AI transcript
0:00:02 – Where do you see SEO going in the world of AI?
0:00:04 – If Warren Buffett was an internet marketer,
0:00:08 he would be like all about SEO 2.0 right now.
0:00:11 It’s all about how do you take a stranger
0:00:13 and make them a raving fan?
0:00:16 If you can do those things, you win at business.
0:00:19 (upbeat music)
0:00:24 – When all your marketing team does is put out fires,
0:00:25 they burn out.
0:00:28 But with HubSpot, they can achieve their best results
0:00:29 without the stress.
0:00:32 Tap into HubSpot’s collection of AI tools,
0:00:35 breeze, to pinpoint leads, capture attention,
0:00:38 and access all your data in one place.
0:00:39 Keep your marketers cool
0:00:42 and your campaign results hotter than ever.
0:00:45 Visit hubspot.com/marketers to learn more.
0:00:48 (upbeat music)
0:00:50 – Hey, welcome to the Next Wave podcast.
0:00:51 I’m Matt Wolf.
0:00:53 I’m here with Nathan Lanz.
0:00:56 And today we’ve got an amazing returning guest
0:01:01 in Greg Eisenberg, he’s the host of the Startup Ideas podcast.
0:01:04 He’s got a holding company called Elite Checkout,
0:01:07 which does over $10 million in revenue per year.
0:01:09 He runs the boring marketing company
0:01:12 where he helps other companies do SEO.
0:01:14 And in this episode, he’s gonna map out
0:01:18 an exact blueprint that you can follow for your SEO
0:01:21 in this new AI world that we’re going into.
0:01:24 He breaks down the SEO 2.0 strategy.
0:01:27 We also talk about all sorts of hot takes
0:01:30 that he has on Twitter and dive into
0:01:32 why he believes those hot takes are true.
0:01:34 And you’re really gonna enjoy this episode.
0:01:35 So let’s just go ahead and dig right in
0:01:37 with Greg Eisenberg.
0:01:38 Hey, Greg, welcome back to the show.
0:01:40 It’s great to have you on.
0:01:42 – I’m ready, I’m ready for it.
0:01:45 – Awesome, well, let’s just get straight into it.
0:01:47 So you have a couple tweets out there right now
0:01:50 that Nathan and I both had some conversations around
0:01:52 and got us talking and we thought,
0:01:54 these are some fun things to bring Greg on
0:01:56 and talk about on the podcast.
0:02:01 And the first one being the concept of SEO 2.0.
0:02:03 So maybe the best place to start is just like,
0:02:05 what’s your definition of SEO 2.0?
0:02:08 Where do you see SEO going in the world of AI?
0:02:12 – Yeah, I mean, a lot of people call it programmatic SEO.
0:02:14 I call it SEO 2.0.
0:02:19 SEO 1.0 to me is basically, I mean, it’s old school SEO.
0:02:21 It’s basically you go and create content pages.
0:02:26 You get humans to do it, top things to do in New York City.
0:02:28 You go and you know all the different restaurants
0:02:30 and bars and you review it.
0:02:34 And then you just sort of hope that, I guess,
0:02:37 you get backlinks from other websites
0:02:40 and then that’s gonna generate traffic.
0:02:44 And that worked pretty well for a long period of time.
0:02:49 And there’s quite literally dozens of billion dollar companies
0:02:52 that rode the SEO 1.0 wave.
0:02:55 Content companies, social networks like Reddit,
0:02:58 these are companies that have great rank.
0:03:00 And then about a year and a half, two years ago,
0:03:02 I started thinking about, hey,
0:03:06 if AI makes it easier to do two things.
0:03:10 One is to get insights and data in a really efficient way.
0:03:14 And two is creating content and creating web pages
0:03:15 in a really efficient way.
0:03:16 Isn’t that me?
0:03:17 What I mean that you’ll be able to just create
0:03:22 like millions of pages and rank for them?
0:03:25 And that’s really what I mean by SEO 2.0.
0:03:27 I’ll tell you why I think it’s exciting to me
0:03:29 and maybe to people listening,
0:03:34 is the same way that there’s dozens of billion dollar companies
0:03:36 and thousands of million dollar companies
0:03:40 that have rode the wave of SEO 1.0,
0:03:43 I believe the same thing is gonna happen for 2.0.
0:03:45 The people that are gonna be able
0:03:50 to efficiently create these web pages, get ranked,
0:03:51 are gonna do really well.
0:03:53 And I know that there’s someone listening,
0:03:55 especially ’cause this is an AI podcast
0:03:58 that’s thinking themselves, well, okay,
0:04:02 but it doesn’t matter because perplexity is the new Google.
0:04:04 And if perplexity is the new Google,
0:04:08 then you know, you ranking doesn’t really matter
0:04:11 on this search engine that no one’s gonna go to.
0:04:14 And the reality of the situation is,
0:04:16 Google is not dead for,
0:04:18 I know a lot of people like to think that Google is dead,
0:04:20 but Google is not dead.
0:04:25 And there’s still huge opportunities to start ranking.
0:04:27 – You know, the first thing that comes to mind
0:04:31 when I think of like an SEO 2.0 concept
0:04:34 and like AI just sort of being able
0:04:36 to sort of mass make content,
0:04:38 is that I feel like the internet might just get flooded
0:04:40 with a lot of junk content.
0:04:43 I mean, how much of SEO 1.0,
0:04:44 especially in the early days,
0:04:47 was people like writing articles, spinning the articles,
0:04:48 and then it would just like change
0:04:50 a certain percentage of words
0:04:52 and they were doing it for the algorithm
0:04:54 and they weren’t really doing it for the reader.
0:04:55 That’s the first thing that comes to mind,
0:04:59 is like, doesn’t it seem like that might just lead
0:05:01 to a whole bunch of clutter and junk all over the internet
0:05:03 that nobody actually wants to read?
0:05:05 – Yeah, so the reality is,
0:05:09 if you go and create AI content today,
0:05:12 you’re gonna get penalized by Google.
0:05:14 So Google has gotten really good
0:05:19 at basically knowing if something is written by AI.
0:05:20 So what you actually have to do
0:05:23 is you kind of have to like deke out Google.
0:05:27 You have to essentially use humans for 20%
0:05:29 and AI for 80%.
0:05:31 And that’s how you do it.
0:05:33 And if you do it that way,
0:05:36 there’s no real way for Google to know
0:05:38 as long as the content is high quality.
0:05:39 Like that’s what Google cares about.
0:05:42 Google cares about high quality content.
0:05:44 And if you can create high quality content
0:05:51 and add some human vibes to it, you’ll still rank.
0:05:53 And I’ll just give you a few examples
0:05:56 of what does it mean to make it more human?
0:06:00 Well, one is Google knows if,
0:06:02 like if you put this article has been written
0:06:06 by Nathan Lanz or Matt Wolf,
0:06:09 and well, Google will know that this is like a human
0:06:10 behind it, they’ll go and search.
0:06:11 What’s the bio of this person?
0:06:13 Is this a reputable person?
0:06:17 So what a lot of people are doing in the AI SEO space
0:06:19 is just creating content.
0:06:23 It just looks like chat GVTs created it.
0:06:26 You know how, and I always notice it
0:06:28 ’cause it’s always like the first word
0:06:31 of every sentence is capitalized.
0:06:34 And there’s like colons for days.
0:06:37 It’s like no regular person
0:06:39 would have that many amount of colons.
0:06:42 It’s funny, I’m actually doing a search right now
0:06:45 and I searched up the word as like as an AI language model.
0:06:49 And like, if you just do a search for that term,
0:06:52 it’s like insane how many results actually show up
0:06:55 because so many people are so lazy
0:06:57 that they literally leave as an AI language model
0:06:59 in some of their content.
0:07:00 Totally.
0:07:02 And the other thing that I’m really into
0:07:06 from an AI SEO 2.0 perspective
0:07:09 is creating things like tools and calculators
0:07:11 and embedding it into the content.
0:07:15 So those are things that maybe you’re creating
0:07:20 a piece of content around mortgages, let’s say,
0:07:22 and you embed a mortgage calculator
0:07:24 into that piece of content.
0:07:26 These are things that helped you rank.
0:07:29 And going back to AI, it’s like,
0:07:34 well, you can just use Clode 3.5 Sonnet
0:07:37 to code you up some of these tools and calculators,
0:07:39 start embedding it into the content
0:07:41 and then just start really just outranking
0:07:44 your typical SEO 1.0 people.
0:07:48 So I think that like why have I been interested in this
0:07:50 in the last year and a half and two years
0:07:53 is just because I think that if it becomes
0:07:57 so much easier to figure out which keywords to rank for
0:08:00 and what the content needs to look like
0:08:03 in order for it to be seen by Google,
0:08:05 when you create these assets,
0:08:07 these are assets that pay forever.
0:08:11 It’s not like paid ads where if you spent $100,000
0:08:15 on your paid ads this month, chances are next month,
0:08:17 nothing’s gonna come from that.
0:08:21 But when you invest in SEO, generally,
0:08:24 it just is a compounding machine
0:08:27 and it’ll increase 10, 20% a month.
0:08:29 And I will say the downside of SEO
0:08:32 is you don’t see results right away.
0:08:35 So if you’re an instant gratification person,
0:08:39 like you might not love it, but once it starts compounding,
0:08:40 it becomes a beautiful thing
0:08:42 and then you just have an asset that’s paying you.
0:08:45 – So kind of, I guess kind of the idea would be like,
0:08:47 if you’re just going out there and using AI
0:08:50 to create content for the purpose of SEO,
0:08:51 a lot of other companies are gonna kind of do
0:08:54 the same thing, but you can actually differentiate
0:08:55 by going and creating tools
0:08:58 that will give people a reason to click to your site, right?
0:09:01 I guess if you’re creating a whole bunch of content,
0:09:03 a lot of people are just gonna use Perplexity
0:09:07 or the new Google AI to search
0:09:08 and the response is just gonna be right there
0:09:10 on the homepage without needing to click in.
0:09:13 But if they have a tool that they can use,
0:09:15 they actually have a reason to click into your site
0:09:18 other than just like a sort of summarized version
0:09:19 that AI gives you.
0:09:20 Is that kind of what you’re getting at with it?
0:09:22 – Yeah, yeah, exactly.
0:09:24 I think like the new content has evolved.
0:09:27 It’s no longer just like static words
0:09:29 and stuff like that, right?
0:09:30 So you have to think about, okay,
0:09:33 what is the future of content
0:09:36 and how do I create the most high quality piece of content?
0:09:40 And it just so happens that when you create tools
0:09:42 and calculators and things like that,
0:09:44 people end up staying on the website longer.
0:09:47 And that’s a signal to Google that,
0:09:50 hey, this is something that’s really valuable,
0:09:52 therefore prioritize it.
0:09:57 And other things like video also keeps time on site.
0:10:01 Therefore, that’s good for Google.
0:10:02 And then you might outrank.
0:10:07 And these are all things that AI could help you create.
0:10:10 Where I mean, you could have created these things.
0:10:11 Someone listening to this is like,
0:10:14 but I could have created this about eight years ago.
0:10:17 Yes, but it would have been really expensive.
0:10:19 Like you would have had to hire a dev team
0:10:23 to create all these tools on a monthly basis.
0:10:25 You’re spending $10,000 to $20,000 a month.
0:10:28 You would have had to hire an SEO firm
0:10:31 just for thinking about what are the SEO keywords
0:10:33 I wanna go after.
0:10:36 And then you’d have to hire a bunch of human beings
0:10:39 to create content that would cost a lot of money.
0:10:40 So all these things would add up.
0:10:43 What I’m saying now is it’s now a fraction of the price
0:10:48 to do it and it allows you to compete with the big boys
0:10:51 which is pretty cool.
0:10:52 I do wonder though.
0:10:55 It feels like now that all that’s so easy to make
0:10:58 that Google will have to lean more and more into authority.
0:11:01 Like looking for sites that are like the CNNs
0:11:03 or the reddits or whatever, like Quora.
0:11:06 The recent Google update,
0:11:07 apparently they changed the algorithm
0:11:09 where they’re highly focused on Reddit now.
0:11:11 And people are now trolling Reddit
0:11:12 and putting content on there.
0:11:13 ‘Cause Google doesn’t know what to do.
0:11:16 They’re like, well, trust Reddit, I guess, or Quora.
0:11:18 There’s tons of real people there.
0:11:20 And then now people are kind of gaming that.
0:11:21 And then they probably have another issue too,
0:11:25 like in terms of programmatic SEO.
0:11:28 Have you seen like the perplexity pages that came out?
0:11:29 – Yeah.
0:11:32 – Yeah, so apparently perplexity’s SEO traffic
0:11:34 is like booming right now.
0:11:36 And it’s all from like literally
0:11:39 the AI is creating all the pages, right?
0:11:40 And so now they have a situation where like,
0:11:42 okay, perplexity’s a new authority,
0:11:44 but they’re using AI to create all the contents.
0:11:47 Yeah, maybe Google’s not dead,
0:11:48 but maybe they end up serving up
0:11:50 most of perplexity’s content.
0:11:53 – Yeah, my guess is,
0:11:55 and this gets into like antitrust territory,
0:11:58 but my guess is they are going to
0:12:02 start suppressing perplexity pages.
0:12:03 I think that if you’re out there
0:12:06 and you’re listening to this and you’re like, okay,
0:12:08 you know, I would love thousands of visits
0:12:11 coming to my website every single month.
0:12:12 The thing to think about is,
0:12:15 how do you create the most high quality content
0:12:17 in the most efficient way?
