0:55:28 that
0:55:29 was
0:55:29 enough
0:55:30 and
0:55:30 then
0:55:30 eventually
0:55:31 it
0:55:31 started
0:55:31 to
0:55:31 add
0:55:32 the
0:55:32 other
0:55:32 features
0:55:33 and
0:55:33 it’s
0:55:33 an
0:55:33 awfully
0:55:34 powerful
0:55:34 computer
0:55:35 these
0:55:35 days
0:55:36 if
0:55:36 you
0:55:37 look
0:55:37 at
0:55:37 blockchain
0:55:38 it’s
0:55:39 slower
0:55:40 it’s
0:55:40 more
0:55:41 complicated
0:55:41 to
0:55:41 program
0:55:42 like
0:55:42 there’s
0:55:42 a lot
0:55:42 of
0:55:43 issues
0:55:43 with
0:55:43 it
0:55:44 but
0:55:44 it’s
0:55:44 got
0:55:44 a
0:55:51 that
0:55:52 code
0:55:53 you
0:55:53 know
0:55:53 says
0:55:54 there’s
0:55:54 only
0:55:54 21
0:55:55 million
0:55:55 bitcoin
0:55:56 you
0:55:57 can
0:55:57 absolutely
0:55:58 count
0:55:58 on
0:55:58 that
0:55:59 in
0:55:59 a
0:55:59 way
0:56:00 that
0:56:00 like
0:56:00 you
0:56:00 can’t
0:56:01 trust
0:56:01 Google
0:56:02 you
0:56:02 can’t
0:56:02 trust
0:56:03 you
0:56:03 know
0:56:04 Facebook
0:56:04 to
0:56:04 say
0:56:05 like
0:56:05 oh
0:56:05 these
0:56:05 are
0:56:06 our
0:56:06 privacy
0:56:07 rules
0:56:08 like
0:56:08 you
0:56:08 can’t
0:56:08 trust
0:56:09 that
0:56:09 at
0:56:09 all
0:56:10 you
0:56:10 can’t
0:56:10 trust
0:56:10 the
0:56:10 US
0:56:11 government
0:56:11 to
0:56:11 say
0:56:11 they’re
0:56:11 not
0:56:11 going
0:56:11 to
0:56:11 print
0:56:12 any
0:56:12 more
0:56:12 money
0:56:13 like
0:56:13 that’s
0:56:13 for
0:56:13 sure
0:56:14 and
0:56:15 so
0:56:15 you
0:56:15 know
0:56:21 count
0:56:21 on
0:56:22 and
0:56:22 you
0:56:22 don’t
0:56:23 have
0:56:23 to
0:56:23 trust
0:56:23 a
0:56:23 company
0:56:24 you
0:56:24 don’t
0:56:24 have
0:56:24 to
0:56:24 trust
0:56:25 a
0:56:26 lawyer
0:56:26 you
0:56:26 just
0:56:27 have
0:56:27 to
0:56:27 trust
0:56:28 the
0:56:28 game
0:56:28 theoretic
0:56:29 mathematical
0:56:30 properties
0:56:30 of the
0:56:30 blockchain
0:56:32 and
0:56:32 that’s
0:56:32 amazing
0:56:33 so
0:56:33 now
0:56:33 you
0:56:33 can
0:56:34 you
0:56:34 know
0:56:35 program
0:56:35 property
0:56:36 rights
0:56:36 and
0:56:36 money
0:56:37 and
0:56:37 law
0:56:37 and
0:56:38 all
0:56:38 these
0:56:38 kinds
0:56:38 of
0:56:39 things
0:56:39 that
0:56:39 you
0:56:39 could
0:56:39 never
0:56:39 do
0:56:40 before
0:56:41 and
0:56:42 so
0:56:42 I
0:56:42 think
0:56:43 that’s
0:56:43 hard
0:56:43 for
0:56:44 normal
0:56:44 people
0:56:44 to
0:56:45 understand
0:56:46 who
0:56:46 aren’t
0:56:47 deep
0:56:52 and
0:56:52 then
0:56:53 you
0:56:53 know
0:56:53 I
0:56:53 think
0:56:54 the
0:56:54 next
0:56:54 wave
0:56:55 was
0:56:55 you
0:56:55 had
0:56:56 look
0:56:57 you
0:56:57 know
0:56:57 it
0:56:58 was
0:56:58 a
0:56:58 very
0:56:59 odd
0:56:59 thing
0:57:01 with
0:57:01 the
0:57:01 Biden
0:57:01 administration
0:57:02 because
0:57:02 he
0:57:02 wasn’t
0:57:03 really
0:57:04 I
0:57:04 think
0:57:04 it’s
0:57:05 come
0:57:05 out
0:57:05 now
0:57:05 he
0:57:05 wasn’t
0:57:06 really
0:57:06 the
0:57:06 president
0:57:06 he
0:57:07 wasn’t
0:57:07 really
0:57:07 making
0:57:07 any
0:57:08 decisions
0:57:09 you
0:57:09 couldn’t
0:57:10 even
0:57:10 get a
0:57:10 meeting
0:57:10 with
0:57:10 him
0:57:11 if
0:57:11 you
0:57:11 were
0:57:11 in
0:57:12 his
0:57:12 cabinet
0:57:13 and
0:57:13 in
0:57:13 terms
0:57:13 of
0:57:14 domestic
0:57:14 policy
0:57:14 that
0:57:15 was
0:57:15 run
0:57:15 by
0:57:16 Elizabeth
0:57:16 Warren
0:57:17 and
0:57:17 then
0:57:17 the
0:57:17 second
0:57:18 confusing
0:57:18 thing
0:57:18 is
0:57:19 Elizabeth
0:57:19 Warren
0:57:20 is
0:57:20 always
0:57:20 calling
0:57:21 like
0:57:21 people
0:57:21 fascist
0:57:22 her
0:57:23 whole
0:57:24 push
0:57:24 with
0:57:25 fintech
0:57:25 and
0:57:25 crypto
0:57:26 was
0:57:27 to
0:57:27 make
0:57:28 sure
0:57:29 that
0:57:29 she
0:57:29 could
0:57:30 kick
0:57:30 people
0:57:30 out
0:57:30 of
0:57:30 the
0:57:31 banking
0:57:31 system
0:57:31 who
0:57:31 are
0:57:32 political
0:57:32 enemies
0:57:33 and
0:57:33 so
0:57:34 in
0:57:34 order
0:57:34 to
0:57:34 do
0:57:34 that
0:57:35 you
0:57:35 have
0:57:35 to
0:57:35 outlaw
0:57:36 new
0:57:36 forms
0:57:36 of
0:57:37 financial
0:57:38 technology
0:57:39 because
0:57:39 those
0:57:40 would
0:57:40 be
0:57:40 kind
0:57:40 of
0:57:41 back
0:57:41 doors
0:57:41 or
0:57:42 side
0:57:42 doors
0:57:42 or
0:57:43 parallels
0:57:44 to
0:57:45 the
0:57:46 g-sibs
0:57:46 and
0:57:46 the
0:57:46 banking
0:57:46 system
0:57:47 which
0:57:47 she
0:57:48 comprehensively
0:57:49 and I
0:57:49 think
0:57:49 this is
0:57:50 coming out
0:57:50 now
0:57:50 could
0:57:50 kick
0:57:51 people
0:57:51 out
0:57:51 of
0:57:52 and
0:57:52 so
0:57:52 when
0:57:52 you
0:57:53 use
0:57:53 when
0:57:54 it’s
0:57:54 a full
0:57:55 top-down
0:57:55 hierarchy
0:57:56 and you
0:57:56 can
0:57:56 use
0:57:57 private
0:57:57 companies
0:57:58 to enforce
0:57:58 your will
0:57:59 that is
0:58:00 the way
0:58:00 fascism
0:58:01 works
0:58:01 and then
0:58:02 the way
0:58:02 she does
0:58:02 it
0:58:02 is
0:58:03 she
0:58:03 sells
0:58:04 this
0:58:04 fake
0:58:05 story
0:58:05 about
0:58:06 you know
0:58:07 it’s
0:58:08 funding
0:58:08 terror
0:58:08 and it
0:58:09 turns out
0:58:09 like
0:58:09 the
0:58:10 USAID
0:58:10 was
0:58:11 funding
0:58:11 the
0:58:11 terrorist
0:58:11 groups
0:58:12 but
0:58:12 that’s
0:58:12 a
0:58:12 different
0:58:12 story
0:58:13 but
0:58:13 you know
0:58:13 it’s
0:58:14 doing
0:58:14 all
0:58:14 these
0:58:15 nefarious
0:58:15 things
0:58:16 which
0:58:16 you know
0:58:17 was just
0:58:17 a very
0:58:18 unfair
0:58:18 portrayal
0:58:19 and so
0:58:19 then
0:58:20 the whole
0:58:20 industry
0:58:21 got this
0:58:21 reputation
0:58:22 as scammy
0:58:22 and this
0:58:23 and that
0:58:23 and the
0:58:23 other
0:58:24 and then
0:58:24 of course
0:58:25 we had
0:58:26 Sam
0:58:26 Bankman
0:58:26 freed
0:58:27 who
0:58:27 didn’t
0:58:27 do
0:58:27 us
0:58:27 any
0:58:28 favors
0:58:28 by
0:58:30 you know
0:58:30 and this
0:58:31 is another
0:58:32 kind of
0:58:32 though
0:58:33 issue
0:58:33 with what
0:58:34 Elizabeth Warren
0:58:34 did
0:58:34 is she
0:58:35 blocked
0:58:35 all
0:58:36 legislation
0:58:38 and so
0:58:38 the
0:58:39 criminals
0:58:39 were running
0:58:40 free
0:58:40 and the
0:58:41 people
0:58:41 doing
0:58:41 things
0:58:42 that
0:58:42 should
0:58:42 have
0:58:42 been
0:58:42 legal
0:58:43 were
0:58:43 getting
0:58:44 terrorized
0:58:44 by the
0:58:45 government
0:58:46 when they
0:58:46 should
0:58:46 have been
0:58:46 looking