0:12:20 And if you’re able to create high quality content,
0:12:24 the CNNs of the world and the high,
0:12:28 the trusted, great backlinks that you want
0:12:31 are going to come after you just organically
0:12:34 because you’ve created high quality content.
0:12:38 And then of course, you have to do some things,
0:12:40 like for example, not many people know this,
0:12:43 but if you submit your startup to Product Hunt,
0:12:46 Product Hunt is actually a high trust worthy site.
0:12:47 Like the number one reason you should submit
0:12:49 to Product Hunt is for SEO.
0:12:51 And so there’s a bunch of things like that
0:12:53 that you can do that, you know,
0:12:56 I’ve learned over the years that are worth doing too.
0:12:59 (upbeat music)
0:13:00 – We’ll be right back,
0:13:02 but first I want to tell you about another great podcast
0:13:03 you’re going to want to listen to.
0:13:07 It’s called Science of Scaling, hosted by Mark Roberers,
0:13:10 and it’s brought to you by the HubSpot Podcast Network,
0:13:13 the audio destination for business professionals.
0:13:15 Each week hosts Mark Roberers,
0:13:18 founding chief revenue officer at HubSpot,
0:13:20 senior lecturer at Harvard Business School,
0:13:22 and co-founder of Stage 2 Capital,
0:13:25 sits down with the most successful sales leaders in tech
0:13:28 to learn the secrets, strategies, and tactics
0:13:30 to scaling your company’s growth.
0:13:32 He recently did a great episode called,
0:13:35 “How Do You Solve for a Siloed Marketing in Sales?”
0:13:37 And I personally learned a lot from it.
0:13:39 You’re going to want to check out the podcast,
0:13:43 listen to Science of Scaling wherever you get your podcasts.
0:13:46 (upbeat music)
0:13:49 Now, what do you think that like, in the future,
0:13:51 people are even going to like click into the links anymore,
0:13:53 though, because I feel like, obviously,
0:13:55 even Google’s going in the direction now
0:13:56 where if you ask it questions,
0:13:57 it just puts the answer on the homepage,
0:14:00 or people can go to chat GPT or Claude
0:14:03 and ask their question and, you know, get an answer.
0:14:06 And I don’t know if I’m just like, I’m in an AI like tech bubble
0:14:08 where that’s just sort of become my habit,
0:14:11 or if the rest of the world is starting to do that as well.
0:14:12 I don’t, I don’t totally know,
0:14:15 but I feel like I find myself clicking
0:14:16 into links less and less
0:14:19 because the response is just right there for me.
0:14:22 So like, do you think the incentive to even do SEO
0:14:24 is going to go away over time?
0:14:27 – I think that the searches that are going to do really well
0:14:29 on like the perplexities of the world,
0:14:34 AI search are going to be very information-based searches.
0:14:38 So it’s like, what is the weather today in Miami?
0:14:41 You know, Hurricane Barrel is coming.
0:14:44 Is it going to land in Houston?
0:14:46 And I think it’s going to do an incredible job
0:14:49 at like going through the internet
0:14:52 and just boom, here’s your answer.
0:14:54 But there’s a huge sub,
0:14:56 there’s a huge segment of searches
0:15:00 that aren’t as simple as,
0:15:02 let me just pull in the information
0:15:06 that almost reminds me a little bit about,
0:15:07 you know, going old school.
0:15:09 It’s like, you walk into a library, you know,
0:15:11 it’s not like there’s just one book in the library.
0:15:13 There’s like different books in the library
0:15:16 or, you know, you turn on Netflix.
0:15:19 It’s not like you go into one channel
0:15:20 and then it, you know,
0:15:23 just knows what you’re looking for automatically.
0:15:26 Sometimes it’s nice to have some sort of interface
0:15:29 that allows you to jump off to different places.
0:15:34 So I think that the future of search
0:15:37 is going to be a lot more like perplexity
0:15:39 than it is currently Google.
0:15:43 But I think the idea of I’m going to jump around
0:15:44 to different places,
0:15:47 a la library, a la Netflix,
0:15:51 a la, you know, linear TV isn’t going anywhere.
0:15:53 – Okay, so is that the game plan that people do now
0:15:55 is like you throw out tons of AI content
0:15:59 and then you go back with humans to improve
0:16:00 the content that’s actually ranking
0:16:01 and actually make it good?
0:16:04 Is that the kind of the playbook that people are doing?
0:16:06 – I mean, that’s a playbook
0:16:08 that a lot of people who are doing well
0:16:09 in the space are doing.
0:16:11 So I think the first thing I always think about is like,
0:16:14 okay, who’s my ideal customer?
0:16:17 And then based on that, it’s okay.
0:16:18 So I want to track these people
0:16:21 and then what do I want these people to do next?
0:16:24 And for you, it’s probably join the newsletter, I’m guessing.
0:16:26 ‘Cause that way you have their,
0:16:27 okay, once around the newsletter,
0:16:30 it’s like then you can send them the podcast, right?
0:16:34 And then they, it’s all about how do you take a stranger
0:16:36 and make them a raving fan?
0:16:38 Like that’s the game of business in a nutshell.
0:16:40 Everything else doesn’t matter, right?
0:16:43 If you can do those things, then,
0:16:47 and you can do it profitably, you win at business.
0:16:49 Like that’s the most barbarian way
0:16:51 of saying what business is, is that.
0:16:56 So I feel like if Warren Buffett was an internet marketer,
0:16:59 he would be like all about SEO 2.0 right now.
0:17:02 – Well, I also feel like with AI,
0:17:05 you have this attention to almost take this reverse route
0:17:08 because you can create so much content at scale with AI, right?
0:17:11 So you could always go and create
0:17:13 a ton of different content pieces
0:17:17 on a bunch of different AI related topics.
0:17:21 It all would make a good sort of lead into your opt-in,
0:17:24 figure out which content drives the most leads in
0:17:26 and then sort of double down on that content.
0:17:28 I guess that would be another approach as well,
0:17:30 sort of cast a wide net,
0:17:35 figure out which piece of the net catches the most fish
0:17:37 and then hone in on that, you know?
0:17:38 – I think that’s right.
0:17:41 – Well, the other thing I wanna talk about with SEO 2.0
0:17:44 is the fact that large language models in AI
0:17:47 could make SEO a lot more personalized, right?
0:17:49 I think you mentioned that in your SEO 2.0 tweet
0:17:54 is that we’re gonna see SEO and the search results
0:17:59 get a lot more personalized and tailored to the individuals.
0:18:03 So how do you see people sort of leaning into that with SEO?
0:18:06 – So yeah, that’s definitely where the world is going.
0:18:09 I mean, it has been going there even pre-AI,
0:18:13 like in, you know, Google started collecting data around,
0:18:15 okay, we know your location is in Miami.
0:18:16 Therefore, if you write weather,
0:18:18 it’s gonna bring up weather Miami.
0:18:21 It’s not gonna bring up weather Kyoto.
0:18:26 So what AI has done is it has just made that,
0:18:27 we just put that on steroids.
0:18:32 So what the next evolution of that is agents
0:18:34 basically coming to me with,
0:18:38 hey, you need to know this piece of data for this reason.
0:18:41 So the way search works today is it’s very,
0:18:45 what I would call lean forward.
0:18:48 You have to lean forward to get data.
0:18:51 And I think probably where we’re gonna go
0:18:53 in the next few years,
0:18:55 it’s gonna be way more lean back.
0:19:00 So instead of me searching for what’s a great,
0:19:04 you know, sushi restaurant in Miami Beach,
0:19:08 it’s going to know that it’s 6 p.m. my time.
0:19:10 I haven’t made any dinner reservations.
0:19:12 I live in Miami Beach.
0:19:16 I, it knows I love sushi
0:19:19 and I usually have sushi on Tuesday nights
0:19:21 and it hasn’t done that.
0:19:23 And I haven’t, you know,
0:19:24 it’s just like weird and knows me so well.
0:19:26 And it’s like, hey, you know, there’s this news,
0:19:28 you know, sushi restaurants are just open.
0:19:30 I think you’re gonna like it.
0:19:33 Say yes if you want me to make a reservation for two.
0:19:37 Oh, by the way, your friend, Jamie’s in town,
0:19:39 do you want me to add them to the calendar invite?
0:19:42 And that’s like literally the direction of where we’re going.
0:19:43 – Yeah, I could see that.
0:19:46 But I also think that, you know, that description,
0:19:47 I think to some of us,
0:19:50 like the early adopter techie types,
0:19:51 that sounds really exciting.
0:19:53 But then to a lot of people,
0:19:56 that also sounds like more of a dystopian future
0:19:58 where all these companies have so much data
0:20:01 and, you know, information on us.
0:20:03 How do you see these companies sort of balancing
0:20:05 the data privacy concerns,
0:20:07 trying to feed people exactly what they want
0:20:09 when they want it?
0:20:13 – The reality around data and privacy is
0:20:17 people are okay giving away their data and privacy
0:20:20 if it makes their lives way more convenient.
0:20:23 So if the product ends up working
0:20:25 and it makes their lives way more convenient,
0:20:28 then they’re clicking the terms of service
0:20:29 and they’re not reading and scrolling
0:20:30 with the terms of services
0:20:33 and they’re just going and making their lives convenient.
0:20:34 Why?
0:20:39 Because life for everyone is very difficult,
0:20:42 stressful, overwhelming,
0:20:46 no matter how wealthy you are from the top to the bottom.
0:20:51 And I think that if I just believe actually
0:20:55 that a lot of these products are going in the direction
0:20:57 that they’re going to make the lives way more convenient.
0:21:00 So of course they’ll be, you know, let’s put it this way,
0:21:04 Google is way bigger than DuckDuckGo, you know?
0:21:05 – Right.
0:21:09 – It doesn’t mean DuckDuckGo is in a great business
0:21:11 and a great product and there’s a subset of people
0:21:14 who care about privacy first,
0:21:16 but there’s a greater amount of people
0:21:18 that are just like, I’ve got a baby,
0:21:20 I’m holding a baby on this hand,
0:21:22 my boss is calling me on the other hand,
0:21:24 I need to Google search this
0:21:27 and I’m not going to DuckDuckGo to do it.
0:21:29 – Yeah, I always feel like I care about privacy
0:21:31 but as soon as you give me some kind of convenience
0:21:34 that makes my life easier, I’m like, yeah, you know.
0:21:35 – Yeah.
0:21:36 – Give it to me.
0:21:38 – Yeah, and I think, you know,
0:21:39 where the world is probably going to go
0:21:44 and AI plays a role into this too is AI is going to scrape
0:21:46 a lot of data that exists on the internet.
0:21:48 So it’s going to know probably my address
0:21:50 that I live in Miami Beach.
0:21:54 And maybe what happens is it finds out
0:21:58 that a friend of mine wants to send me an invitation
0:22:01 via mail to a wedding, let’s say.
0:22:02 And it’s just going to be like,
0:22:05 hey, do you live at 123 Main Street Miami Beach?
0:22:06 Yes or no.
0:22:10 And so I think the future of a lot of these products
0:22:13 is what I call contextual onboarding
0:22:15 versus upfront onboarding.
0:22:18 Upfront onboarding is when you ask for a million things
0:22:22 upfront, your phone number, your, you know, where you live,
0:22:25 all these things to, and then you say yes to, you know,
0:22:26 each of those things.
0:22:30 I think that with respect to search and other products too,
0:22:33 but let’s just talk about search for a second.
0:22:36 As it gets to know you, it’s convenience.
0:22:37 So it’s going to be like, hey,
0:22:39 do you want us to do this thing for you?
0:22:41 All you have to do is press yes.
0:22:45 And we have this data because it’s scraped.
0:22:46 – Yeah, I mean, at the end of the day,
0:22:48 all of the companies could get access to the data
0:22:49 if they wanted to.
0:22:53 – I think that whether you believe in SEO 2.0 or not,
0:22:57 play with the tools, like go and create AI content,
0:23:01 human layer it, add a calculator to it, add a tool,
0:23:03 like just go and like try things out
0:23:06 and just see how you feel with it.
0:23:10 And through that, you’re going to learn in your niche,
0:23:13 oh my God, you know, this type of calculator,
0:23:15 this kind of tool works.
0:23:18 And to me, like I invest in things
0:23:21 when I can see like a crazy amount of upside return.
0:23:24 And this is something to me that’s like not a lot of,
0:23:26 like it’s not going to take you that long to learn
0:23:29 this sort of stuff and to test this,
0:23:33 but the return on investment, if it works is so large.
0:23:37 – So I want to talk about some of your spicy takes.
0:23:40 You had another tweet where you gave like 25
0:23:42 different spicy takes and Nathan and I actually talked
0:23:44 about it and we pretty much agree with all
0:23:46 of your spicy takes, but I thought there was some fun ones
0:23:49 that we can dive into because they’re obviously
0:23:51 more nuanced than, you know, the single sentence
0:23:53 for each take that you put.
0:23:55 – By the way, for context, whenever I do
0:24:00 those spicy takes things, I like drink a massive
0:24:05 like cold brew or something or like an energy drink
0:24:09 and just like bang them out in like 10 minutes,
0:24:12 press tweet and then just like go for a walk
0:24:14 and then see what happens in like 30 minutes.
0:24:16 – Can’t see how many people disagree
0:24:18 or agree with what you just said.
0:24:18 – Yeah.
0:24:19 – That’s awesome.
0:24:21 Well, you had one of the spicy takes
0:24:23 and I actually commented on your post about it was,
0:24:26 you mentioned like tools like Otter that are going to be
0:24:29 like a note-taking assistant are probably going to flop.
0:24:31 And I’m curious why you think that.