0:58:47 they should
0:58:48 have been
0:58:48 looking at
0:58:49 FTX
0:58:49 they were
0:58:50 looking at
0:58:50 Coinbase
0:58:51 which was
0:58:51 totally
0:58:52 compliant
0:58:52 public
0:58:53 company
0:58:54 you know
0:58:54 begging
0:58:55 for
0:58:55 feedback
0:58:56 tell us
0:58:56 what you
0:58:56 want us
0:58:57 to do
0:58:57 exactly
0:58:59 yeah
0:58:59 that
0:59:00 whole
0:59:00 thing
0:59:00 was
0:59:00 crazy
0:59:01 so
0:59:01 given
0:59:02 that
0:59:02 AI
0:59:02 is
0:59:03 putting
0:59:03 us
0:59:03 on
0:59:03 a
0:59:03 collision
0:59:04 course
0:59:04 with
0:59:05 I
0:59:05 don’t
0:59:05 know
0:59:06 who’s
0:59:06 real
0:59:06 I
0:59:06 don’t
0:59:06 know
0:59:07 what’s
0:59:07 fake
0:59:07 do
0:59:07 you
0:59:08 think
0:59:08 that
0:59:08 blockchain
0:59:09 is
0:59:09 about
0:59:09 to
0:59:09 have
0:59:09 its
0:59:09 day
0:59:10 like
0:59:10 in
0:59:10 the
0:59:10 next
0:59:11 12
0:59:11 to
0:59:12 24
0:59:12 months
0:59:13 or
0:59:13 is
0:59:13 this
0:59:13 still
0:59:14 something
0:59:14 that
0:59:14 it’s
0:59:14 so
0:59:15 embedded
0:59:15 deep
0:59:15 in
0:59:15 the
0:59:16 infrastructure
0:59:16 it’s
0:59:16 going
0:59:16 to
0:59:17 take
0:59:17 a
0:59:17 long
0:59:17 time
0:59:17 to
0:59:18 really
0:59:18 have
0:59:19 its
0:59:19 I
0:59:19 told
0:59:19 you
0:59:19 so
0:59:20 moment
0:59:21 you
0:59:21 know
0:59:21 I
0:59:22 think
0:59:22 it’s
0:59:23 I
0:59:23 think
0:59:23 it’s
0:59:24 within
0:59:24 24
0:59:24 months
0:59:24 for
0:59:25 sure
0:59:25 I
0:59:25 mean
0:59:25 I
0:59:25 think
0:59:26 that
0:59:27 there’s
0:59:27 enough
0:59:28 you
0:59:29 know
0:59:29 there
0:59:29 were
0:59:29 like
0:59:30 actual
0:59:30 like
0:59:30 if
0:59:30 you
0:59:30 like
0:59:31 kind
0:59:31 of
0:59:31 the
0:59:31 last
0:59:31 wave
0:59:31 of
0:59:32 blockchain
0:59:33 there
0:59:33 were
0:59:34 real
0:59:35 technological
0:59:35 limitations
0:59:36 that
0:59:36 made
0:59:36 it
0:59:37 you
0:59:37 know
0:59:38 I
0:59:38 think
0:59:39 we’re
0:59:39 slowing
0:59:39 it
0:59:40 down
0:59:40 from
0:59:40 getting
0:59:40 broader
0:59:41 adoption
0:59:41 so
0:59:42 you
0:59:42 know
0:59:42 very
0:59:43 obvious
0:59:43 usability
0:59:44 challenges
0:59:45 the
0:59:46 fees
0:59:46 were
0:59:46 really
0:59:46 high
0:59:47 the
0:59:47 blockchains
0:59:47 were
0:59:48 slow
0:59:49 so
0:59:49 there
0:59:49 were
0:59:49 just
0:59:49 a
0:59:50 lot
0:59:50 of
0:59:50 use
0:59:50 cases
0:59:50 that
0:59:50 you
0:59:51 just
0:59:51 couldn’t
0:59:51 do
0:59:51 on
0:59:51 them
0:59:53 I
0:59:53 think
0:59:54 that’s
0:59:54 changing
0:59:55 very
0:59:55 very
0:59:55 fast
0:59:56 things
0:59:56 are
0:59:56 you
0:59:56 know
0:59:57 the
0:59:57 chains
0:59:58 are
0:59:58 much
0:59:58 faster
0:59:59 the
1:00:00 layer
1:00:00 two
1:00:00 stuff
1:00:01 makes
1:00:01 them
1:00:02 very
1:00:02 fast
1:00:02 and
1:00:03 cheap
1:00:05 you
1:00:05 know
1:00:05 people
1:00:05 are
1:00:05 doing
1:00:06 a
1:00:06 lot
1:00:07 on
1:00:07 usability
1:00:09 you
1:00:09 know
1:00:09 for
1:00:10 wallets
1:00:10 and
1:00:10 these
1:00:11 kinds
1:00:11 of
1:00:11 things
1:00:11 so
1:00:12 I
1:00:13 think
1:00:13 we’re
1:00:13 getting
1:00:13 pretty
1:00:14 close
1:00:14 and
1:00:14 then
1:00:14 I
1:00:19 world
1:00:19 coin
1:00:20 you
1:00:20 know
1:00:20 to
1:00:21 me
1:00:21 the
1:00:22 difference
1:00:23 between
1:00:23 that
1:00:23 thing
1:00:23 being
1:00:24 very
1:00:24 broadly
1:00:25 adopted
1:00:25 and
1:00:26 where
1:00:26 it
1:00:26 is
1:00:26 now
1:00:26 where
1:00:26 it’s
1:00:27 I
1:00:27 think
1:00:27 half
1:00:27 the
1:00:27 people
1:00:28 in
1:00:28 Buenos
1:00:28 Aires
1:00:29 use
1:00:29 it
1:00:30 daily
1:00:31 so
1:00:31 it’s
1:00:32 widely
1:00:33 adopted
1:00:33 where
1:00:33 it’s
1:00:33 been
1:00:34 legal
1:00:35 I
1:00:35 think
1:00:35 that
1:00:36 if
1:00:36 they
1:00:37 are
1:00:37 able
1:00:37 to
1:00:38 get
1:00:38 integrated
1:00:39 into
1:00:39 some
1:00:39 of
1:00:39 the
1:00:39 big
1:00:40 social
1:00:41 platforms
1:00:42 then
1:00:43 everybody
1:00:44 needs
1:00:44 proof
1:00:44 of
1:00:45 human
1:00:47 and
1:00:49 you
1:00:49 know
1:00:49 like
1:00:49 it
1:00:49 would
1:00:50 make
1:00:50 the
1:00:51 experience
1:00:52 online
1:00:52 so
1:00:52 much
1:00:52 better
1:00:53 if
1:00:53 you
1:00:53 knew
1:00:53 who
1:00:53 was
1:00:53 human
1:00:54 and
1:00:54 who
1:00:54 was
1:00:54 not
1:00:55 and
1:00:55 right
1:00:55 now
1:00:56 like
1:00:56 you
1:00:56 can’t
1:00:57 tell
1:00:57 at
1:00:58 all
1:00:59 and
1:00:59 so
1:00:59 and
1:01:00 that
1:01:00 problem
1:01:00 is
1:01:00 going
1:01:00 to
1:01:00 get
1:01:01 worse
1:01:01 and
1:01:01 then
1:01:01 the
1:01:02 solution
1:01:02 is
1:01:02 really
1:01:02 here
1:01:04 so
1:01:04 I
1:01:05 think
1:01:05 it’s
1:01:05 going
1:01:05 to
1:01:06 start
1:01:06 to
1:01:06 take
1:01:06 off
1:01:06 and
1:01:06 you
1:01:06 only
1:01:07 need
1:01:07 one
1:01:07 or
1:01:08 two
1:01:09 big
1:01:09 use
1:01:10 cases
1:01:10 to
1:01:11 start
1:01:11 getting
1:01:11 the
1:01:12 whole
1:01:12 infrastructure
1:01:13 deployed
1:01:14 and
1:01:14 once
1:01:15 the
1:01:15 infrastructure
1:01:15 is
1:01:16 deployed
1:01:16 I
1:01:17 think
1:01:17 we’ll
1:01:17 certainly
1:01:18 rely
1:01:18 on
1:01:18 it
1:01:19 and
1:01:20 if
1:01:20 you
1:01:20 look
1:01:20 at
1:01:21 actually
1:01:21 the
1:01:21 curve
1:01:22 of
1:01:22 people
1:01:22 who
1:01:22 have
1:01:23 active
1:01:23 wallets
1:01:24 and
1:01:25 the
1:01:25 curve
1:01:25 of
1:01:26 internet
1:01:27 adoption
1:01:27 they’re
1:01:28 pretty
1:01:28 similar
1:01:32 it’s
1:01:32 about
1:01:32 I think
1:01:33 blockchain
1:01:33 is
1:01:33 growing
1:01:34 a little
1:01:34 faster
1:01:34 than
1:01:34 the
1:01:35 internet
1:01:35 did
1:01:36 initially
1:01:37 and
1:01:37 so
1:01:37 I
1:01:37 think
1:01:38 we’ll
1:01:38 get
1:01:38 to
1:01:39 a
1:01:39 place
1:01:39 where
1:01:40 certainly
1:01:40 everybody
1:01:40 in
1:01:40 the
1:01:41 US
1:01:41 will
1:01:41 be
1:01:42 on
1:01:42 it
1:01:44 which
1:01:44 by
1:01:44 the
1:01:44 way
1:01:45 could
1:01:45 be
1:01:46 great
1:01:46 from
1:01:46 a
1:01:46 government
1:01:47 standpoint
1:01:47 you
1:01:47 know
1:01:48 Elon
1:01:48 has
1:01:48 talked
1:01:49 about
1:01:49 putting
1:01:49 all
1:01:49 the
1:01:50 government
1:01:50 payments
1:01:50 on
1:01:50 the
1:01:51 blockchain
1:01:51 which
1:01:51 I
1:01:51 think
1:01:52 would
1:01:52 be
1:01:52 really
1:01:53 good
1:01:53 for
1:01:53 transparency
1:01:55 we never
1:01:55 get into
1:01:56 this weird
1:01:56 situation