0:24:34 And the reason I ask is because I actually use Otter,
0:24:35 I actually agree that it’s going to flop,
0:24:36 but I think it’s for a different reason
0:24:38 than you mentioned in your tweet.
0:24:40 So I’m curious why you think so.
0:24:43 – So, and our team, by the way, we have like 120 people
0:24:46 on our team and like we use Otter in all these meetings
0:24:48 and whenever I’m in there, I’m always like, it’s me,
0:24:51 it’s like, I like stare down the Otter, you know,
0:24:56 I’m kind of like, you don’t like me and I don’t like you.
0:24:59 And I think a lot of people have that feeling
0:25:02 with respect to an AI note-taking app.
0:25:06 It gives the convenience to the person
0:25:08 who wants to write notes, of course.
0:25:11 So you can be like more present in the meeting.
0:25:14 Like that’s the, that’s who people who like tools like that,
0:25:15 that’s what they say to me.
0:25:17 They’re like, yeah, like I just want to be present
0:25:20 in the meeting, therefore this is going to help me,
0:25:21 you know, be present.
0:25:25 And it’s like, homie, can’t you just do two things at once?
0:25:28 Like, you know, I remember being in the third grade
0:25:32 taking notes, you know, in a classroom, you know,
0:25:35 I think what people don’t like about AI note-taking tools,
0:25:38 and this, I do believe this is 90 plus percent of people,
0:25:43 is they can’t be free.
0:25:45 They can’t, you know, maybe they want to swear
0:25:47 and they can’t, they don’t want to swear.
0:25:49 And I think it hurts more productivity
0:25:51 than it does help productivity.
0:25:53 That’s my spicy take with it.
0:25:56 And I also just think that, yeah,
0:25:58 the only person that enjoys the Otter
0:26:02 is the person who is supposed to be taking notes.
0:26:05 And I also think, yeah, I’m just, I don’t,
0:26:09 I think go take notes, man, just go take some notes.
0:26:10 – I mean, at the end of the day too,
0:26:11 if you’re sitting there taking notes,
0:26:13 that’s sort of turning it into a multimodality
0:26:15 sort of engagement, right?
0:26:18 You’re listening and you’re also writing,
0:26:19 and because you’re doing both,
0:26:21 you’re going to help lock a lot of those concepts in.
0:26:22 – My take is there’s going to be a backlash
0:26:24 on this sort of stuff.
0:26:27 I feel like people now are kind of just like
0:26:31 politely smiling and they’re like, that’s cute.
0:26:34 But I think in a year or two,
0:26:37 people are going to be like, hey, hey, do you mind just,
0:26:39 I mean, you’re starting to see it a little bit,
0:26:41 but you know, I’m starting to see like,
0:26:42 hey, do you mind just like removing that recorder
0:26:44 from that meeting?
0:26:48 I think you’re going to see instances of that, 50X.
0:26:53 – I’ll tell you, so here’s my sort of killer use
0:26:55 that I love using Otter for.
0:26:57 I actually go to like a lot of conferences
0:26:59 to make content around them for my YouTube channel.
0:27:02 So, you know, I was at Cisco Live
0:27:03 and I was at Augmented World Expo
0:27:07 and Google I/O, Microsoft Build, all these events.
0:27:10 And a lot of times I’m sitting in these conferences
0:27:14 while they’re speaking and it’s hard to take
0:27:15 all the notes on everything they’re talking about.
0:27:17 So I usually open up Otter on my phone,
0:27:20 just set it next to me on my chair
0:27:23 and let Otter actually record the entire, you know,
0:27:26 keynote or panel or whatever I’m sitting and listening to.
0:27:29 And then I’ll take like the summary and the transcript
0:27:32 and I’ll pull it over into a clod or a chat GPT
0:27:35 and say, hey, I need to make a video about this.
0:27:37 What are some of the talking points from this presentation
0:27:39 that I should bring up in this video?
0:27:41 And it will actually help me sort of like
0:27:43 outline rough draft a presentation
0:27:45 about the keynote that I just saw.
0:27:47 And I found that to be really helpful.
0:27:49 Saying that, I also agree,
0:27:51 it’s probably not going to be around for a long time
0:27:54 and it’s going to flop, but I think the reason is more
0:27:56 because I think it’s just going to be a feature
0:27:58 instead of an actual app in a lot of these devices.
0:28:02 I think we’re going to see Google just roll it into Android
0:28:03 as part of the operating system.
0:28:05 I think we’re going to see Apple intelligence
0:28:07 just roll that kind of feature in.
0:28:10 And I can’t see companies like Otter existing
0:28:13 when it’s just like a native built-in feature
0:28:15 of these devices.
0:28:19 – So, well, where I’ll agree with you is in the,
0:28:21 what I’ll call like broadcast setting
0:28:23 if like you’re in a conference
0:28:26 and there’s tons of people around you
0:28:29 in groups of larger than 10,
0:28:31 something like Otter super important.
0:28:34 Like I can see it being super valuable, super important.
0:28:38 But within the context of I’m a company
0:28:39 and I’ve got all these meetings
0:28:41 and a lot of the meetings are just
0:28:46 hanging out or talking about one issue.
0:28:49 And they’re smaller.
0:28:52 Like maybe the average meeting is four people.
0:28:57 That’s where I think Otter is going to really flop.
0:28:58 – Yeah, yeah.
0:28:59 I can see that take too.
0:29:02 I mean, I don’t use it on like Zoom calls and things like that.
0:29:04 So it’s not actually one of the use cases
0:29:05 I’ve actually used it for.
0:29:08 So I can definitely see that take.
0:29:10 – So you said AI is taking jobs
0:29:12 not just making people more productive.
0:29:13 Companies want profits.
0:29:16 The idea that AI frees you to focus on your passion
0:29:18 is pure fantasy for most workers.
0:29:19 You know, I kind of agree with this.
0:29:21 I feel it’s a situation where everyone in AI right now
0:29:24 they can’t say that though.
0:29:25 There’s an odd thing there, right?
0:29:26 Where you just, you feel like you can’t say
0:29:28 the truth about the thing.
0:29:30 ‘Cause it leads to like, okay, well, you know,
0:29:31 does this change capitalism?
0:29:32 Like I’m a capitalist.
0:29:34 Like does capitalism make sense in the future?
0:29:37 Does our current government and economy
0:29:40 make sense in the future where, yeah,
0:29:42 you can use AI for most things
0:29:43 and maybe everyone doesn’t have to work the same way.
0:29:45 Is that kind of like the gist of,
0:29:47 is that how you see it as well or?
0:29:50 – Sometimes I’m scrolling Twitter
0:29:53 and I’m seeing like these tweets where it’s like,
0:29:57 AI is gonna make, you know, there’s gonna be more jobs.
0:29:58 You know, it’s making more jobs.
0:30:02 And I’m like, what are you smoking then?
0:30:04 Like honestly, I feel like I’m living
0:30:05 in a completely different planet.
0:30:10 I’m like, okay, there’s this new paradigm shift
0:30:13 that has basically made it 100 times easier
0:30:18 to do basically anything from creative to software,
0:30:23 to marketing, to all these different things.
0:30:24 So that’s happening.
0:30:26 Everything’s become easier.
0:30:30 At the same time, companies are looking
0:30:34 to become more profitable and especially larger companies,
0:30:36 especially VC-backed companies
0:30:38 who are trying to exit or get acquired
0:30:41 and publicly traded companies
0:30:44 whose only job is to increase shareholder value.
0:30:47 Once they get comfortable to the point
0:30:48 where people are using these tools
0:30:50 and they’re becoming way more productive,
0:30:51 they’re not gonna be like,
0:30:55 oh, hey, let’s go hire another thousand people.
0:30:56 So yeah, I just think that people
0:30:58 are gonna be way more productive.
0:31:00 And if people are more productive,
0:31:04 there’s going to be less of those types of jobs.
0:31:08 – So what do you think, just like very theoretical, right?
0:31:10 What do you think the job market looks like
0:31:12 in several years from now?
0:31:14 – There’s a lot less engineers.
0:31:16 There’s a lot less designers.
0:31:19 I think there’s a lot less marketers.
0:31:23 I think the top 1% ends up getting paid 10 times.
0:31:27 I think the middle gets stripped out.
0:31:28 I think that’s what happens.
0:31:32 I think the middle of these white collar jobs
0:31:34 becomes stripped out.
0:31:37 You’re either junior and you’re like very low paid
0:31:41 or you’re senior and you’re very high paid.
0:31:44 You know, I wish I wasn’t saying this.
0:31:45 I agree with you.
0:31:46 Yeah, I think we both agree.
0:31:47 – It’s hard to say, right?
0:31:48 Cause we don’t know the answer.
0:31:50 Like we don’t know the answer for those people.
0:31:53 Like what do they do and how do they earn a living?
0:31:56 And yeah, sure the AI tools will make their life better,
0:32:00 but they also still need food on the table, right?
0:32:02 – Right, exactly.
0:32:05 – So do you think this, I mean, obviously,
0:32:06 none of us know where it’s all headed,
0:32:09 but do you think this leads to like a UBI scenario?
0:32:12 Do you think like, where do your predictions lie
0:32:13 if you had to make predictions?
0:32:15 – I think it’s important to just share your opinion
0:32:17 and it’s okay if your opinion changes over time
0:32:18 in real time.
0:32:20 And as you learn about a particular topic,
0:32:22 and I think that if more people do that,
0:32:24 like the better society is,
0:32:26 as long as you can be open-minded about, okay,
0:32:28 the data has changed.
0:32:31 Therefore my stance on something has changed.
0:32:33 As far as like do, you know,
0:32:37 in a situation where millions of people are out of work,
0:32:39 is there some sort of UBI?
0:32:43 I mean, I don’t wanna speak for Sam Altman,
0:32:46 but like, why is he creating Worldcoin
0:32:49 if he isn’t thinking about that?
0:32:53 That UBI is gonna be a more important part of,
0:32:56 you know, as a mechanism to pay certain people.
0:32:58 You know, what do I think?
0:33:00 I stop there.
0:33:03 Like, I don’t know how, you know, I’m not a politician.
0:33:05 I’m not an economist.
0:33:07 I’m a technologist that knows
0:33:08 where the technology is going.
0:33:09 Where it goes from there?
0:33:11 Like, I’m not, it’s above my pay grade.
0:33:14 Well, yeah, you have to be like a philosopher,
0:33:15 psychologist, everything, right?
0:33:18 Like, it’s like, how do humans, you know,
0:33:20 interact or how humans live if they don’t have work?
0:33:21 And how does that work?
0:33:23 And how do you feel motivated?
0:33:26 And especially if other people are getting 10x, right?
0:33:29 And they’re living this amazing new life,
0:33:30 and then you’re kind of like living
0:33:33 in your virtual reality world or whatever.
0:33:36 Yeah, I mean, I don’t think there’s gonna be UBI.
0:33:41 I think people will find different types of work
0:33:45 and be productive to society.
0:33:47 What that work will look like,
0:33:48 I think will be different, very different
0:33:50 than what it is today.
0:33:54 I think that the wealth disparity, unfortunately, will grow.
0:33:56 And that’s why I think it’s important for everyone
0:33:58 to just stay ahead.
0:34:00 That’s why this podcast is great, right?
0:34:03 Like, learn, get your hands dirty,
0:34:08 and you’ll never have to worry if that’s the case
0:34:13 because you’ll always be one step ahead of other people.
0:34:16 People don’t put in the work.
0:34:18 People just don’t put in the work, right?
0:34:19 Even when the work gets easier.
0:34:24 They much rather watch episode 124 of Love Island
0:34:25 than episode 21.
0:34:27 Is that tonight?
0:34:28 The next wave.
0:34:29 I think that’s true.
0:34:32 I think it, you know, it’s people wanna,
0:34:33 you know, we talked about this before.
0:34:34 It’s like, life is hard, right?
0:34:36 And it’s like, if you wanna numb,
0:34:38 you wanna numb your brain sometimes,
0:34:39 but sometimes like,
0:34:42 sometimes you gotta keep your brain stimulated
0:34:44 in order for you to stay ahead.
0:34:48 – Yeah, I mean, along that same lines,
0:34:49 I’m gonna actually jump ahead
0:34:51 to one of your other hot takes
0:34:53 because it’s very relevant to what you just said.
0:34:56 You said more young millionaires will get minted
0:34:58 over the next 10 years than any point in the history.
0:35:00 But watch YouTube, listen to podcasts,
0:35:03 and learn to take ideas into reality thanks to AI.
0:35:06 It’ll make many people jealous.
0:35:08 So, I mean, there, you just kind of mentioned
0:35:10 there’s gonna be a growing disparity
0:35:12 between I guess, you know, the haves and have nots.
0:35:14 The people with a lot of money
0:35:16 versus people with not a lot of money.
0:35:19 But also, to sort of counteract that point,
0:35:20 it also sounds like you’re saying
0:35:23 there’s no easier time in history to make money.
0:35:25 – Absolutely.
0:35:27 I mean, you’re connected to the internet.
0:35:30 And so you have access to a library
0:35:34 of the smartest people on the planet, basically,
0:35:35 in your pocket.
0:35:40 So, and then when you wanna create something,
0:35:43 now you’ve got, you know, AI tools and no-code tools
0:35:46 that you can actually just go and press some buttons,
0:35:49 not have to have, you know, gone to Stanford
0:35:53 and studied computer science with, you know,
0:35:54 all these well-known people.
0:35:57 And you can just put stuff on the internet
0:35:59 and then literally billions of people
0:36:03 have their credit cards on the internet
0:36:05 and you can sell to anyone.