1:01:57 now
1:01:57 where
1:01:58 half
1:01:58 the
1:01:58 country
1:01:59 wants
1:01:59 to
1:02:00 tear
1:02:00 down
1:02:01 all
1:02:01 the
1:02:01 government
1:02:02 services
1:02:02 and half
1:02:02 them
1:02:03 wants
1:02:03 to
1:02:03 keep
1:02:03 it
1:02:03 because
1:02:04 nobody
1:02:04 knows
1:02:04 what
1:02:04 the
1:02:05 spending
1:02:05 is
1:02:06 but
1:02:07 that
1:02:07 would
1:02:07 be
1:02:07 great
1:02:08 but
1:02:08 you know
1:02:09 beyond
1:02:09 that
1:02:09 like
1:02:09 if
1:02:10 you
1:02:10 think
1:02:10 about
1:02:11 well
1:02:11 why
1:02:11 is
1:02:11 there
1:02:12 so
1:02:12 much
1:02:12 waste
1:02:12 and
1:02:12 fraud
1:02:14 well
1:02:14 part
1:02:14 of
1:02:14 it
1:02:14 is
1:02:14 you know
1:02:15 like
1:02:16 you know
1:02:17 I
1:02:17 get
1:02:18 taxed
1:02:18 I
1:02:18 give
1:02:19 my
1:02:19 money
1:02:19 to
1:02:19 the
1:02:20 IRS
1:02:21 the
1:02:21 IRS
1:02:22 gives
1:02:22 it
1:02:23 to
1:02:23 Congress
1:02:23 they
1:02:24 you know
1:02:24 do
1:02:24 whatever
1:02:24 they
1:02:24 do
1:02:25 with
1:02:25 it
1:02:25 and
1:02:25 so
1:02:25 forth
1:02:26 and
1:02:27 well
1:02:27 how
1:02:27 does
1:02:27 it
1:02:27 get
1:02:27 to
1:02:27 the
1:02:28 people
1:02:28 who
1:02:28 need
1:02:29 it
1:02:30 you know
1:02:30 that’s
1:02:30 a
1:02:30 very
1:02:31 lossy
1:02:31 process
1:02:32 you know
1:02:32 and
1:02:32 we
1:02:32 don’t
1:02:32 even
1:02:33 know
1:02:33 they
1:02:33 are
1:02:33 and
1:02:33 it’s
1:02:33 very
1:02:34 you know
1:02:35 one of the things we found out during COVID
1:02:38 the government’s not very good at sending people money
1:02:39 it’s good at taking money
1:02:41 it’s not good at sending them money
1:02:41 right
1:02:44 we lost like 400 billion dollars trying to give people
1:02:44 stimulus
1:02:45 ridiculous
1:02:46 yeah
1:02:46 crazy
1:02:47 ridiculous
1:02:48 you know
1:02:48 like
1:02:48 if
1:02:49 everybody
1:02:49 in the US
1:02:50 had
1:02:51 an address
1:02:52 on the blockchain
1:02:54 you could just tell me
1:02:56 okay here’s 10,000 people who need money
1:02:58 please send them
1:02:59 you know
1:03:00 $5,000 each
1:03:02 well probably that’s too much money for me
1:03:02 but
1:03:03 you know
1:03:04 something like that
1:03:05 you know
1:03:07 whatever my tax bill is
1:03:09 or whatever that portion of wealth redistribution is
1:03:11 that would be 100%
1:03:12 zero loss
1:03:13 by the way
1:03:14 I’d feel a lot better about it
1:03:16 because I’d know I’d be helping people
1:03:17 and you know
1:03:17 look
1:03:18 maybe even
1:03:19 somebody would go
1:03:20 hey this is great
1:03:21 thank you
1:03:21 and we’re like
1:03:22 maybe it would bring us
1:03:24 we wouldn’t have this crazy class warfare
1:03:25 because everybody would go
1:03:26 hey we’re all integrated
1:03:26 you know
1:03:27 like I’m helping you
1:03:28 you’re helping me
1:03:30 and then
1:03:31 like if you had that
1:03:32 then you’d fix the whole
1:03:33 kind of
1:03:35 democracy integrity problem
1:03:35 because
1:03:37 everybody could vote
1:03:38 off that address
1:03:39 and by the way
1:03:40 everybody would have an address
1:03:42 because everybody would want the money
1:03:44 so what bigger incentive
1:03:46 to register to vote
1:03:47 than
1:03:48 you
1:03:49 in order to get money
1:03:50 you have to have an address
1:03:52 which registers you to vote
1:03:53 and so that kind of thing
1:03:53 I think
1:03:55 could get us to
1:03:57 just like a much
1:03:58 higher trust
1:04:00 in our own institutions
1:04:01 hmm
1:04:02 yeah
1:04:02 so
1:04:03 wow
1:04:04 speaking of that
1:04:05 I wanted to
1:04:05 absolutely
1:04:06 scream
1:04:08 into the abyss
1:04:08 when I heard
1:04:09 that people couldn’t retire
1:04:10 from the government
1:04:12 faster than the elevator
1:04:13 would lower their records
1:04:15 down into a mine
1:04:16 I was like
1:04:17 what is happening
1:04:19 what do you take away from
1:04:20 why does Elon
1:04:22 want to do this
1:04:22 why is he sleeping
1:04:23 in hallways
1:04:25 um
1:04:26 why
1:04:27 why is he doing this
1:04:28 is it just to get
1:04:28 government contracts
1:04:29 and it’s nefarious
1:04:31 in the way that so many people
1:04:31 think it is
1:04:32 or is there something
1:04:33 positive there
1:04:34 what’s the
1:04:35 what’s the game
1:04:36 I think there’s a
1:04:38 couple of different things
1:04:38 so
1:04:40 one is
1:04:41 the strong thing
1:04:41 is he
1:04:42 he truly believes
1:04:43 that
1:04:44 um
1:04:45 America’s the best
1:04:46 country in the world
1:04:46 you know
1:04:47 he is an immigrant
1:04:48 um
1:04:49 and that
1:04:50 it’s not guaranteed
1:04:51 to stay that way
1:04:52 and
1:04:54 and we have been
1:04:56 in danger of losing it
1:04:57 uh
1:04:58 and
1:04:59 you know
1:05:00 and so the
1:05:01 most important thing
1:05:01 for him to do
1:05:02 in order for his
1:05:03 companies to be relevant
1:05:04 in order for going to Mars
1:05:05 to be relevant
1:05:06 in order for anything
1:05:07 he wants to do in life
1:05:08 to be relevant
1:05:08 is
1:05:09 we’ve got to stabilize
1:05:10 U.S. government
1:05:11 I think that’s the main thing
1:05:12 driving him
1:05:12 you know
1:05:13 so then you say
1:05:13 well how did he get
1:05:14 to that conclusion
1:05:15 that
1:05:16 you know
1:05:17 the whole country
1:05:18 is in jeopardy
1:05:19 and it really
1:05:19 like
1:05:21 it’s a pretty
1:05:22 uh
1:05:22 it was a pretty
1:05:23 interesting thing
1:05:24 to watch
1:05:24 um
1:05:25 because
1:05:26 right in 2021
1:05:28 I think he was a democrat
1:05:29 and he was certainly
1:05:30 pretty apolitical
1:05:32 and uh
1:05:32 I was actually
1:05:34 in a chat group
1:05:34 with him
1:05:35 um
1:05:36 when he got
1:05:37 the idea
1:05:38 or posed
1:05:38 the question
1:05:39 should he buy
1:05:40 Twitter
1:05:41 um
1:05:42 and
1:05:44 a lot of it
1:05:45 stemmed from
1:05:45 um
1:05:46 you know
1:05:47 it started with
1:05:48 the U.S. government
1:05:50 just harassing him
1:05:51 uh
1:05:51 which was a very odd
1:05:52 thing right
1:05:53 like here’s your
1:05:54 I mean I think
1:05:55 you could very well
1:05:56 argue he was our
1:05:57 most productive citizen
1:05:58 he um
1:05:59 was our entire
1:06:00 space program
1:06:02 he advanced
1:06:02 he advanced
1:06:02 the state
1:06:03 of electric
1:06:03 cars
1:06:04 by 20 years
1:06:05 he’s still
1:06:06 like 95%
1:06:06 of there’s
1:06:07 something like
1:06:07 that
1:06:07 percent
1:06:08 of the electric
1:06:09 cars sold
1:06:09 in the U.S.
1:06:12 you know he’s
1:06:13 you know
1:06:14 done the things
1:06:15 with Neuralink
1:06:16 to you know
1:06:17 help people
1:06:18 like uh
1:06:19 use their arms
1:06:20 and legs
1:06:20 who have been
1:06:21 paralyzed
1:06:21 and this kind
1:06:22 of thing
1:06:22 so you know
1:06:23 you’ve got
1:06:24 um
1:06:25 you know
1:06:25 he’s a really
1:06:26 kind of remarkable
1:06:27 person to want
1:06:28 to pick on
1:06:29 but uh
1:06:30 what happened
1:06:31 was uh
1:06:33 because he got
1:06:34 like this PR
1:06:35 for being very
1:06:35 wealthy
1:06:36 um
1:06:37 the Biden
1:06:37 administration
1:06:38 targeted him
1:06:39 um
1:06:39 you know
1:06:40 and again
1:06:40 they’re fascists
1:06:41 so uh
1:06:42 it’s really
1:06:43 a power struggle
1:06:44 always with the
1:06:44 fascists and
1:06:45 anybody who looks
1:06:46 like they’re
1:06:46 becoming powerful
1:06:48 and yeah
1:06:48 some of the
1:06:49 things they did
1:06:50 and you know
1:06:50 one of the ones
1:06:51 that’s talked
1:06:51 about a lot
1:06:52 was um
1:06:54 they sued him
1:06:55 the Biden
1:06:55 Department of
1:06:56 Justice sued
1:06:58 him for uh
1:07:00 discriminating
1:07:01 against refugees
1:07:02 um
1:07:03 but he had a
1:07:04 contract with
1:07:04 the U.S.
1:07:05 Department of
1:07:05 Defense
1:07:06 that required
1:07:06 him to only
1:07:07 hire U.S.
1:07:07 citizens
1:07:09 so he was
1:07:10 breaking the law
1:07:10 either way
1:07:11 uh
1:07:11 and they
1:07:12 they never
1:07:13 dropped the lawsuit
1:07:14 even after it was
1:07:14 pointed out
1:07:15 even after
1:07:16 you know
1:07:16 it was pointed
1:07:17 out by Congress
1:07:18 uh
1:07:18 so it was
1:07:19 clear harassment
1:07:20 and I think
1:07:21 that his
1:07:22 conclusion from
1:07:23 that was
1:07:24 you know
1:07:24 this
1:07:25 we are
1:07:26 like ironically
1:07:27 um
1:07:28 I think his
1:07:29 conclusion was
1:07:29 we’re losing
1:07:30 the democracy
1:07:31 um
1:07:32 you know
1:07:33 we’re going
1:07:34 into this
1:07:34 very very
1:07:35 strange world
1:07:36 uh
1:07:36 where the
1:07:36 incentives
1:07:37 are all
1:07:37 upside down
1:07:38 and
1:07:39 you know
1:07:40 the way
1:07:41 Elon thinks
1:07:42 is
1:07:42 it’s up to
1:07:43 him to
1:07:43 save it
1:07:44 um
1:07:45 and so he
1:07:45 got like
1:07:46 extremely
1:07:46 involved
1:07:47 and then I
1:07:47 think the
1:07:48 more involved
1:07:49 he got
1:07:49 the more
1:07:51 he both
1:07:52 realized like
1:07:53 a lot of the
1:07:53 things really
1:07:54 were dangerous
1:07:54 and then
1:07:55 secondly
1:07:56 that
1:07:57 um
1:07:57 he
1:07:58 personally
1:07:59 uh
1:08:00 would be
1:08:01 somebody who
1:08:01 would know
1:08:01 how to fix
1:08:01 it
1:08:02 and
1:08:03 you go
1:08:03 like
1:08:03 well why
1:08:04 the hell
1:08:04 would Elon
1:08:05 must know
1:08:05 how to
1:08:05 fix the
1:08:06 government
1:08:06 all this
1:08:07 and you
1:08:07 know
1:08:08 this is
1:08:08 the thing
1:08:09 that everybody’s
1:08:09 saying now
1:08:10 uh
1:08:11 and it’s
1:08:11 funny
1:08:11 because I
1:08:12 told this
1:08:13 to um
1:08:14 Andreessen
1:08:15 years ago
1:08:15 uh
1:08:16 was
1:08:17 because I
1:08:18 I’m a big
1:08:19 fan of
1:08:19 Isaac Newton
1:08:20 and you
1:08:20 know like
1:08:21 we always
1:08:21 talked about
1:08:22 like who
1:08:22 is Elon
1:08:22 like
1:08:23 you know
1:08:23 like
1:08:23 what
1:08:24 entrepreneur
1:08:25 comes to
1:08:25 mind
1:08:26 you know
1:08:26 and it
1:08:26 really
1:08:27 wasn’t
1:08:27 maybe
1:08:28 Thomas
1:08:28 Edison
1:08:28 but not
1:08:29 really
1:08:30 um
1:08:31 but Isaac
1:08:32 Newton
1:08:33 uh
1:08:33 whoa
1:08:34 was really
1:08:34 the one
1:08:35 that I
1:08:35 always
1:08:36 thought
1:08:37 he was
1:08:37 most
1:08:37 like
1:08:38 um
1:08:38 you know
1:08:39 because
1:08:39 it’s
1:08:39 like okay
1:08:39 who
1:08:40 can build
1:08:41 like rockets
1:08:41 and cars
1:08:42 and this
1:08:42 and that
1:08:42 and the
1:08:42 other
1:08:43 but the
1:08:43 reason I
1:08:43 thought he
1:08:44 was like
1:08:44 Isaac
1:08:44 Newton
1:08:44 was
1:08:45 at the
1:08:46 end of
1:08:46 Isaac
1:08:46 Newton’s
1:08:46 life
1:08:47 um
1:08:48 but I
1:08:48 think
1:08:48 he was
1:08:49 he was
1:08:49 in his
1:08:49 late
1:08:50 60s
1:08:51 uh
1:08:51 maybe
1:08:52 like
1:08:52 67
1:08:53 68
1:08:53 and this
1:08:54 is you
1:08:54 know
1:08:54 for those
1:08:54 of you
1:08:54 don’t
1:08:55 know
1:08:56 Isaac
1:08:56 Newton
1:08:56 like
1:08:56 he
1:08:56 figured
1:08:57 out
1:08:57 how
1:08:57 the
1:08:57 entire
1:08:58 world
1:08:58 works
1:08:58 and
1:08:59 wrote
1:08:59 it
1:08:59 down
1:08:59 in
1:08:59 a
1:08:59 book
1:09:00 um
1:09:01 you
1:09:01 know
1:09:01 called
1:09:02 Principia
1:09:02 Mathematica
1:09:03 which is
1:09:03 probably
1:09:04 the most
1:09:04 amazing
1:09:05 work
1:09:05 in the
1:09:05 history
1:09:05 of
1:09:05 science
1:09:06 uh
1:09:07 and he
1:09:07 did
1:09:07 it
1:09:08 like
1:09:08 entirely
1:09:08 by
1:09:09 himself
1:09:10 like
1:09:11 he didn’t
1:09:11 even talk
1:09:11 to anybody
1:09:12 at this
1:09:12 you know
1:09:12 at the
1:09:13 time he
1:09:13 wrote it
1:09:13 I think
1:09:14 he was
1:09:14 trying to
1:09:15 figure out
1:09:16 what God
1:09:16 was or
1:09:16 something
1:09:17 like that
1:09:18 um
1:09:19 but
1:09:19 as you
1:09:19 do
1:09:20 but
1:09:20 so he
1:09:20 gets to
1:09:21 be
1:09:21 you know
1:09:21 like
1:09:22 in his
1:09:22 late
1:09:22 60s
1:09:23 and
1:09:23 the
1:09:24 bank
1:09:24 of
1:09:24 England
1:09:24 has
1:09:25 a
1:09:25 crisis
1:09:25 which
1:09:26 is
1:09:26 causing
1:09:27 a
1:09:27 huge
1:09:27 crisis
1:09:27 for
1:09:28 the
1:09:28 whole
1:09:28 country
1:09:29 um
1:09:29 which
1:09:29 is
1:09:30 uh
1:09:30 there’s
1:09:31 a
1:09:31 giant
1:09:32 counterfeiting
1:09:32 problem
1:09:33 so the
1:09:33 currency