0:36:08 It used to be that, you know, my dad had a store
0:36:13 and he had a store in this, you know, little area.
0:36:16 And if more people moved to the area,
0:36:18 there’d be more people and if people left,
0:36:23 then less people would buy at the store, you know?
0:36:25 Now, the store is everyone
0:36:29 and you don’t need to go and spend two years
0:36:31 building a building.
0:36:34 You can spend two minutes to build a building.
0:36:38 So I think there’s gonna be a lot of low-quality
0:36:43 and medium-quality startups, content, communities
0:36:45 that are gonna exist.
0:36:48 And it’s just gonna feel like,
0:36:52 it’s just like how in marketing, like if you go,
0:36:55 I saw like a video of like 1919 in New York City
0:36:59 and there’s like no billboards and there’s like no marketing.
0:37:03 And then if you go in 2024, New York, like of course,
0:37:06 there’s like your eyes constantly looking
0:37:08 at a million different marketing, you know?
0:37:11 I think that’s gonna, that’s the world we’re headed in
0:37:12 in terms of the internet.
0:37:14 It’s gonna look a lot less like 1919,
0:37:16 a lot more like 2024 times square.
0:37:20 And if you need, if you wanna stand out, you know,
0:37:23 those are where the young millionaires are gonna mint it.
0:37:27 And I think why are they gonna be young?
0:37:31 It’s because the young people are gonna be native
0:37:33 to the tools to create the stuff.
0:37:36 – Yeah, one thing I feel like, you know,
0:37:38 even though this technology is gonna create a,
0:37:41 maybe a bigger divide in terms of the haves and haves nots,
0:37:43 I do feel like it’s gonna equalize things in a way
0:37:47 where you don’t have to be the smartest person
0:37:49 to use AI and make something cool.
0:37:51 So like a lot of times I feel like it’s gonna matter more
0:37:54 about like how motivated is the person, right?
0:37:55 Yeah, they didn’t go to Stanford,
0:37:56 they didn’t have the highest GPA,
0:37:58 but they can, if they’re motivated,
0:38:00 they’ll still be able to use AI tools
0:38:02 and build a company and build stuff, right?
0:38:05 So I mean, I find that personally inspiring.
0:38:08 And like, you know, if people really wanna make it,
0:38:10 AI is gonna help them make it more than ever before.
0:38:14 – Absolutely, absolutely.
0:38:18 I think if you’re motivated and you’re ambitious,
0:38:19 you’re gonna do well.
0:38:20 You have nothing to worry about.
0:38:22 There’s one last topic I wanted to talk about
0:38:25 before we let you go, and that’s the climate issue.
0:38:27 ‘Cause you mentioned that there’s going to be
0:38:29 a major climate backlash against AI,
0:38:32 it’s gonna become the current thing to talk about.
0:38:34 And Nathan and I, even before this call,
0:38:35 we’re actually talking about that.
0:38:36 We both agree with you on that.
0:38:39 I think that’s gonna be the big narrative.
0:38:42 But when you made that hot take, like,
0:38:43 what was your perspective?
0:38:45 Where were you coming from on it when you wrote it?
0:38:50 – Every major wave of technology, computer,
0:38:55 computers, the internet, social, mobile, crypto,
0:39:01 there’s been a point in time where those technologies
0:39:06 get so big that there becomes a climate backlash.
0:39:10 If you look at the maximums of the,
0:39:13 like if you were to plot out, you know,
0:39:14 number of articles, let’s say,
0:39:18 that have the climate backlash,
0:39:20 every peak has gotten higher.
0:39:25 So we saw this, I think, in 2021 with crypto,
0:39:30 there started to become a huge backlash
0:39:32 that, and people started saying, you know,
0:39:35 if we shut down Bitcoin, like we can power,
0:39:37 like, you know, all of Ukraine
0:39:39 and all the, you know, different places, right?
0:39:43 So, and I think that someone is going to
0:39:47 start telling the story that these technologies
0:39:51 take compute and compute isn’t free.
0:39:56 It requires electricity, requires the actual GPUs,
0:39:58 and that affects the environment.
0:40:03 So for me, it’s not a matter of if,
0:40:04 it’s a matter of when.
0:40:05 – Yeah.
0:40:07 – And, you know, from an opportunity perspective,
0:40:09 like, there’s probably opportunities to create more,
0:40:14 like, climate-friendly, clean, carbon-neutral AI technologies.
0:40:18 So if, you know, we were talking about DuckDuckGo,
0:40:20 DuckDuckGo versus Google,
0:40:23 I think that if, you know, that was the,
0:40:25 you know, riding the privacy-first trend,
0:40:27 I think there’s also a climate-first trend.
0:40:29 So I think that there’s an opportunity to create,
0:40:32 like, what does the green version of perplexity look like?
0:40:34 – Yeah, it feels like, unfortunately, this makes,
0:40:37 you know, AI is going to get political for this reason,
0:40:40 which I hate, you know, because there’s obviously a divide
0:40:43 of, like, okay, are we going to stay only on Earth,
0:40:46 and are we going to, you know, solve climate issues
0:40:49 by kind of being more efficient and having less people
0:40:52 and less waste, or are we going to solve climate problems
0:40:55 by using technology to solve those problems
0:40:58 and going to other planets, right?
0:40:59 And so a lot of the techno-optimists, like, you know,
0:41:03 we believe, like, that obviously AI, you know,
0:41:05 if it takes more energy, that’s okay.
0:41:07 Like, AI is gonna help us solve the problems,
0:41:08 it’s gonna help build the technologies
0:41:09 we don’t even know exist yet
0:41:11 that are gonna solve the problems,
0:41:14 and you can’t get out of it just by having less,
0:41:17 you know, plastic straws or whatever.
0:41:19 But obviously, there’s a huge divide there,
0:41:20 and it gets really political, and so yeah,
0:41:23 I hate that, unfortunately, AI is probably gonna get
0:41:24 pretty political in the next year or so,
0:41:25 and I think it will.
0:41:28 – Yeah, yeah, I’ve been hearing a lot of this stuff
0:41:31 about, like, you know, the climate issues
0:41:33 and how much power all of this uses,
0:41:34 and all of that kind of stuff,
0:41:36 but I’ve also been seeing both sides of the narrative.
0:41:39 I’ve also been seeing a lot of articles and reports
0:41:42 and stuff that say it’s way overblown, too.
0:41:44 I’ve seen things saying that in 2020,
0:41:46 when everybody was in their homes
0:41:50 and, like, everybody started playing video games more often,
0:41:53 that used as much compute as what AI is using right now,
0:41:55 right, and then there’s also the sort of narrative
0:41:57 of, like, NVIDIA and Qualcomm
0:42:00 and all these companies that are making the processing,
0:42:02 their number one goal, anytime they give a keynote,
0:42:05 is to talk about how they’re trying to make it more efficient
0:42:06 and bring down the energy usage
0:42:08 and get more and more powerful compute,
0:42:11 but for less the energy use.
0:42:13 So I do think a lot of these problems will be solved,
0:42:17 but I also think this is gonna be, like, the narrative,
0:42:19 the popular narrative against AI,
0:42:21 but I also think a lot of this,
0:42:24 there’s solutions being worked on right now.
0:42:25 – It is gonna get more efficient,
0:42:27 the graphics charge is gonna get more efficient,
0:42:28 but there’s a reason that, like, Elon Musk
0:42:31 is, like, doubling down on building, you know,
0:42:34 factories and stuff in Texas, right?
0:42:35 Like, aligning himself with, like,
0:42:37 people who are, like, more pro-energy
0:42:40 and, like, have more energy, you know, consumption,
0:42:42 is because that’s probably where things are gonna go.
0:42:44 Like, yeah, in the current state,
0:42:46 yeah, it doesn’t use too much energy,
0:42:48 but when we get to AGI, (laughs)
0:42:50 ASI, yeah, you’re probably talking about
0:42:52 massive amounts of energy consumption,
0:42:55 probably the likes that humanity’s never seen before.
0:42:56 – So this is really interesting.
0:43:00 I asked, I prompted perplexity and I said,
0:43:03 “What’s the ecological impact of a chat GPT prompt?”
0:43:06 I was gonna say perplexity prompt,
0:43:09 ’cause I was like, I want something unbiased, right?
0:43:10 (laughs)
0:43:11 So, it says this.
0:43:15 The ecological impact of a single chat GPT prompt
0:43:19 is relatively small, but it can add up significantly
0:43:21 when considering the massive scale of usage.
0:43:22 Here’s the breakdown.
0:43:25 Each chat GPT query is estimated to produce
0:43:28 approximately 4.32 grams of CO2,
0:43:29 but here’s where it gets really interesting.
0:43:32 I didn’t realize, but, I mean, silly me,
0:43:35 that there’s water consumption with chat, with this.
0:43:38 So, chat GPT’s water usage is particularly noteworthy.
0:43:40 For every 20 to 50 queries,
0:43:43 the system consumes about 500 milliliters of water,
0:43:46 equivalent to a standard bottle of water.
0:43:50 This water is primarily used for cooling the data centers
0:43:53 that power the AII model, so really interesting.
0:43:56 And then, there’s more, and then there’s energy usage.
0:44:00 So, the energy consumption of chat GPT is considerable.
0:44:03 The system runs on an estimated 30,000 GPUs,
0:44:07 which requires significant power to operate.
0:44:10 So, to put this in perspective,
0:44:13 15 queries are equivalent to watching one hour video.
0:44:16 16 queries consume as much energy as boiling one kettle.
0:44:19 139 queries use as much energy
0:44:22 as watching one load of laundry.
0:44:25 And if current growth trends continue by 2027,
0:44:27 chat GPT’s electricity consumption
0:44:30 could rival entire nations like Sweden, Argentina,
0:44:32 and the Netherlands.
0:44:36 – Well, I think the important part of this whole conversation
0:44:39 is that people have conversations about it, right?
0:44:41 Because I think when it comes to AI right now,
0:44:43 the pro AI people want to pretend
0:44:46 like these negatives don’t exist.
0:44:49 And the anti AI people want to act like
0:44:51 solutions can’t exist.
0:44:55 And I think the important thing is that we,
0:44:59 I don’t feel like AI should be this divisive topic.
0:45:02 I feel like it should be a topic
0:45:05 that we’re all talking about and collaborating on
0:45:07 and figuring out solutions together,
0:45:09 both on the pro and anti AI side.
0:45:11 How do we find this middle ground
0:45:12 that’s gonna make everybody happy?
0:45:15 That’s going to solve a lot of these problems.
0:45:16 And I think one of the frustrations
0:45:21 that I’ve had recently just being so deep in the AI space
0:45:22 is it’s all so binary.
0:45:24 It seems like everybody’s either,
0:45:27 I’m totally against AI or I’m totally for AI,
0:45:28 but I believe there’s a spectrum there
0:45:31 and you can land anywhere on that spectrum.
0:45:33 Well, I think that’s just the bigger issue
0:45:36 with human beings, especially now is like,
0:45:39 you have to be either this side or that side
0:45:41 and there’s no in between, there’s no gray,
0:45:43 like gray is dead now.
0:45:48 And that’s just, it’s just being applied to AI.
0:45:54 So let’s leave it on a positive note.
0:45:56 (laughing)
0:45:57 Go create some stuff.
0:46:00 You know what I mean?
0:46:04 You are listening to this, you understand the tools,
0:46:07 you just gotta go build some stuff
0:46:10 and go and create stuff that people are gonna love
0:46:12 that are gonna put smile on their faces.
0:46:14 It’s gonna add value to their lives,
0:46:16 make them happier and healthier
0:46:20 and contribute to making the world a better
0:46:23 and happier place and you can do that now.
0:46:26 Well, on that note, you make a lot of great content.
0:46:29 Everybody needs to go follow Greg on Twitter X.
0:46:30 Great content there.
0:46:32 What’s the name of your podcast again?
0:46:33 – The startup ideas podcast.
0:46:36 – The startup ideas podcast, go check that out.
0:46:37 Well, thanks so much for joining us.
0:46:39 This has been an amazing conversation.
0:46:41 I’m sure it won’t be the last mode we have on this show
0:46:44 and I appreciate you joining us.
0:46:45 – Thanks for having me.
0:46:48 (upbeat music)
0:46:51 (upbeat music)
0:46:53 (upbeat music)
0:46:56 (upbeat music)
0:46:58 (upbeat music)
0:47:01 (upbeat music)
0:47:11 [BLANK_AUDIO]
0:00:04 – If Warren Buffett was an internet marketer,
0:00:08 he would be like all about SEO 2.0 right now.
0:00:11 It’s all about how do you take a stranger
0:00:13 and make them a raving fan?
0:00:16 If you can do those things, you win at business.
0:00:19 (upbeat music)
0:00:24 – When all your marketing team does is put out fires,
0:00:25 they burn out.
0:00:28 But with HubSpot, they can achieve their best results
0:00:29 without the stress.
0:00:32 Tap into HubSpot’s collection of AI tools,
0:00:35 breeze, to pinpoint leads, capture attention,
0:00:38 and access all your data in one place.
0:00:39 Keep your marketers cool
0:00:42 and your campaign results hotter than ever.
0:00:45 Visit hubspot.com/marketers to learn more.
0:00:48 (upbeat music)
0:00:50 – Hey, welcome to the Next Wave podcast.
0:00:51 I’m Matt Wolf.
0:00:53 I’m here with Nathan Lanz.
0:00:56 And today we’ve got an amazing returning guest
0:01:01 in Greg Eisenberg, he’s the host of the Startup Ideas podcast.
0:01:04 He’s got a holding company called Elite Checkout,
0:01:07 which does over $10 million in revenue per year.
0:01:09 He runs the boring marketing company
0:01:12 where he helps other companies do SEO.
0:01:14 And in this episode, he’s gonna map out
0:01:18 an exact blueprint that you can follow for your SEO
0:01:21 in this new AI world that we’re going into.
0:01:24 He breaks down the SEO 2.0 strategy.
0:01:27 We also talk about all sorts of hot takes
0:01:30 that he has on Twitter and dive into
0:01:32 why he believes those hot takes are true.
0:01:34 And you’re really gonna enjoy this episode.
0:01:35 So let’s just go ahead and dig right in
0:01:37 with Greg Eisenberg.
0:01:38 Hey, Greg, welcome back to the show.
0:01:40 It’s great to have you on.
0:01:42 – I’m ready, I’m ready for it.
0:01:45 – Awesome, well, let’s just get straight into it.
0:01:47 So you have a couple tweets out there right now
0:01:50 that Nathan and I both had some conversations around
0:01:52 and got us talking and we thought,
0:01:54 these are some fun things to bring Greg on
0:01:56 and talk about on the podcast.
0:02:01 And the first one being the concept of SEO 2.0.
0:02:03 So maybe the best place to start is just like,
0:02:05 what’s your definition of SEO 2.0?
0:02:08 Where do you see SEO going in the world of AI?
0:02:12 – Yeah, I mean, a lot of people call it programmatic SEO.
0:02:14 I call it SEO 2.0.
0:02:19 SEO 1.0 to me is basically, I mean, it’s old school SEO.
0:02:21 It’s basically you go and create content pages.
0:02:26 You get humans to do it, top things to do in New York City.
0:02:28 You go and you know all the different restaurants
0:02:30 and bars and you review it.
0:02:34 And then you just sort of hope that, I guess,
0:02:37 you get backlinks from other websites
0:02:40 and then that’s gonna generate traffic.
0:02:44 And that worked pretty well for a long period of time.
0:02:49 And there’s quite literally dozens of billion dollar companies
0:02:52 that rode the SEO 1.0 wave.
0:02:55 Content companies, social networks like Reddit,
0:02:58 these are companies that have great rank.
0:03:00 And then about a year and a half, two years ago,
0:03:02 I started thinking about, hey,
0:03:06 if AI makes it easier to do two things.
0:03:10 One is to get insights and data in a really efficient way.
0:03:14 And two is creating content and creating web pages
0:03:15 in a really efficient way.
0:03:16 Isn’t that me?
0:03:17 What I mean that you’ll be able to just create
0:03:22 like millions of pages and rank for them?
0:03:25 And that’s really what I mean by SEO 2.0.
0:03:27 I’ll tell you why I think it’s exciting to me
0:03:29 and maybe to people listening,
0:03:34 is the same way that there’s dozens of billion dollar companies
0:03:36 and thousands of million dollar companies
0:03:40 that have rode the wave of SEO 1.0,
0:03:43 I believe the same thing is gonna happen for 2.0.
0:03:45 The people that are gonna be able
0:03:50 to efficiently create these web pages, get ranked,
0:03:51 are gonna do really well.
0:03:53 And I know that there’s someone listening,
0:03:55 especially ’cause this is an AI podcast
0:03:58 that’s thinking themselves, well, okay,
0:04:02 but it doesn’t matter because perplexity is the new Google.
0:04:04 And if perplexity is the new Google,
0:04:08 then you know, you ranking doesn’t really matter
0:04:11 on this search engine that no one’s gonna go to.
0:04:14 And the reality of the situation is,
0:04:16 Google is not dead for,
0:04:18 I know a lot of people like to think that Google is dead,
0:04:20 but Google is not dead.
0:04:25 And there’s still huge opportunities to start ranking.
0:04:27 – You know, the first thing that comes to mind
0:04:31 when I think of like an SEO 2.0 concept
0:04:34 and like AI just sort of being able
0:04:36 to sort of mass make content,
0:04:38 is that I feel like the internet might just get flooded
0:04:40 with a lot of junk content.
0:04:43 I mean, how much of SEO 1.0,
0:04:44 especially in the early days,
0:04:47 was people like writing articles, spinning the articles,
0:04:48 and then it would just like change
0:04:50 a certain percentage of words
0:04:52 and they were doing it for the algorithm
0:04:54 and they weren’t really doing it for the reader.
0:04:55 That’s the first thing that comes to mind,
0:04:59 is like, doesn’t it seem like that might just lead
0:05:01 to a whole bunch of clutter and junk all over the internet
0:05:03 that nobody actually wants to read?
0:05:05 – Yeah, so the reality is,
0:05:09 if you go and create AI content today,
0:05:12 you’re gonna get penalized by Google.
0:05:14 So Google has gotten really good
0:05:19 at basically knowing if something is written by AI.
0:05:20 So what you actually have to do
0:05:23 is you kind of have to like deke out Google.
0:05:27 You have to essentially use humans for 20%
0:05:29 and AI for 80%.
0:05:31 And that’s how you do it.
0:05:33 And if you do it that way,
0:05:36 there’s no real way for Google to know
0:05:38 as long as the content is high quality.
0:05:39 Like that’s what Google cares about.
0:05:42 Google cares about high quality content.
0:05:44 And if you can create high quality content
0:05:51 and add some human vibes to it, you’ll still rank.
0:05:53 And I’ll just give you a few examples
0:05:56 of what does it mean to make it more human?
0:06:00 Well, one is Google knows if,
0:06:02 like if you put this article has been written
0:06:06 by Nathan Lanz or Matt Wolf,
0:06:09 and well, Google will know that this is like a human
0:06:10 behind it, they’ll go and search.
0:06:11 What’s the bio of this person?
0:06:13 Is this a reputable person?
0:06:17 So what a lot of people are doing in the AI SEO space
0:06:19 is just creating content.
0:06:23 It just looks like chat GVTs created it.
0:06:26 You know how, and I always notice it
0:06:28 ’cause it’s always like the first word
0:06:31 of every sentence is capitalized.
0:06:34 And there’s like colons for days.
0:06:37 It’s like no regular person
0:06:39 would have that many amount of colons.
0:06:42 It’s funny, I’m actually doing a search right now
0:06:45 and I searched up the word as like as an AI language model.
0:06:49 And like, if you just do a search for that term,
0:06:52 it’s like insane how many results actually show up
0:06:55 because so many people are so lazy
0:06:57 that they literally leave as an AI language model
0:06:59 in some of their content.
0:07:00 Totally.
0:07:02 And the other thing that I’m really into
0:07:06 from an AI SEO 2.0 perspective
0:07:09 is creating things like tools and calculators
0:07:11 and embedding it into the content.
0:07:15 So those are things that maybe you’re creating
0:07:20 a piece of content around mortgages, let’s say,
0:07:22 and you embed a mortgage calculator
0:07:24 into that piece of content.
0:07:26 These are things that helped you rank.
0:07:29 And going back to AI, it’s like,
0:07:34 well, you can just use Clode 3.5 Sonnet
0:07:37 to code you up some of these tools and calculators,
0:07:39 start embedding it into the content
0:07:41 and then just start really just outranking
0:07:44 your typical SEO 1.0 people.
0:07:48 So I think that like why have I been interested in this
0:07:50 in the last year and a half and two years
0:07:53 is just because I think that if it becomes
0:07:57 so much easier to figure out which keywords to rank for
0:08:00 and what the content needs to look like
0:08:03 in order for it to be seen by Google,
0:08:05 when you create these assets,
0:08:07 these are assets that pay forever.
0:08:11 It’s not like paid ads where if you spent $100,000
0:08:15 on your paid ads this month, chances are next month,
0:08:17 nothing’s gonna come from that.
0:08:21 But when you invest in SEO, generally,
0:08:24 it just is a compounding machine
0:08:27 and it’ll increase 10, 20% a month.
0:08:29 And I will say the downside of SEO
0:08:32 is you don’t see results right away.
0:08:35 So if you’re an instant gratification person,
0:08:39 like you might not love it, but once it starts compounding,
0:08:40 it becomes a beautiful thing
0:08:42 and then you just have an asset that’s paying you.
0:08:45 – So kind of, I guess kind of the idea would be like,
0:08:47 if you’re just going out there and using AI
0:08:50 to create content for the purpose of SEO,
0:08:51 a lot of other companies are gonna kind of do
0:08:54 the same thing, but you can actually differentiate
0:08:55 by going and creating tools
0:08:58 that will give people a reason to click to your site, right?
0:09:01 I guess if you’re creating a whole bunch of content,
0:09:03 a lot of people are just gonna use Perplexity
0:09:07 or the new Google AI to search
0:09:08 and the response is just gonna be right there
0:09:10 on the homepage without needing to click in.
0:09:13 But if they have a tool that they can use,
0:09:15 they actually have a reason to click into your site
0:09:18 other than just like a sort of summarized version
0:09:19 that AI gives you.
0:09:20 Is that kind of what you’re getting at with it?
0:09:22 – Yeah, yeah, exactly.
0:09:24 I think like the new content has evolved.
0:09:27 It’s no longer just like static words
0:09:29 and stuff like that, right?
0:09:30 So you have to think about, okay,
0:09:33 what is the future of content
0:09:36 and how do I create the most high quality piece of content?
0:09:40 And it just so happens that when you create tools
0:09:42 and calculators and things like that,
0:09:44 people end up staying on the website longer.
0:09:47 And that’s a signal to Google that,
0:09:50 hey, this is something that’s really valuable,
0:09:52 therefore prioritize it.
0:09:57 And other things like video also keeps time on site.
0:10:01 Therefore, that’s good for Google.
0:10:02 And then you might outrank.
0:10:07 And these are all things that AI could help you create.
0:10:10 Where I mean, you could have created these things.
0:10:11 Someone listening to this is like,
0:10:14 but I could have created this about eight years ago.
0:10:17 Yes, but it would have been really expensive.
0:10:19 Like you would have had to hire a dev team
0:10:23 to create all these tools on a monthly basis.
0:10:25 You’re spending $10,000 to $20,000 a month.
0:10:28 You would have had to hire an SEO firm
0:10:31 just for thinking about what are the SEO keywords
0:10:33 I wanna go after.
0:10:36 And then you’d have to hire a bunch of human beings
0:10:39 to create content that would cost a lot of money.
0:10:40 So all these things would add up.
0:10:43 What I’m saying now is it’s now a fraction of the price
0:10:48 to do it and it allows you to compete with the big boys
0:10:51 which is pretty cool.
0:10:52 I do wonder though.
0:10:55 It feels like now that all that’s so easy to make
0:10:58 that Google will have to lean more and more into authority.
0:11:01 Like looking for sites that are like the CNNs
0:11:03 or the reddits or whatever, like Quora.
0:11:06 The recent Google update,
0:11:07 apparently they changed the algorithm
0:11:09 where they’re highly focused on Reddit now.
0:11:11 And people are now trolling Reddit
0:11:12 and putting content on there.
0:11:13 ‘Cause Google doesn’t know what to do.
0:11:16 They’re like, well, trust Reddit, I guess, or Quora.
0:11:18 There’s tons of real people there.
0:11:20 And then now people are kind of gaming that.
0:11:21 And then they probably have another issue too,
0:11:25 like in terms of programmatic SEO.
0:11:28 Have you seen like the perplexity pages that came out?
0:11:29 – Yeah.
0:11:32 – Yeah, so apparently perplexity’s SEO traffic
0:11:34 is like booming right now.
0:11:36 And it’s all from like literally
0:11:39 the AI is creating all the pages, right?
0:11:40 And so now they have a situation where like,
0:11:42 okay, perplexity’s a new authority,
0:11:44 but they’re using AI to create all the contents.
0:11:47 Yeah, maybe Google’s not dead,
0:11:48 but maybe they end up serving up
0:11:50 most of perplexity’s content.
0:11:53 – Yeah, my guess is,
0:11:55 and this gets into like antitrust territory,
0:11:58 but my guess is they are going to
0:12:02 start suppressing perplexity pages.
0:12:03 I think that if you’re out there
0:12:06 and you’re listening to this and you’re like, okay,
0:12:08 you know, I would love thousands of visits
0:12:11 coming to my website every single month.
0:12:12 The thing to think about is,
0:12:15 how do you create the most high quality content
0:12:17 in the most efficient way?
0:12:20 And if you’re able to create high quality content,
0:12:24 the CNNs of the world and the high,
0:12:28 the trusted, great backlinks that you want
0:12:31 are going to come after you just organically
0:12:34 because you’ve created high quality content.
0:12:38 And then of course, you have to do some things,
0:12:40 like for example, not many people know this,
0:12:43 but if you submit your startup to Product Hunt,
0:12:46 Product Hunt is actually a high trust worthy site.
0:12:47 Like the number one reason you should submit
0:12:49 to Product Hunt is for SEO.
0:12:51 And so there’s a bunch of things like that
0:12:53 that you can do that, you know,
0:12:56 I’ve learned over the years that are worth doing too.
0:12:59 (upbeat music)
0:13:00 – We’ll be right back,
0:13:02 but first I want to tell you about another great podcast
0:13:03 you’re going to want to listen to.
0:13:07 It’s called Science of Scaling, hosted by Mark Roberers,
0:13:10 and it’s brought to you by the HubSpot Podcast Network,
0:13:13 the audio destination for business professionals.
0:13:15 Each week hosts Mark Roberers,
0:13:18 founding chief revenue officer at HubSpot,
0:13:20 senior lecturer at Harvard Business School,
0:13:22 and co-founder of Stage 2 Capital,
0:13:25 sits down with the most successful sales leaders in tech
0:13:28 to learn the secrets, strategies, and tactics
0:13:30 to scaling your company’s growth.