1:09:33 is
1:09:34 going to
1:09:34 be
1:09:34 undermined
1:09:35 and
1:09:35 uh
1:09:36 England’s
1:09:36 going to
1:09:36 basically
1:09:37 go
1:09:37 bankrupt
1:09:38 and
1:09:39 so
1:09:39 they
1:09:39 had
1:09:39 no
1:09:40 idea
1:09:40 what
1:09:40 to
1:09:40 do
1:09:41 about
1:09:41 it
1:09:41 um
1:09:41 so
1:09:42 they
1:09:42 call
1:09:42 Isaac
1:09:42 Newton
1:09:43 because
1:09:43 he’s
1:09:43 the
1:09:43 smartest
1:09:44 man
1:09:44 in the
1:09:44 world
1:09:45 of
1:09:45 course
1:09:45 you’re
1:09:45 going
1:09:45 to
1:09:45 call
1:09:46 him
1:09:46 so
1:09:46 Isaac
1:09:47 Newton
1:09:48 67
1:09:48 year
1:09:48 old
1:09:49 like
1:09:50 hermit
1:09:51 physicist
1:09:52 goes
1:09:52 in
1:09:53 and
1:09:53 he
1:09:53 says
1:09:54 okay
1:09:54 I
1:09:54 can
1:09:54 help
1:09:55 with
1:09:55 the
1:09:55 problem
1:09:56 make
1:09:56 me
1:09:56 CEO
1:09:57 of
1:09:57 the
1:09:57 mint
1:09:57 so
1:09:58 they
1:09:58 make
1:09:58 him
1:09:58 CEO
1:09:58 of
1:09:59 the
1:09:59 mint
1:09:59 you
1:09:59 know
1:09:59 kind
1:10:00 of
1:10:00 head
1:10:00 of
1:10:00 doge
1:10:00 whatever
1:10:02 uh
1:10:02 and
1:10:03 he
1:10:04 reorganizes
1:10:04 the
1:10:04 mint
1:10:05 in like
1:10:05 a week
1:10:06 and
1:10:06 then
1:10:06 fixes
1:10:06 the
1:10:07 technology
1:10:07 in a
1:10:07 month
1:10:08 and
1:10:09 completely
1:10:09 makes
1:10:09 it
1:10:10 impossible
1:10:10 to
1:10:10 counterfeit
1:10:11 then
1:10:12 he
1:10:12 becomes
1:10:13 a
1:10:13 private
1:10:13 eye
1:10:15 and
1:10:15 goes
1:10:15 into
1:10:16 all
1:10:16 the
1:10:16 pubs
1:10:16 where
1:10:16 the
1:10:17 counterfeiters
1:10:17 are
1:10:18 arrests
1:10:18 all
1:10:18 of
1:10:18 them
1:10:19 then
1:10:20 he
1:10:20 learns
1:10:20 the
1:10:20 law
1:10:21 and
1:10:21 becomes
1:10:21 the
1:10:22 prosecutor
1:10:23 and
1:10:23 prosecutes
1:10:24 all
1:10:24 the
1:10:24 counterfeiters
1:10:25 and has
1:10:26 100%
1:10:26 conviction
1:10:27 record
1:10:28 and
1:10:28 that
1:10:29 by
1:10:29 the
1:10:29 way
1:10:30 that’s
1:10:30 Elon
1:10:30 so
1:10:32 if
1:10:32 you
1:10:33 I
1:10:33 didn’t
1:10:33 know
1:10:34 that
1:10:34 part
1:10:34 of
1:10:34 his
1:10:34 story
1:10:35 oh
1:10:35 yeah
1:10:35 yeah
1:10:36 so
1:10:36 it’s
1:10:36 an
1:10:37 amazing
1:10:37 thing
1:10:39 and
1:10:39 if
1:10:39 you
1:10:39 look
1:10:40 at
1:10:40 Elon
1:10:40 and
1:10:41 Doge
1:10:41 to
1:10:42 me
1:10:42 the
1:10:42 most
1:10:43 remarkable
1:10:43 thing
1:10:43 about
1:10:44 Doge
1:10:45 is
1:10:46 how
1:10:46 he’s
1:10:47 done
1:10:47 it
1:10:47 so
1:10:48 if
1:10:48 you
1:10:48 or
1:10:49 I
1:10:50 were
1:10:50 to
1:10:51 say
1:10:52 okay
1:10:53 let’s
1:10:53 go in
1:10:54 and
1:10:54 kind
1:10:54 of
1:10:55 get
1:10:55 the
1:10:55 waste
1:10:55 and
1:10:56 fraud
1:10:56 out
1:10:56 of
1:10:56 the
1:10:56 government
1:10:57 we
1:10:57 would
1:10:58 like
1:10:58 audit
1:10:58 the
1:10:59 departments
1:10:59 or
1:10:59 this
1:10:59 and
1:11:00 that
1:11:00 the
1:11:00 other
1:11:00 and
1:11:00 so
1:11:01 forth
1:11:02 no
1:11:02 no
1:11:02 no
1:11:03 like
1:11:03 that’s
1:11:03 how
1:11:03 he
1:11:06 like
1:11:07 how
1:11:07 do
1:11:07 the
1:11:07 checks
1:11:07 go
1:11:08 out
1:11:10 like
1:11:11 how
1:11:11 is
1:11:11 the
1:11:11 system
1:11:12 designed
1:11:12 like
1:11:12 when
1:11:13 does
1:11:13 the
1:11:13 money
1:11:13 leave
1:11:13 the
1:11:14 building
1:11:15 and
1:11:15 then
1:11:15 oh
1:11:16 all
1:11:16 comes
1:11:16 out
1:11:16 of
1:11:17 one
1:11:17 system
1:11:18 let
1:11:18 me
1:11:18 have
1:11:19 access
1:11:19 to
1:11:19 that
1:11:19 system
1:11:20 and
1:11:20 I’ll
1:11:20 look
1:11:20 at
1:11:20 all
1:11:20 the
1:11:21 payments
1:11:22 I’m
1:11:22 not
1:11:22 asking
1:11:22 anybody
1:11:23 what
1:11:23 they’re
1:11:23 spending
1:11:24 I’m
1:11:24 looking
1:11:25 at
1:11:25 what
1:11:25 they’re
1:11:25 spending
1:11:26 like
1:11:26 I’m
1:11:26 getting
1:11:26 to
1:11:26 ground
1:11:27 truth
1:11:27 and
1:11:28 then
1:11:28 I’m
1:11:28 going
1:11:28 to
1:11:28 work
1:11:28 my
1:11:29 way
1:11:29 backwards
1:11:29 from
1:11:30 there
1:11:30 and
1:11:30 he’s
1:11:31 probably
1:11:32 and
1:11:32 you know
1:11:33 so
1:11:34 not only
1:11:34 is he
1:11:35 not
1:11:35 unqualified
1:11:36 he’s
1:11:36 maybe the
1:11:37 only person
1:11:37 qualified
1:11:38 to figure
1:11:38 out
1:11:38 like
1:11:39 how
1:11:46 you know
1:11:46 he’s
1:11:47 just
1:11:47 a
1:11:47 very
1:11:48 unique
1:11:48 individual
1:11:51 he’s
1:11:51 also
1:11:52 a troll
1:11:53 he also
1:11:53 likes
1:11:54 upsetting
1:11:54 people
1:11:55 I get
1:11:55 all
1:11:56 that
1:11:57 but
1:11:58 what
1:11:58 he
1:11:58 brings
1:11:58 to
1:11:58 the
1:11:59 table
1:11:59 is
1:11:59 pretty
1:11:59 interesting
1:12:00 I would
1:12:00 say
1:12:01 I would
1:12:01 say
1:12:02 very
1:12:02 extraordinary
1:12:03 you
1:12:04 have
1:12:04 also
1:12:04 written
1:12:04 about
1:12:05 another
1:12:05 extraordinary
1:12:06 historical
1:12:07 figure
1:12:09 from
1:12:09 the
1:12:10 Haitian
1:12:10 revolution
1:12:11 a guy
1:12:11 named
1:12:12 Toussaint
1:12:16 Tell us
1:12:16 about him
1:12:16 because
1:12:16 there’s
1:12:17 something
1:12:17 about
1:12:17 this
1:12:18 moment
1:12:18 about
1:12:19 being
1:12:19 a
1:12:19 master
1:12:20 strategist
1:12:20 about
1:12:22 using
1:12:22 what you
1:12:23 have
1:12:23 being
1:12:24 creative
1:12:24 that
1:12:25 feels
1:12:25 like
1:12:25 it’s
1:12:26 very
1:12:27 apropos
1:12:27 to
1:12:28 this
1:12:28 moment
1:12:31 what
1:12:31 made
1:12:32 his
1:12:32 story
1:12:32 special
1:12:33 yeah
1:12:33 so
1:12:34 Toussaint
1:12:34 was
1:12:35 another
1:12:35 one
1:12:35 of
1:12:35 these
1:12:36 characters
1:12:36 in
1:12:36 history
1:12:37 like
1:12:37 there
1:12:37 are
1:12:37 certain
1:12:38 I
1:12:39 call
1:12:39 them
1:12:39 like
1:12:39 once
1:12:39 in
1:12:40 every
1:12:40 400