0:13:32 He recently did a great episode called,
0:13:35 “How Do You Solve for a Siloed Marketing in Sales?”
0:13:37 And I personally learned a lot from it.
0:13:39 You’re going to want to check out the podcast,
0:13:43 listen to Science of Scaling wherever you get your podcasts.
0:13:46 (upbeat music)
0:13:49 Now, what do you think that like, in the future,
0:13:51 people are even going to like click into the links anymore,
0:13:53 though, because I feel like, obviously,
0:13:55 even Google’s going in the direction now
0:13:56 where if you ask it questions,
0:13:57 it just puts the answer on the homepage,
0:14:00 or people can go to chat GPT or Claude
0:14:03 and ask their question and, you know, get an answer.
0:14:06 And I don’t know if I’m just like, I’m in an AI like tech bubble
0:14:08 where that’s just sort of become my habit,
0:14:11 or if the rest of the world is starting to do that as well.
0:14:12 I don’t, I don’t totally know,
0:14:15 but I feel like I find myself clicking
0:14:16 into links less and less
0:14:19 because the response is just right there for me.
0:14:22 So like, do you think the incentive to even do SEO
0:14:24 is going to go away over time?
0:14:27 – I think that the searches that are going to do really well
0:14:29 on like the perplexities of the world,
0:14:34 AI search are going to be very information-based searches.
0:14:38 So it’s like, what is the weather today in Miami?
0:14:41 You know, Hurricane Barrel is coming.
0:14:44 Is it going to land in Houston?
0:14:46 And I think it’s going to do an incredible job
0:14:49 at like going through the internet
0:14:52 and just boom, here’s your answer.
0:14:54 But there’s a huge sub,
0:14:56 there’s a huge segment of searches
0:15:00 that aren’t as simple as,
0:15:02 let me just pull in the information
0:15:06 that almost reminds me a little bit about,
0:15:07 you know, going old school.
0:15:09 It’s like, you walk into a library, you know,
0:15:11 it’s not like there’s just one book in the library.
0:15:13 There’s like different books in the library
0:15:16 or, you know, you turn on Netflix.
0:15:19 It’s not like you go into one channel
0:15:20 and then it, you know,
0:15:23 just knows what you’re looking for automatically.
0:15:26 Sometimes it’s nice to have some sort of interface
0:15:29 that allows you to jump off to different places.
0:15:34 So I think that the future of search
0:15:37 is going to be a lot more like perplexity
0:15:39 than it is currently Google.
0:15:43 But I think the idea of I’m going to jump around
0:15:44 to different places,
0:15:47 a la library, a la Netflix,
0:15:51 a la, you know, linear TV isn’t going anywhere.
0:15:53 – Okay, so is that the game plan that people do now
0:15:55 is like you throw out tons of AI content
0:15:59 and then you go back with humans to improve
0:16:00 the content that’s actually ranking
0:16:01 and actually make it good?
0:16:04 Is that the kind of the playbook that people are doing?
0:16:06 – I mean, that’s a playbook
0:16:08 that a lot of people who are doing well
0:16:09 in the space are doing.
0:16:11 So I think the first thing I always think about is like,
0:16:14 okay, who’s my ideal customer?
0:16:17 And then based on that, it’s okay.
0:16:18 So I want to track these people
0:16:21 and then what do I want these people to do next?
0:16:24 And for you, it’s probably join the newsletter, I’m guessing.
0:16:26 ‘Cause that way you have their,
0:16:27 okay, once around the newsletter,
0:16:30 it’s like then you can send them the podcast, right?
0:16:34 And then they, it’s all about how do you take a stranger
0:16:36 and make them a raving fan?
0:16:38 Like that’s the game of business in a nutshell.
0:16:40 Everything else doesn’t matter, right?
0:16:43 If you can do those things, then,
0:16:47 and you can do it profitably, you win at business.
0:16:49 Like that’s the most barbarian way
0:16:51 of saying what business is, is that.
0:16:56 So I feel like if Warren Buffett was an internet marketer,
0:16:59 he would be like all about SEO 2.0 right now.
0:17:02 – Well, I also feel like with AI,
0:17:05 you have this attention to almost take this reverse route
0:17:08 because you can create so much content at scale with AI, right?
0:17:11 So you could always go and create
0:17:13 a ton of different content pieces
0:17:17 on a bunch of different AI related topics.
0:17:21 It all would make a good sort of lead into your opt-in,
0:17:24 figure out which content drives the most leads in
0:17:26 and then sort of double down on that content.
0:17:28 I guess that would be another approach as well,
0:17:30 sort of cast a wide net,
0:17:35 figure out which piece of the net catches the most fish
0:17:37 and then hone in on that, you know?
0:17:38 – I think that’s right.
0:17:41 – Well, the other thing I wanna talk about with SEO 2.0
0:17:44 is the fact that large language models in AI
0:17:47 could make SEO a lot more personalized, right?
0:17:49 I think you mentioned that in your SEO 2.0 tweet
0:17:54 is that we’re gonna see SEO and the search results
0:17:59 get a lot more personalized and tailored to the individuals.
0:18:03 So how do you see people sort of leaning into that with SEO?
0:18:06 – So yeah, that’s definitely where the world is going.
0:18:09 I mean, it has been going there even pre-AI,
0:18:13 like in, you know, Google started collecting data around,
0:18:15 okay, we know your location is in Miami.
0:18:16 Therefore, if you write weather,
0:18:18 it’s gonna bring up weather Miami.
0:18:21 It’s not gonna bring up weather Kyoto.
0:18:26 So what AI has done is it has just made that,
0:18:27 we just put that on steroids.
0:18:32 So what the next evolution of that is agents
0:18:34 basically coming to me with,
0:18:38 hey, you need to know this piece of data for this reason.
0:18:41 So the way search works today is it’s very,
0:18:45 what I would call lean forward.
0:18:48 You have to lean forward to get data.
0:18:51 And I think probably where we’re gonna go
0:18:53 in the next few years,
0:18:55 it’s gonna be way more lean back.
0:19:00 So instead of me searching for what’s a great,
0:19:04 you know, sushi restaurant in Miami Beach,
0:19:08 it’s going to know that it’s 6 p.m. my time.
0:19:10 I haven’t made any dinner reservations.
0:19:12 I live in Miami Beach.
0:19:16 I, it knows I love sushi
0:19:19 and I usually have sushi on Tuesday nights
0:19:21 and it hasn’t done that.
0:19:23 And I haven’t, you know,
0:19:24 it’s just like weird and knows me so well.
0:19:26 And it’s like, hey, you know, there’s this news,
0:19:28 you know, sushi restaurants are just open.
0:19:30 I think you’re gonna like it.
0:19:33 Say yes if you want me to make a reservation for two.
0:19:37 Oh, by the way, your friend, Jamie’s in town,
0:19:39 do you want me to add them to the calendar invite?
0:19:42 And that’s like literally the direction of where we’re going.
0:19:43 – Yeah, I could see that.
0:19:46 But I also think that, you know, that description,
0:19:47 I think to some of us,
0:19:50 like the early adopter techie types,
0:19:51 that sounds really exciting.
0:19:53 But then to a lot of people,
0:19:56 that also sounds like more of a dystopian future
0:19:58 where all these companies have so much data
0:20:01 and, you know, information on us.
0:20:03 How do you see these companies sort of balancing
0:20:05 the data privacy concerns,
0:20:07 trying to feed people exactly what they want
0:20:09 when they want it?
0:20:13 – The reality around data and privacy is
0:20:17 people are okay giving away their data and privacy
0:20:20 if it makes their lives way more convenient.
0:20:23 So if the product ends up working
0:20:25 and it makes their lives way more convenient,
0:20:28 then they’re clicking the terms of service
0:20:29 and they’re not reading and scrolling
0:20:30 with the terms of services
0:20:33 and they’re just going and making their lives convenient.
0:20:34 Why?
0:20:39 Because life for everyone is very difficult,
0:20:42 stressful, overwhelming,
0:20:46 no matter how wealthy you are from the top to the bottom.
0:20:51 And I think that if I just believe actually
0:20:55 that a lot of these products are going in the direction
0:20:57 that they’re going to make the lives way more convenient.
0:21:00 So of course they’ll be, you know, let’s put it this way,
0:21:04 Google is way bigger than DuckDuckGo, you know?
0:21:05 – Right.
0:21:09 – It doesn’t mean DuckDuckGo is in a great business
0:21:11 and a great product and there’s a subset of people
0:21:14 who care about privacy first,
0:21:16 but there’s a greater amount of people
0:21:18 that are just like, I’ve got a baby,
0:21:20 I’m holding a baby on this hand,
0:21:22 my boss is calling me on the other hand,
0:21:24 I need to Google search this
0:21:27 and I’m not going to DuckDuckGo to do it.
0:21:29 – Yeah, I always feel like I care about privacy
0:21:31 but as soon as you give me some kind of convenience
0:21:34 that makes my life easier, I’m like, yeah, you know.
0:21:35 – Yeah.
0:21:36 – Give it to me.
0:21:38 – Yeah, and I think, you know,
0:21:39 where the world is probably going to go
0:21:44 and AI plays a role into this too is AI is going to scrape
0:21:46 a lot of data that exists on the internet.
0:21:48 So it’s going to know probably my address
0:21:50 that I live in Miami Beach.
0:21:54 And maybe what happens is it finds out
0:21:58 that a friend of mine wants to send me an invitation
0:22:01 via mail to a wedding, let’s say.
0:22:02 And it’s just going to be like,
0:22:05 hey, do you live at 123 Main Street Miami Beach?
0:22:06 Yes or no.
0:22:10 And so I think the future of a lot of these products
0:22:13 is what I call contextual onboarding
0:22:15 versus upfront onboarding.
0:22:18 Upfront onboarding is when you ask for a million things
0:22:22 upfront, your phone number, your, you know, where you live,
0:22:25 all these things to, and then you say yes to, you know,
0:22:26 each of those things.
0:22:30 I think that with respect to search and other products too,
0:22:33 but let’s just talk about search for a second.
0:22:36 As it gets to know you, it’s convenience.
0:22:37 So it’s going to be like, hey,
0:22:39 do you want us to do this thing for you?
0:22:41 All you have to do is press yes.
0:22:45 And we have this data because it’s scraped.
0:22:46 – Yeah, I mean, at the end of the day,
0:22:48 all of the companies could get access to the data
0:22:49 if they wanted to.
0:22:53 – I think that whether you believe in SEO 2.0 or not,
0:22:57 play with the tools, like go and create AI content,
0:23:01 human layer it, add a calculator to it, add a tool,
0:23:03 like just go and like try things out
0:23:06 and just see how you feel with it.
0:23:10 And through that, you’re going to learn in your niche,
0:23:13 oh my God, you know, this type of calculator,
0:23:15 this kind of tool works.
0:23:18 And to me, like I invest in things
0:23:21 when I can see like a crazy amount of upside return.
0:23:24 And this is something to me that’s like not a lot of,
0:23:26 like it’s not going to take you that long to learn
0:23:29 this sort of stuff and to test this,
0:23:33 but the return on investment, if it works is so large.
0:23:37 – So I want to talk about some of your spicy takes.
0:23:40 You had another tweet where you gave like 25
0:23:42 different spicy takes and Nathan and I actually talked
0:23:44 about it and we pretty much agree with all
0:23:46 of your spicy takes, but I thought there was some fun ones
0:23:49 that we can dive into because they’re obviously
0:23:51 more nuanced than, you know, the single sentence
0:23:53 for each take that you put.
0:23:55 – By the way, for context, whenever I do
0:24:00 those spicy takes things, I like drink a massive
0:24:05 like cold brew or something or like an energy drink
0:24:09 and just like bang them out in like 10 minutes,
0:24:12 press tweet and then just like go for a walk
0:24:14 and then see what happens in like 30 minutes.
0:24:16 – Can’t see how many people disagree
0:24:18 or agree with what you just said.
0:24:18 – Yeah.
0:24:19 – That’s awesome.
0:24:21 Well, you had one of the spicy takes
0:24:23 and I actually commented on your post about it was,
0:24:26 you mentioned like tools like Otter that are going to be
0:24:29 like a note-taking assistant are probably going to flop.
0:24:31 And I’m curious why you think that.
0:24:34 And the reason I ask is because I actually use Otter,
0:24:35 I actually agree that it’s going to flop,
0:24:36 but I think it’s for a different reason
0:24:38 than you mentioned in your tweet.
0:24:40 So I’m curious why you think so.
0:24:43 – So, and our team, by the way, we have like 120 people
0:24:46 on our team and like we use Otter in all these meetings
0:24:48 and whenever I’m in there, I’m always like, it’s me,
0:24:51 it’s like, I like stare down the Otter, you know,
0:24:56 I’m kind of like, you don’t like me and I don’t like you.
0:24:59 And I think a lot of people have that feeling
0:25:02 with respect to an AI note-taking app.
0:25:06 It gives the convenience to the person
0:25:08 who wants to write notes, of course.
0:25:11 So you can be like more present in the meeting.
0:25:14 Like that’s the, that’s who people who like tools like that,
0:25:15 that’s what they say to me.
0:25:17 They’re like, yeah, like I just want to be present
0:25:20 in the meeting, therefore this is going to help me,
0:25:21 you know, be present.
0:25:25 And it’s like, homie, can’t you just do two things at once?
0:25:28 Like, you know, I remember being in the third grade
0:25:32 taking notes, you know, in a classroom, you know,
0:25:35 I think what people don’t like about AI note-taking tools,
0:25:38 and this, I do believe this is 90 plus percent of people,
0:25:43 is they can’t be free.