1:12:40 year
1:12:41 type
1:12:41 people
1:12:43 where
1:12:43 you
1:12:43 just
1:12:43 don’t
1:12:43 see
1:12:43 them
1:12:44 that
1:12:44 often
1:12:44 but
1:12:45 so
1:12:46 it
1:12:46 turns
1:12:46 out
1:12:46 like
1:12:46 in
1:12:47 the
1:12:47 history
1:12:47 of
1:12:48 humanity
1:12:50 there’s
1:12:50 been
1:12:51 one
1:12:51 kind
1:12:51 of
1:12:52 successful
1:12:53 slave
1:12:53 revolt
1:12:53 that
1:12:54 like
1:12:54 entered
1:12:55 in
1:12:55 an
1:12:55 independent
1:12:56 state
1:12:57 which
1:12:57 you
1:12:57 know
1:12:57 if
1:12:57 you
1:12:58 think
1:12:58 about
1:12:59 the
1:12:59 history
1:12:59 of
1:13:00 slavery
1:13:00 which
1:13:00 goes
1:13:01 back
1:13:02 thousands
1:13:02 of
1:13:02 years
1:13:02 really
1:13:03 kind
1:13:03 of
1:13:04 from
1:13:04 the
1:13:05 beginning
1:13:05 of
1:13:05 written
1:13:06 history
1:13:06 like
1:13:07 we’ve
1:13:07 had
1:13:07 slavery
1:13:07 so
1:13:08 it’s
1:13:09 like
1:13:09 a
1:13:09 pretty
1:13:10 old
1:13:11 time
1:13:11 construct
1:13:12 and
1:13:13 there’s
1:13:13 a lot
1:13:13 of
1:13:14 motivation
1:13:14 to have
1:13:14 a
1:13:14 revolt
1:13:15 if
1:13:15 you’re
1:13:15 a
1:13:15 slave
1:13:17 but
1:13:17 why
1:13:17 only
1:13:18 one
1:13:18 successful
1:13:18 one
1:13:18 and
1:13:19 it
1:13:19 turns
1:13:19 out
1:13:19 it’s
1:13:20 it’s
1:13:20 really
1:13:21 hard
1:13:23 you know
1:13:24 for
1:13:25 to
1:13:25 generate
1:13:26 an
1:13:26 effective
1:13:27 revolt
1:13:27 if
1:13:27 you’re
1:13:28 slaves
1:13:28 because
1:13:30 slave
1:13:30 culture
1:13:31 is
1:13:32 difficult
1:13:32 because
1:13:32 you
1:13:32 don’t
1:13:32 own
1:13:33 right
1:13:33 if
1:13:33 you
1:13:34 don’t
1:13:34 have
1:13:36 any
1:13:37 sense
1:13:37 of
1:13:38 you know
1:13:38 owning
1:13:38 anything
1:13:39 you don’t
1:13:39 own
1:13:39 your
1:13:39 own
1:13:40 will
1:13:40 right
1:13:40 like
1:13:40 you
1:13:41 are
1:13:42 at
1:13:42 the
1:13:42 kind
1:13:43 of
1:13:43 pleasure
1:13:44 of
1:13:44 who’s
1:13:45 ever
1:13:45 running
1:13:45 things
1:13:46 so
1:13:47 long
1:13:47 term
1:13:48 thinking
1:13:48 doesn’t
1:13:48 make
1:13:49 sense
1:13:50 and
1:13:51 what
1:13:51 it
1:13:52 because
1:13:52 like
1:13:52 why
1:13:53 plan
1:13:53 for
1:13:53 next
1:13:53 week
1:13:54 it
1:13:54 doesn’t
1:13:55 matter
1:13:55 what
1:13:55 you
1:13:55 plan
1:13:56 like
1:13:56 it’s
1:13:56 not
1:13:57 yours
1:13:58 so
1:13:58 everything’s
1:13:58 going to be
1:13:59 very short
1:13:59 term
1:14:00 and short
1:14:00 termism
1:14:01 is
1:14:02 difficult
1:14:02 in a
1:14:02 military
1:14:03 context
1:14:04 because
1:14:06 in order
1:14:06 to have
1:14:06 an effective
1:14:07 military
1:14:07 there needs
1:14:08 to be
1:14:09 a
1:14:09 trust
1:14:10 right
1:14:10 like
1:14:10 a
1:14:10 trust
1:14:12 you know
1:14:12 you have
1:14:12 to be
1:14:12 able
1:14:12 to
1:14:13 trust
1:14:13 people
1:14:13 to
1:14:14 execute
1:14:14 the
1:14:14 order
1:14:14 like
1:14:15 I give
1:14:15 an
1:14:15 order
1:14:16 it’s
1:14:16 kind
1:14:16 of
1:14:16 like
1:14:17 the
1:14:17 Byzantine
1:14:17 generals
1:14:18 problem
1:14:19 to go
1:14:19 back
1:14:19 to
1:14:19 crypto
1:14:21 where
1:14:22 like
1:14:23 I have
1:14:23 to
1:14:23 trust
1:14:23 that
1:14:24 you’re
1:14:24 going
1:14:24 to
1:14:24 do
1:14:24 the
1:14:24 order
1:14:25 you
1:14:25 have
1:14:25 to
1:14:25 trust
1:14:25 that
1:14:26 I’m
1:14:26 giving
1:14:26 the
1:14:26 correct
1:14:27 order
1:14:28 but
1:14:28 trust
1:14:28 is
1:14:28 a
1:14:29 long
1:14:29 term
1:14:29 idea
1:14:29 because
1:14:30 it
1:14:30 comes
1:14:30 from
1:14:30 okay
1:14:31 I’m
1:14:31 going to
1:14:31 do
1:14:32 something
1:14:32 for
1:14:32 you
1:14:32 today
1:14:33 because
1:14:33 I
1:14:33 trust
1:14:33 that
1:14:33 down
1:14:33 the
1:14:34 line
1:14:34 you’ll
1:14:34 do
1:14:34 something
1:14:34 for
1:14:35 me
1:14:35 that
1:14:36 doesn’t
1:14:36 really
1:14:36 exist
1:14:36 in
1:14:47 to
1:14:47 running
1:14:47 a
1:14:48 successful
1:14:48 revolution
1:14:49 and
1:14:49 then
1:14:49 if
1:14:49 you
1:14:49 look
1:14:49 at
1:14:50 Haiti
1:14:50 at
1:14:50 the
1:14:51 time
1:14:53 you
1:14:53 had
1:14:53 the
1:14:54 French
1:14:54 army
1:14:54 the
1:14:55 British
1:14:55 army
1:14:57 and
1:14:57 the
1:14:57 Spanish
1:14:58 army
1:14:58 all
1:14:58 in
1:14:58 there
1:14:59 fighting
1:14:59 for it
1:15:00 so
1:15:00 really
1:15:01 well
1:15:01 developed
1:15:03 the strongest
1:15:04 militaries
1:15:04 of the
1:15:04 era
1:15:05 all
1:15:06 in
1:15:06 that
1:15:06 region
1:15:07 all
1:15:07 very
1:15:08 interested
1:15:08 in
1:15:09 the
1:15:09 sugar
1:15:11 which
1:15:11 was
1:15:12 quite
1:15:12 valuable
1:15:12 at
1:15:12 the
1:15:13 time
1:15:13 so
1:15:14 how
1:15:14 in
1:15:14 the
1:15:14 world
1:15:14 would
1:15:15 you ever
1:15:15 get out
1:15:15 of
1:15:15 that
1:15:17 and
1:15:17 it
1:15:17 turned
1:15:17 out
1:15:18 he
1:15:19 was
1:15:20 probably
1:15:20 the
1:15:21 great
1:15:21 cultural
1:15:22 genius
1:15:22 of
1:15:22 the
1:15:22 last
1:15:23 you know
1:15:25 maybe
1:15:25 in
1:15:25 history
1:15:26 but
1:15:26 certainly
1:15:26 the
1:15:26 last
1:15:27 several
1:15:27 hundred
1:15:28 years
1:15:30 and
1:15:31 he
1:15:32 was
1:15:32 able
1:15:33 because
1:15:34 he
1:15:34 was
1:15:34 a
1:15:35 person
1:15:35 who
1:15:35 although
1:15:35 he
1:15:36 was
1:15:36 born
1:15:36 a
1:15:36 slave
1:15:37 was
1:15:37 very
1:15:38 very
1:15:38 integrated
1:15:38 into
1:15:39 European
1:15:39 culture
1:15:40 because
1:15:40 he
1:15:40 was
1:15:40 so
1:15:41 smart
1:15:42 and
1:15:43 so
1:15:44 the
1:15:44 person
1:15:44 who
1:15:45 ran
1:15:45 the
1:15:45 plantation
1:15:47 kind
1:15:47 of
1:15:48 took
1:15:48 him
1:15:48 to
1:15:48 all
1:15:48 the
1:15:49 diplomatic
1:15:49 meetings
1:15:50 around
1:15:50 and
1:15:50 so
1:15:51 forth
1:15:51 and
1:15:51 he
1:15:52 got