0:25:45 They can’t, you know, maybe they want to swear
0:25:47 and they can’t, they don’t want to swear.
0:25:49 And I think it hurts more productivity
0:25:51 than it does help productivity.
0:25:53 That’s my spicy take with it.
0:25:56 And I also just think that, yeah,
0:25:58 the only person that enjoys the Otter
0:26:02 is the person who is supposed to be taking notes.
0:26:05 And I also think, yeah, I’m just, I don’t,
0:26:09 I think go take notes, man, just go take some notes.
0:26:10 – I mean, at the end of the day too,
0:26:11 if you’re sitting there taking notes,
0:26:13 that’s sort of turning it into a multimodality
0:26:15 sort of engagement, right?
0:26:18 You’re listening and you’re also writing,
0:26:19 and because you’re doing both,
0:26:21 you’re going to help lock a lot of those concepts in.
0:26:22 – My take is there’s going to be a backlash
0:26:24 on this sort of stuff.
0:26:27 I feel like people now are kind of just like
0:26:31 politely smiling and they’re like, that’s cute.
0:26:34 But I think in a year or two,
0:26:37 people are going to be like, hey, hey, do you mind just,
0:26:39 I mean, you’re starting to see it a little bit,
0:26:41 but you know, I’m starting to see like,
0:26:42 hey, do you mind just like removing that recorder
0:26:44 from that meeting?
0:26:48 I think you’re going to see instances of that, 50X.
0:26:53 – I’ll tell you, so here’s my sort of killer use
0:26:55 that I love using Otter for.
0:26:57 I actually go to like a lot of conferences
0:26:59 to make content around them for my YouTube channel.
0:27:02 So, you know, I was at Cisco Live
0:27:03 and I was at Augmented World Expo
0:27:07 and Google I/O, Microsoft Build, all these events.
0:27:10 And a lot of times I’m sitting in these conferences
0:27:14 while they’re speaking and it’s hard to take
0:27:15 all the notes on everything they’re talking about.
0:27:17 So I usually open up Otter on my phone,
0:27:20 just set it next to me on my chair
0:27:23 and let Otter actually record the entire, you know,
0:27:26 keynote or panel or whatever I’m sitting and listening to.
0:27:29 And then I’ll take like the summary and the transcript
0:27:32 and I’ll pull it over into a clod or a chat GPT
0:27:35 and say, hey, I need to make a video about this.
0:27:37 What are some of the talking points from this presentation
0:27:39 that I should bring up in this video?
0:27:41 And it will actually help me sort of like
0:27:43 outline rough draft a presentation
0:27:45 about the keynote that I just saw.
0:27:47 And I found that to be really helpful.
0:27:49 Saying that, I also agree,
0:27:51 it’s probably not going to be around for a long time
0:27:54 and it’s going to flop, but I think the reason is more
0:27:56 because I think it’s just going to be a feature
0:27:58 instead of an actual app in a lot of these devices.
0:28:02 I think we’re going to see Google just roll it into Android
0:28:03 as part of the operating system.
0:28:05 I think we’re going to see Apple intelligence
0:28:07 just roll that kind of feature in.
0:28:10 And I can’t see companies like Otter existing
0:28:13 when it’s just like a native built-in feature
0:28:15 of these devices.
0:28:19 – So, well, where I’ll agree with you is in the,
0:28:21 what I’ll call like broadcast setting
0:28:23 if like you’re in a conference
0:28:26 and there’s tons of people around you
0:28:29 in groups of larger than 10,
0:28:31 something like Otter super important.
0:28:34 Like I can see it being super valuable, super important.
0:28:38 But within the context of I’m a company
0:28:39 and I’ve got all these meetings
0:28:41 and a lot of the meetings are just
0:28:46 hanging out or talking about one issue.
0:28:49 And they’re smaller.
0:28:52 Like maybe the average meeting is four people.
0:28:57 That’s where I think Otter is going to really flop.
0:28:58 – Yeah, yeah.
0:28:59 I can see that take too.
0:29:02 I mean, I don’t use it on like Zoom calls and things like that.
0:29:04 So it’s not actually one of the use cases
0:29:05 I’ve actually used it for.
0:29:08 So I can definitely see that take.
0:29:10 – So you said AI is taking jobs
0:29:12 not just making people more productive.
0:29:13 Companies want profits.
0:29:16 The idea that AI frees you to focus on your passion
0:29:18 is pure fantasy for most workers.
0:29:19 You know, I kind of agree with this.
0:29:21 I feel it’s a situation where everyone in AI right now
0:29:24 they can’t say that though.
0:29:25 There’s an odd thing there, right?
0:29:26 Where you just, you feel like you can’t say
0:29:28 the truth about the thing.
0:29:30 ‘Cause it leads to like, okay, well, you know,
0:29:31 does this change capitalism?
0:29:32 Like I’m a capitalist.
0:29:34 Like does capitalism make sense in the future?
0:29:37 Does our current government and economy
0:29:40 make sense in the future where, yeah,
0:29:42 you can use AI for most things
0:29:43 and maybe everyone doesn’t have to work the same way.
0:29:45 Is that kind of like the gist of,
0:29:47 is that how you see it as well or?
0:29:50 – Sometimes I’m scrolling Twitter
0:29:53 and I’m seeing like these tweets where it’s like,
0:29:57 AI is gonna make, you know, there’s gonna be more jobs.
0:29:58 You know, it’s making more jobs.
0:30:02 And I’m like, what are you smoking then?
0:30:04 Like honestly, I feel like I’m living
0:30:05 in a completely different planet.
0:30:10 I’m like, okay, there’s this new paradigm shift
0:30:13 that has basically made it 100 times easier
0:30:18 to do basically anything from creative to software,
0:30:23 to marketing, to all these different things.
0:30:24 So that’s happening.
0:30:26 Everything’s become easier.
0:30:30 At the same time, companies are looking
0:30:34 to become more profitable and especially larger companies,
0:30:36 especially VC-backed companies
0:30:38 who are trying to exit or get acquired
0:30:41 and publicly traded companies
0:30:44 whose only job is to increase shareholder value.
0:30:47 Once they get comfortable to the point
0:30:48 where people are using these tools
0:30:50 and they’re becoming way more productive,
0:30:51 they’re not gonna be like,
0:30:55 oh, hey, let’s go hire another thousand people.
0:30:56 So yeah, I just think that people
0:30:58 are gonna be way more productive.
0:31:00 And if people are more productive,
0:31:04 there’s going to be less of those types of jobs.
0:31:08 – So what do you think, just like very theoretical, right?
0:31:10 What do you think the job market looks like
0:31:12 in several years from now?
0:31:14 – There’s a lot less engineers.
0:31:16 There’s a lot less designers.
0:31:19 I think there’s a lot less marketers.
0:31:23 I think the top 1% ends up getting paid 10 times.
0:31:27 I think the middle gets stripped out.
0:31:28 I think that’s what happens.
0:31:32 I think the middle of these white collar jobs
0:31:34 becomes stripped out.
0:31:37 You’re either junior and you’re like very low paid
0:31:41 or you’re senior and you’re very high paid.
0:31:44 You know, I wish I wasn’t saying this.
0:31:45 I agree with you.
0:31:46 Yeah, I think we both agree.
0:31:47 – It’s hard to say, right?
0:31:48 Cause we don’t know the answer.
0:31:50 Like we don’t know the answer for those people.
0:31:53 Like what do they do and how do they earn a living?
0:31:56 And yeah, sure the AI tools will make their life better,
0:32:00 but they also still need food on the table, right?
0:32:02 – Right, exactly.
0:32:05 – So do you think this, I mean, obviously,
0:32:06 none of us know where it’s all headed,
0:32:09 but do you think this leads to like a UBI scenario?
0:32:12 Do you think like, where do your predictions lie
0:32:13 if you had to make predictions?
0:32:15 – I think it’s important to just share your opinion
0:32:17 and it’s okay if your opinion changes over time
0:32:18 in real time.
0:32:20 And as you learn about a particular topic,
0:32:22 and I think that if more people do that,
0:32:24 like the better society is,
0:32:26 as long as you can be open-minded about, okay,
0:32:28 the data has changed.
0:32:31 Therefore my stance on something has changed.
0:32:33 As far as like do, you know,
0:32:37 in a situation where millions of people are out of work,
0:32:39 is there some sort of UBI?
0:32:43 I mean, I don’t wanna speak for Sam Altman,
0:32:46 but like, why is he creating Worldcoin
0:32:49 if he isn’t thinking about that?
0:32:53 That UBI is gonna be a more important part of,
0:32:56 you know, as a mechanism to pay certain people.
0:32:58 You know, what do I think?
0:33:00 I stop there.
0:33:03 Like, I don’t know how, you know, I’m not a politician.
0:33:05 I’m not an economist.
0:33:07 I’m a technologist that knows
0:33:08 where the technology is going.
0:33:09 Where it goes from there?
0:33:11 Like, I’m not, it’s above my pay grade.
0:33:14 Well, yeah, you have to be like a philosopher,
0:33:15 psychologist, everything, right?
0:33:18 Like, it’s like, how do humans, you know,
0:33:20 interact or how humans live if they don’t have work?
0:33:21 And how does that work?
0:33:23 And how do you feel motivated?
0:33:26 And especially if other people are getting 10x, right?
0:33:29 And they’re living this amazing new life,
0:33:30 and then you’re kind of like living
0:33:33 in your virtual reality world or whatever.
0:33:36 Yeah, I mean, I don’t think there’s gonna be UBI.
0:33:41 I think people will find different types of work
0:33:45 and be productive to society.
0:33:47 What that work will look like,
0:33:48 I think will be different, very different
0:33:50 than what it is today.
0:33:54 I think that the wealth disparity, unfortunately, will grow.
0:33:56 And that’s why I think it’s important for everyone
0:33:58 to just stay ahead.
0:34:00 That’s why this podcast is great, right?
0:34:03 Like, learn, get your hands dirty,
0:34:08 and you’ll never have to worry if that’s the case
0:34:13 because you’ll always be one step ahead of other people.
0:34:16 People don’t put in the work.
0:34:18 People just don’t put in the work, right?
0:34:19 Even when the work gets easier.
0:34:24 They much rather watch episode 124 of Love Island
0:34:25 than episode 21.
0:34:27 Is that tonight?
0:34:28 The next wave.
0:34:29 I think that’s true.
0:34:32 I think it, you know, it’s people wanna,
0:34:33 you know, we talked about this before.
0:34:34 It’s like, life is hard, right?
0:34:36 And it’s like, if you wanna numb,
0:34:38 you wanna numb your brain sometimes,
0:34:39 but sometimes like,
0:34:42 sometimes you gotta keep your brain stimulated
0:34:44 in order for you to stay ahead.
0:34:48 – Yeah, I mean, along that same lines,
0:34:49 I’m gonna actually jump ahead
0:34:51 to one of your other hot takes
0:34:53 because it’s very relevant to what you just said.
0:34:56 You said more young millionaires will get minted
0:34:58 over the next 10 years than any point in the history.
0:35:00 But watch YouTube, listen to podcasts,
0:35:03 and learn to take ideas into reality thanks to AI.
0:35:06 It’ll make many people jealous.
0:35:08 So, I mean, there, you just kind of mentioned
0:35:10 there’s gonna be a growing disparity
0:35:12 between I guess, you know, the haves and have nots.
0:35:14 The people with a lot of money
0:35:16 versus people with not a lot of money.
0:35:19 But also, to sort of counteract that point,
0:35:20 it also sounds like you’re saying
0:35:23 there’s no easier time in history to make money.
0:35:25 – Absolutely.
0:35:27 I mean, you’re connected to the internet.
0:35:30 And so you have access to a library
0:35:34 of the smartest people on the planet, basically,
0:35:35 in your pocket.
0:35:40 So, and then when you wanna create something,
0:35:43 now you’ve got, you know, AI tools and no-code tools
0:35:46 that you can actually just go and press some buttons,
0:35:49 not have to have, you know, gone to Stanford
0:35:53 and studied computer science with, you know,
0:35:54 all these well-known people.
0:35:57 And you can just put stuff on the internet
0:35:59 and then literally billions of people
0:36:03 have their credit cards on the internet
0:36:05 and you can sell to anyone.
0:36:08 It used to be that, you know, my dad had a store
0:36:13 and he had a store in this, you know, little area.
0:36:16 And if more people moved to the area,
0:36:18 there’d be more people and if people left,
0:36:23 then less people would buy at the store, you know?
0:36:25 Now, the store is everyone
0:36:29 and you don’t need to go and spend two years
0:36:31 building a building.
0:36:34 You can spend two minutes to build a building.
0:36:38 So I think there’s gonna be a lot of low-quality
0:36:43 and medium-quality startups, content, communities
0:36:45 that are gonna exist.
0:36:48 And it’s just gonna feel like,
0:36:52 it’s just like how in marketing, like if you go,
0:36:55 I saw like a video of like 1919 in New York City
0:36:59 and there’s like no billboards and there’s like no marketing.
0:37:03 And then if you go in 2024, New York, like of course,
0:37:06 there’s like your eyes constantly looking
0:37:08 at a million different marketing, you know?
0:37:11 I think that’s gonna, that’s the world we’re headed in
0:37:12 in terms of the internet.
0:37:14 It’s gonna look a lot less like 1919,
0:37:16 a lot more like 2024 times square.
0:37:20 And if you need, if you wanna stand out, you know,
0:37:23 those are where the young millionaires are gonna mint it.
0:37:27 And I think why are they gonna be young?
0:37:31 It’s because the young people are gonna be native
0:37:33 to the tools to create the stuff.