1:15:52 very
1:15:52 involved
1:15:53 and
1:15:53 kind
1:15:53 of
1:15:53 mastered
1:15:55 European
1:15:55 culture
1:15:57 so to
1:15:57 speak
1:15:57 and
1:15:57 the
1:15:57 different
1:15:58 subtleties
1:15:58 around
1:15:59 it
1:16:00 and
1:16:00 he
1:16:01 started
1:16:01 adopting
1:16:01 those
1:16:02 things
1:16:02 and
1:16:02 applying
1:16:03 them
1:16:03 to
1:16:03 his
1:16:04 leadership
1:16:04 and
1:16:05 then
1:16:06 furthermore
1:16:07 incorporated
1:16:08 Europeans
1:16:09 into
1:16:09 the slave
1:16:10 army
1:16:10 so he
1:16:10 would
1:16:11 capture
1:16:12 you know
1:16:13 he
1:16:13 would
1:16:14 defeat
1:16:14 the
1:16:14 Spanish
1:16:15 capture
1:16:16 some
1:16:16 guys
1:16:16 rather
1:16:17 than
1:16:17 kill
1:16:17 them
1:16:17 he
1:16:18 incorporated
1:16:19 the best
1:16:19 leaders
1:16:19 into
1:16:20 his
1:16:20 army
1:16:21 and
1:16:21 he
1:16:21 built
1:16:21 this
1:16:22 very
1:16:23 advanced
1:16:23 hybrid
1:16:24 fighting
1:16:24 system
1:16:24 where
1:16:25 they
1:16:25 used
1:16:25 a lot
1:16:25 of
1:16:25 the
1:16:26 guerrilla
1:16:27 techniques
1:16:28 that
1:16:28 he
1:16:29 had
1:16:29 brought
1:16:29 over
1:16:29 from
1:16:30 Africa
1:16:31 and
1:16:32 then
1:16:32 he
1:16:33 had
1:16:34 combined
1:16:34 that
1:16:35 with
1:16:35 some
1:16:36 of
1:16:36 the
1:16:37 more
1:16:38 regimented
1:16:40 discipline
1:16:41 strategies
1:16:42 of
1:16:42 the
1:16:43 Europeans
1:16:44 and
1:16:44 in
1:16:44 building
1:16:44 all
1:16:44 that
1:16:45 he
1:16:46 ended
1:16:46 up
1:16:46 building
1:16:46 this
1:16:47 massive
1:16:47 army
1:16:48 and
1:16:49 defeated
1:16:49 Napoleon
1:16:50 and
1:16:51 everyone
1:16:51 else
1:16:51 and
1:16:52 it
1:16:52 was
1:16:52 just
1:16:53 quite
1:16:53 a
1:16:54 remarkable
1:16:54 story
1:16:55 about
1:16:55 how
1:16:55 he
1:16:56 figured
1:16:56 everything
1:16:57 out
1:16:57 from
1:16:57 his
1:16:58 principles
1:16:58 and
1:16:58 in
1:16:58 a
1:16:59 way
1:16:59 that
1:16:59 was
1:17:01 very
1:17:02 much
1:17:02 like
1:17:03 Elon
1:17:03 in
1:17:04 that
1:17:04 sense
1:17:05 one
1:17:05 of
1:17:06 the
1:17:06 things
1:17:06 I heard
1:17:06 you talk
1:17:07 about
1:17:10 that
1:17:10 I
1:17:10 thought
1:17:11 was
1:17:11 so
1:17:11 ingenious
1:17:12 was
1:17:12 he
1:17:12 would
1:17:12 basically
1:17:13 use
1:17:13 song
1:17:14 and
1:17:15 sound
1:17:16 as
1:17:16 like
1:17:16 encrypted
1:17:17 language
1:17:18 it’s
1:17:19 really
1:17:19 yeah
1:17:19 so
1:17:20 that
1:17:20 was
1:17:21 like
1:17:21 a
1:17:21 very
1:17:21 cool
1:17:22 thing
1:17:22 so
1:17:23 right
1:17:24 remember
1:17:24 that
1:17:24 this
1:17:24 is
1:17:24 in
1:17:25 the
1:17:25 days
1:17:25 before
1:17:26 telephony
1:17:26 or the
1:17:27 internet
1:17:27 or any
1:17:28 of
1:17:28 these
1:17:28 things
1:17:28 you know
1:17:28 it’s
1:17:29 pre
1:17:29 Alexander
1:17:30 Graham Bell
1:17:30 and all
1:17:30 that
1:17:30 kind
1:17:30 of
1:17:31 thing
1:17:32 and
1:17:33 so
1:17:34 you know
1:17:34 they were
1:17:34 literally
1:17:35 you know
1:17:35 the Europeans
1:17:36 were on
1:17:36 like
1:17:37 you know
1:17:37 notes
1:17:38 carrier pigeons
1:17:40 guys running
1:17:41 you know
1:17:41 back and
1:17:42 forth
1:17:42 and so
1:17:42 forth
1:17:43 and so
1:17:43 as a
1:17:43 result
1:17:45 you know
1:17:45 you kind
1:17:45 of needed
1:17:46 the army
1:17:47 together
1:17:47 in one
1:17:48 place
1:17:48 just
1:17:48 so you
1:17:49 could
1:17:49 communicate
1:17:49 the
1:17:50 order
1:17:52 Toussaint
1:17:53 basically
1:17:54 you know
1:17:55 had these
1:17:57 drummers
1:17:57 and these
1:17:58 songs
1:17:59 which
1:17:59 he could
1:18:00 put on
1:18:00 top of
1:18:00 like
1:18:02 the hill
1:18:02 who could
1:18:03 be very
1:18:03 very loud
1:18:04 and then
1:18:04 he would
1:18:04 separate his
1:18:05 army
1:18:05 you know
1:18:06 into like
1:18:06 six or
1:18:07 seven groups
1:18:09 but in
1:18:10 the song
1:18:11 would be
1:18:11 embedded
1:18:13 the order
1:18:13 of when
1:18:14 to attack
1:18:14 and you
1:18:15 know
1:18:15 when to
1:18:16 retreat
1:18:16 and all
1:18:17 these kinds
1:18:17 of things
1:18:17 so he
1:18:18 had this
1:18:18 like
1:18:18 super
1:18:19 advanced
1:18:20 you know
1:18:20 wide area
1:18:21 communication
1:18:22 system
1:18:23 that nobody
1:18:24 else had
1:18:24 and that
1:18:24 you know
1:18:25 that was a
1:18:26 big advantage
1:18:27 for him
1:18:28 yeah that to
1:18:29 me the reason
1:18:29 that that
1:18:30 comes up
1:18:30 for me
1:18:30 now is
1:18:31 we have
1:18:31 all these
1:18:32 new
1:18:32 technologies
1:18:33 that are
1:18:33 coming
1:18:33 online
1:18:33 and the
1:18:34 person
1:18:34 that’s
1:18:34 going to
1:18:34 be able
1:18:34 to get
1:18:35 outside
1:18:35 that box
1:18:37 and see
1:18:37 something
1:18:38 new
1:18:38 and fresh
1:18:39 is going
1:18:39 to be able
1:18:39 to use
1:18:40 this in
1:18:40 totally
1:18:41 different
1:18:41 ways
1:18:41 and while
1:18:42 in the
1:18:43 final analysis
1:18:43 I think
1:18:44 you and I
1:18:44 see it
1:18:44 very differently
1:18:45 in terms
1:18:46 of AI’s
1:18:46 ability
1:18:47 to ultimately
1:18:48 gobble up
1:18:49 what humans
1:18:49 can do
1:18:50 but right
1:18:51 now
1:18:52 AI is this
1:18:53 incredible
1:18:54 tool that
1:18:54 as an
1:18:54 entrepreneur
1:18:55 for me
1:18:56 it has
1:18:56 been
1:18:58 ridiculously
1:18:59 exciting
1:19:00 to one
1:19:01 see how
1:19:01 much farther
1:19:02 each of my
1:19:02 employees
1:19:03 can push
1:19:03 their own
1:19:04 abilities
1:19:04 by using
1:19:04 AI
1:19:05 and then
1:19:05 it does
1:19:06 not take
1:19:06 much to
1:19:07 prognosticate
1:19:08 out you
1:19:08 know 12
1:19:09 18 months
1:19:10 to understand
1:19:11 where the
1:19:11 tools are
1:19:12 going to be
1:19:12 and how
1:19:12 much more
1:19:13 they’re going
1:19:13 to let
1:19:13 you do
1:19:13 because
1:19:13 we’re
1:19:14 largely
1:19:14 an
1:19:15 entertainment
1:19:15 company
1:19:16 so for
1:19:17 us to
1:19:17 look at
1:19:17 that
1:19:18 and just
1:19:19 the
1:19:20 revolutionary
1:19:21 changes
1:19:21 but you
1:19:22 can’t be
1:19:22 trapped
1:19:23 inside the
1:19:23 old way
1:19:23 of thinking
1:19:24 you’ve got
1:19:25 to like
1:19:25 you said
1:19:25 build up
1:19:26 from first
1:19:26 principles
1:19:27 yeah it’s
1:19:27 a new
1:19:28 creative
1:19:28 canvas
1:19:29 I think
1:19:29 that’s
1:19:29 like a
1:19:30 really great
1:19:30 way of
1:19:30 thinking
1:19:31 about it
1:19:31 in that
1:19:33 it’s
1:19:33 like
1:19:34 well is
1:19:34 your
1:19:35 creativity
1:19:36 going to
1:19:37 be used
1:19:37 on
1:19:39 you know
1:19:40 kind of
1:19:41 frame
1:19:42 by frame
1:19:42 editing
1:19:43 of like
1:19:44 a video
1:19:45 or will
1:19:45 it be
1:19:46 thinking
1:19:46 of like
1:19:47 incredible
1:19:47 new
1:19:47 things
1:19:48 you can
1:19:48 do
1:19:48 in a
1:19:48 video
1:19:49 ad
1:19:50 that
1:19:50 you
1:19:50 could
1:19:50 never
1:19:50 do
1:19:51 before
1:19:51 and
1:19:51 have
1:19:52 the
1:19:52 AI
1:19:52 do
1:19:52 that
1:19:52 for
1:19:53 you
1:19:53 you know
1:19:53 like
1:19:54 and so
1:19:54 it’s
1:19:54 a little
1:19:54 bit
1:19:55 of a
1:19:55 readjustment
1:19:56 of
1:19:57 where
1:19:58 you put
1:19:58 your
1:19:59 creative
1:20:00 energy
1:20:00 into
1:20:01 and
1:20:01 the
1:20:01 things
1:20:01 that
1:20:01 are
1:20:02 possible
1:20:02 and
1:20:02 so
1:20:02 forth
1:20:03 and
1:20:03 I
1:20:03 think
1:20:03 that’s
1:20:04 you know
1:20:04 we’re
1:20:04 really
1:20:05 seeing
1:20:05 that
1:20:05 across
1:20:05 the
1:20:05 board
1:20:10 is
1:20:10 this
1:20:11 going
1:20:11 to
1:20:11 mean
1:20:12 you
1:20:12 know
1:20:12 like
1:20:13 you
1:20:13 don’t
1:20:13 have
1:20:13 human
1:20:14 investors
1:20:14 anymore
1:20:15 and
1:20:15 it’s
1:20:15 actually
1:20:15 been
1:20:16 like
1:20:16 totally
1:20:16 the
1:20:17 opposite
1:20:17 like
1:20:18 instead
1:20:19 of
1:20:19 this
1:20:20 like
1:20:22 painstakingly
1:20:23 collecting
1:20:24 you know
1:20:24 all
1:20:24 all
1:20:25 the
1:20:25 data
1:20:25 needed
1:20:25 to
1:20:26 put
1:20:26 the
1:20:26 investment
1:20:27 memo
1:20:27 together
1:20:28 like
1:20:28 yeah
1:20:28 it
1:20:28 just
1:20:28 does
1:20:29 that
1:20:29 for
1:20:29 you
1:20:29 and
1:20:30 then
1:20:30 you’re
1:20:30 just
1:20:30 thinking
1:20:30 about
1:20:31 like
1:20:31 okay
1:20:31 what
1:20:31 are
1:20:32 the
1:20:32 like
1:20:32 the
1:20:33 really
1:20:33 compelling
1:20:33 things
1:20:34 about
1:20:34 this
1:20:35 or
1:20:35 rather
1:20:36 than
1:20:36 you know
1:20:37 trying
1:20:37 to
1:20:37 track
1:20:38 every
1:20:39 entrepreneur
1:20:40 and
1:20:41 like
1:20:41 great
1:20:41 engineer
1:20:42 in our
1:20:42 database
1:20:43 the AI
1:20:43 is
1:20:43 just
1:20:44 tracking
1:20:44 all
1:20:44 those
1:20:44 people
1:20:45 and
1:20:45 letting
1:20:45 you
1:20:45 know
1:20:45 hey
1:20:46 that
1:20:46 guy
1:20:46 just
1:20:46 updated
1:20:47 his
1:20:47 LinkedIn
1:20:47 profile
1:20:48 or
1:20:54 like
1:20:54 a
1:20:54 much
1:20:55 more
1:20:55 kind
1:20:56 of
1:20:56 fun
1:20:56 part
1:20:56 of
1:20:56 the
1:20:57 game
1:20:58 and
1:20:59 and
1:21:00 so
1:21:00 you know
1:21:00 look
1:21:01 I
1:21:01 would
1:21:02 say
1:21:02 the
1:21:03 best
1:21:03 predictor
1:21:04 of
1:21:04 kind
1:21:04 of
1:21:05 how
1:21:05 things
1:21:05 are
1:21:05 going
1:21:05 to
1:21:06 go
1:21:06 are
1:21:07 more
1:21:07 like
1:21:07 what’s
1:21:08 happening
1:21:08 now
1:21:08 than
1:21:10 like
1:21:10 the
1:21:10 most
1:21:11 dystopian
1:21:11 view
1:21:12 of it
1:21:12 that
1:21:13 we
1:21:13 can
1:21:14 possibly
1:21:14 think
1:21:14 of
1:21:15 which
1:21:15 I
1:21:15 think
1:21:16 is
1:21:16 where
1:21:16 a lot
1:21:16 of
1:21:16 people
1:21:16 go
1:21:17 to
1:21:17 and
1:21:17 like
1:21:17 I
1:21:17 said
1:21:17 I
1:21:17 think
1:21:17 some
1:21:18 of
1:21:18 that’s
1:21:18 the
1:21:18 name
1:21:19 you
1:21:19 know
1:21:20 artificial
1:21:21 intelligence
1:21:21 we
1:21:22 hate
1:21:23 everything
1:21:23 artificial
1:21:24 so
1:21:25 why
1:21:25 do
1:21:25 we
1:21:26 name
1:21:26 it
1:21:26 artificial
1:21:28 that’s
1:21:28 too
1:21:28 true
1:21:29 Ben
1:21:29 I’ve
1:21:30 enjoyed
1:21:30 every
1:21:30 minute
1:21:30 of
1:21:31 this
1:21:31 where
1:21:31 can
1:21:31 people
1:21:31 keep
1:21:32 up
1:21:32 with
1:21:32 you
1:21:33 yeah
1:21:33 well
1:21:33 I
1:21:34 am
1:21:35 B
1:21:35 Horowitz
1:21:35 on
1:21:36 AX
1:21:37 and
1:21:38 you
1:21:39 know
1:21:40 that’s
1:21:40 probably
1:21:40 the
1:21:40 best
1:21:41 thing
1:21:41 we’re
1:21:42 a16z.com
1:21:43 and
1:21:45 hope
1:21:45 you
1:21:45 enjoyed
1:21:45 it
1:21:46 and
1:21:46 that
1:21:46 was
1:21:47 great
1:21:47 fun
1:21:47 good
1:21:48 fun
1:21:48 catching
1:21:48 it
1:21:49 was
1:21:49 indeed
1:21:49 and
1:21:49 then
1:21:49 you
1:21:49 also
1:21:50 have
1:21:51 multiple
1:21:51 books
1:21:52 that
1:21:52 people
1:21:52 can
1:21:52 read
1:21:52 that
1:21:53 are
1:21:53 extraordinarily
1:21:54 well
1:21:55 respected
1:21:55 in the
1:21:55 field
1:21:56 so
1:21:57 also
1:21:57 thank
1:21:57 you
1:21:57 for
1:21:57 those
1:21:58 absolutely
1:22:00 awesome
1:22:00 thanks
1:22:00 all right
1:22:01 well thank
1:22:01 you brother
1:22:02 I appreciate
1:22:02 it
1:22:02 all right
1:22:02 everybody
1:22:03 if you
1:22:03 have
1:22:03 not
1:22:04 already
1:22:04 be
1:22:04 sure
1:22:04 to
1:22:04 subscribe
1:22:05 and
1:22:05 until
1:22:05 next
1:22:05 time
1:22:06 my
1:22:06 friends
1:22:06 be
1:22:07 legendary
1:22:07 take
1:22:07 care
1:22:08 peace
1:22:20 we’ve
1:22:20 got
1:22:20 more
1:22:20 great
1:22:21 conversations
1:22:21 coming
1:22:21 your
1:22:21 way
1:22:22 see
1:22:22 you
1:22:22 next
1:22:22 time