0:37:36 – Yeah, one thing I feel like, you know,
0:37:38 even though this technology is gonna create a,
0:37:41 maybe a bigger divide in terms of the haves and haves nots,
0:37:43 I do feel like it’s gonna equalize things in a way
0:37:47 where you don’t have to be the smartest person
0:37:49 to use AI and make something cool.
0:37:51 So like a lot of times I feel like it’s gonna matter more
0:37:54 about like how motivated is the person, right?
0:37:55 Yeah, they didn’t go to Stanford,
0:37:56 they didn’t have the highest GPA,
0:37:58 but they can, if they’re motivated,
0:38:00 they’ll still be able to use AI tools
0:38:02 and build a company and build stuff, right?
0:38:05 So I mean, I find that personally inspiring.
0:38:08 And like, you know, if people really wanna make it,
0:38:10 AI is gonna help them make it more than ever before.
0:38:14 – Absolutely, absolutely.
0:38:18 I think if you’re motivated and you’re ambitious,
0:38:19 you’re gonna do well.
0:38:20 You have nothing to worry about.
0:38:22 There’s one last topic I wanted to talk about
0:38:25 before we let you go, and that’s the climate issue.
0:38:27 ‘Cause you mentioned that there’s going to be
0:38:29 a major climate backlash against AI,
0:38:32 it’s gonna become the current thing to talk about.
0:38:34 And Nathan and I, even before this call,
0:38:35 we’re actually talking about that.
0:38:36 We both agree with you on that.
0:38:39 I think that’s gonna be the big narrative.
0:38:42 But when you made that hot take, like,
0:38:43 what was your perspective?
0:38:45 Where were you coming from on it when you wrote it?
0:38:50 – Every major wave of technology, computer,
0:38:55 computers, the internet, social, mobile, crypto,
0:39:01 there’s been a point in time where those technologies
0:39:06 get so big that there becomes a climate backlash.
0:39:10 If you look at the maximums of the,
0:39:13 like if you were to plot out, you know,
0:39:14 number of articles, let’s say,
0:39:18 that have the climate backlash,
0:39:20 every peak has gotten higher.
0:39:25 So we saw this, I think, in 2021 with crypto,
0:39:30 there started to become a huge backlash
0:39:32 that, and people started saying, you know,
0:39:35 if we shut down Bitcoin, like we can power,
0:39:37 like, you know, all of Ukraine
0:39:39 and all the, you know, different places, right?
0:39:43 So, and I think that someone is going to
0:39:47 start telling the story that these technologies
0:39:51 take compute and compute isn’t free.
0:39:56 It requires electricity, requires the actual GPUs,
0:39:58 and that affects the environment.
0:40:03 So for me, it’s not a matter of if,
0:40:04 it’s a matter of when.
0:40:05 – Yeah.
0:40:07 – And, you know, from an opportunity perspective,
0:40:09 like, there’s probably opportunities to create more,
0:40:14 like, climate-friendly, clean, carbon-neutral AI technologies.
0:40:18 So if, you know, we were talking about DuckDuckGo,
0:40:20 DuckDuckGo versus Google,
0:40:23 I think that if, you know, that was the,
0:40:25 you know, riding the privacy-first trend,
0:40:27 I think there’s also a climate-first trend.
0:40:29 So I think that there’s an opportunity to create,
0:40:32 like, what does the green version of perplexity look like?
0:40:34 – Yeah, it feels like, unfortunately, this makes,
0:40:37 you know, AI is going to get political for this reason,
0:40:40 which I hate, you know, because there’s obviously a divide
0:40:43 of, like, okay, are we going to stay only on Earth,
0:40:46 and are we going to, you know, solve climate issues
0:40:49 by kind of being more efficient and having less people
0:40:52 and less waste, or are we going to solve climate problems
0:40:55 by using technology to solve those problems
0:40:58 and going to other planets, right?
0:40:59 And so a lot of the techno-optimists, like, you know,
0:41:03 we believe, like, that obviously AI, you know,
0:41:05 if it takes more energy, that’s okay.
0:41:07 Like, AI is gonna help us solve the problems,
0:41:08 it’s gonna help build the technologies
0:41:09 we don’t even know exist yet
0:41:11 that are gonna solve the problems,
0:41:14 and you can’t get out of it just by having less,
0:41:17 you know, plastic straws or whatever.
0:41:19 But obviously, there’s a huge divide there,
0:41:20 and it gets really political, and so yeah,
0:41:23 I hate that, unfortunately, AI is probably gonna get
0:41:24 pretty political in the next year or so,
0:41:25 and I think it will.
0:41:28 – Yeah, yeah, I’ve been hearing a lot of this stuff
0:41:31 about, like, you know, the climate issues
0:41:33 and how much power all of this uses,
0:41:34 and all of that kind of stuff,
0:41:36 but I’ve also been seeing both sides of the narrative.
0:41:39 I’ve also been seeing a lot of articles and reports
0:41:42 and stuff that say it’s way overblown, too.
0:41:44 I’ve seen things saying that in 2020,
0:41:46 when everybody was in their homes
0:41:50 and, like, everybody started playing video games more often,
0:41:53 that used as much compute as what AI is using right now,
0:41:55 right, and then there’s also the sort of narrative
0:41:57 of, like, NVIDIA and Qualcomm
0:42:00 and all these companies that are making the processing,
0:42:02 their number one goal, anytime they give a keynote,
0:42:05 is to talk about how they’re trying to make it more efficient
0:42:06 and bring down the energy usage
0:42:08 and get more and more powerful compute,
0:42:11 but for less the energy use.
0:42:13 So I do think a lot of these problems will be solved,
0:42:17 but I also think this is gonna be, like, the narrative,
0:42:19 the popular narrative against AI,
0:42:21 but I also think a lot of this,
0:42:24 there’s solutions being worked on right now.
0:42:25 – It is gonna get more efficient,
0:42:27 the graphics charge is gonna get more efficient,
0:42:28 but there’s a reason that, like, Elon Musk
0:42:31 is, like, doubling down on building, you know,
0:42:34 factories and stuff in Texas, right?
0:42:35 Like, aligning himself with, like,
0:42:37 people who are, like, more pro-energy
0:42:40 and, like, have more energy, you know, consumption,
0:42:42 is because that’s probably where things are gonna go.
0:42:44 Like, yeah, in the current state,
0:42:46 yeah, it doesn’t use too much energy,
0:42:48 but when we get to AGI, (laughs)
0:42:50 ASI, yeah, you’re probably talking about
0:42:52 massive amounts of energy consumption,
0:42:55 probably the likes that humanity’s never seen before.
0:42:56 – So this is really interesting.
0:43:00 I asked, I prompted perplexity and I said,
0:43:03 “What’s the ecological impact of a chat GPT prompt?”
0:43:06 I was gonna say perplexity prompt,
0:43:09 ’cause I was like, I want something unbiased, right?
0:43:10 (laughs)
0:43:11 So, it says this.
0:43:15 The ecological impact of a single chat GPT prompt
0:43:19 is relatively small, but it can add up significantly
0:43:21 when considering the massive scale of usage.
0:43:22 Here’s the breakdown.
0:43:25 Each chat GPT query is estimated to produce
0:43:28 approximately 4.32 grams of CO2,
0:43:29 but here’s where it gets really interesting.
0:43:32 I didn’t realize, but, I mean, silly me,
0:43:35 that there’s water consumption with chat, with this.
0:43:38 So, chat GPT’s water usage is particularly noteworthy.
0:43:40 For every 20 to 50 queries,
0:43:43 the system consumes about 500 milliliters of water,
0:43:46 equivalent to a standard bottle of water.
0:43:50 This water is primarily used for cooling the data centers
0:43:53 that power the AII model, so really interesting.
0:43:56 And then, there’s more, and then there’s energy usage.
0:44:00 So, the energy consumption of chat GPT is considerable.
0:44:03 The system runs on an estimated 30,000 GPUs,
0:44:07 which requires significant power to operate.
0:44:10 So, to put this in perspective,
0:44:13 15 queries are equivalent to watching one hour video.
0:44:16 16 queries consume as much energy as boiling one kettle.
0:44:19 139 queries use as much energy
0:44:22 as watching one load of laundry.
0:44:25 And if current growth trends continue by 2027,
0:44:27 chat GPT’s electricity consumption
0:44:30 could rival entire nations like Sweden, Argentina,
0:44:32 and the Netherlands.
0:44:36 – Well, I think the important part of this whole conversation
0:44:39 is that people have conversations about it, right?
0:44:41 Because I think when it comes to AI right now,
0:44:43 the pro AI people want to pretend
0:44:46 like these negatives don’t exist.
0:44:49 And the anti AI people want to act like
0:44:51 solutions can’t exist.
0:44:55 And I think the important thing is that we,
0:44:59 I don’t feel like AI should be this divisive topic.
0:45:02 I feel like it should be a topic
0:45:05 that we’re all talking about and collaborating on
0:45:07 and figuring out solutions together,
0:45:09 both on the pro and anti AI side.
0:45:11 How do we find this middle ground
0:45:12 that’s gonna make everybody happy?
0:45:15 That’s going to solve a lot of these problems.
0:45:16 And I think one of the frustrations
0:45:21 that I’ve had recently just being so deep in the AI space
0:45:22 is it’s all so binary.
0:45:24 It seems like everybody’s either,
0:45:27 I’m totally against AI or I’m totally for AI,
0:45:28 but I believe there’s a spectrum there
0:45:31 and you can land anywhere on that spectrum.
0:45:33 Well, I think that’s just the bigger issue
0:45:36 with human beings, especially now is like,
0:45:39 you have to be either this side or that side
0:45:41 and there’s no in between, there’s no gray,
0:45:43 like gray is dead now.
0:45:48 And that’s just, it’s just being applied to AI.
0:45:54 So let’s leave it on a positive note.
0:45:56 (laughing)
0:45:57 Go create some stuff.
0:46:00 You know what I mean?
0:46:04 You are listening to this, you understand the tools,
0:46:07 you just gotta go build some stuff
0:46:10 and go and create stuff that people are gonna love
0:46:12 that are gonna put smile on their faces.
0:46:14 It’s gonna add value to their lives,
0:46:16 make them happier and healthier
0:46:20 and contribute to making the world a better
0:46:23 and happier place and you can do that now.
0:46:26 Well, on that note, you make a lot of great content.
0:46:29 Everybody needs to go follow Greg on Twitter X.
0:46:30 Great content there.
0:46:32 What’s the name of your podcast again?
0:46:33 – The startup ideas podcast.
0:46:36 – The startup ideas podcast, go check that out.
0:46:37 Well, thanks so much for joining us.
0:46:39 This has been an amazing conversation.
0:46:41 I’m sure it won’t be the last mode we have on this show
0:46:44 and I appreciate you joining us.
0:46:45 – Thanks for having me.
0:46:48 (upbeat music)
0:46:51 (upbeat music)
0:46:53 (upbeat music)
0:46:56 (upbeat music)
0:46:58 (upbeat music)
0:47:01 (upbeat music)
0:47:11 [BLANK_AUDIO]
Episode 16: How is AI transforming the future of SEO and job markets? Matt Wolfe (https://x.com/mreflow) and Nathan Lands (https://x.com/NathanLands) are joined by innovator Greg Isenberg (https://x.com/gregisenberg), founder of Late Checkout and Boring Marketer. Greg hosts “The Startup Ideas Podcast”.
In this episode, the trio delves into the shift from traditional SEO practices to AI-powered SEO 2.0. Greg explains the role of AI in creating personalized content, the pros and cons of AI note-taking tools, and the potential impacts of AI and automation on job markets. They also touch on the environmental and political implications of AI advancements and explore strategies for generating high-quality, interactive content that ranks well on search engines.
Check out The Next Wave YouTube Channel if you want to see Matt and Nathan on screen: https://lnk.to/thenextwavepd
—
Show Notes:
- (00:00) AI to revolutionize content creation and ranking.
- (06:30) Maximizing SEO through embedding tools and calculators.
- (08:13) AI for SEO needs differentiation and interactivity.
- (11:21) Create high-quality content for organic website traffic.
- (14:56) Turning a stranger into a fan – the business game.
- (20:18) AI will scrape internet data for convenience.
- (21:30) Explore SEO 2.0, create AI content.
- (25:32) Using Otter app for easy conference note-taking.
- (28:28) There will be less certain types of jobs.
- (30:54) Share evolving opinions openly and be open-minded.
- (34:05) Access to global knowledge and market simplified.
- (40:08) Climate, AI power usage, tech solutions narrative.
- (43:13) AI conversation should unite, not divide.
- (44:12) Humans polarized, but AI offers positivity.
—
Mentions:
- Greg Isenberg: https://www.gregisenberg.com/
- Late Checkout: https://www.latecheckout.studio/
- Boring Marketer: https://www.boringmarketing.com/
- The Startup Ideas Podcast: https://podcasts.apple.com/us/podcast/the-startup-ideas-podcast/id1593424985
- Otter.ai: https://otter.ai/
- Product Hunt: https://www.producthunt.com/
- Perplexity: https://www.perplexity.ai/
—
Check Out Matt’s Stuff:
• Future Tools – https://futuretools.beehiiv.com/
• Blog – https://www.mattwolfe.com/
• YouTube- https://www.youtube.com/@mreflow
—
Check Out Nathan’s Stuff:
- Newsletter: https://news.lore.com/
- Blog – https://lore.com/
The Next Wave is a HubSpot Original Podcast // Brought to you by The HubSpot Podcast Network // Production by Darren Clarke // Editing by Ezra Bakker Trupiano