Chit Chat Stocks - The Future of Artificial Intelligence (AI) and Cloud Computing With Shawn Wang (OpenAI, Google, + More)

Episode Date: December 18, 2024

On this episode of Chit Chat Stocks, Ryan and Brett speak to Shawn Wang of Latent Space, who is an expert and on the ground working in the AI start-up space. We discuss: (00:00) Introduction to AI and Cloud Computing (02:21) Expertise in AI and Foundation Models (04:43) Changes in Cloud Revenue Growth (06:45) The Cost of AI Infrastructure (10:36) The Landscape of AI Startups (14:09) Building Moats in AI (15:45) Evaluating Major AI Players (16:41) OpenAI's Current Standing (20:10) First Mover Advantage in AI (25:48) Google's AI Strategy and Challenges (32:51) Meta's AI Transformation (36:59) Amazon and Microsoft's AI Strategies (40:16) Startup Spending and Growth Justification (43:02) The Risks and Challenges of AI Investment (46:27) Cost Reduction and the Future of AI Training (49:22) The Rise of AI Engineers (54:46) Autonomous Agents and Their Impact (01:01:22) The Future of Robotics and AI Integration (01:05:400 Industries Affected by AI Advancements SUBSCRIBE TO LATENT SPACE:https://www.latent.space/ ***************************************************** JOIN OUR FREE CHAT COMMUNITY:https://chitchatstocks.substack.com/  ********************************************************************* Sign-up for a bond account atPublic.com/chitchatstocks  A Bond Account is a self-directed brokerage account with Public Investing, member FINRA/SIPC. Deposits into this account are used to purchase 10 investment-grade and high-yield bonds. As of 9/26/24, the average, annualized yield to worst (YTW) across the Bond Account is greater than 6%. A bond’s yield is a function of its market price, which can fluctuate; therefore, a bond’s YTW is not “locked in” until the bond is purchased, and your yield at time of purchase may be different from the yield shown here. The “locked in” YTW is not guaranteed; you may receive less than the YTW of the bonds in the Bond Account if you sell any of the bonds before maturity or if the issuer defaults on the bond. Public Investing charges a markup on each bond trade. See ourFee Schedule. Bond Accounts are not recommendations of individual bonds or default allocations. The bonds in the Bond Account have not been selected based on your needs or risk profile. Seehttps://public.com/disclosures/bond-account to learn more. ********************************************************************* FinChat.io is The Complete Stock Research Platform for fundamental investors. With its beautiful design and institutional-quality data, FinChat is incredibly powerful and easy to use. Use our LINK and get 15% off any premium plan:⁠finchat.io/chitchat  ********************************************************************* Sign up for YellowBrick Investing to track the best investing pitches across the internet:joinyellowbrick.com/chitchat ********************************************************************* Bluechippers Club is a tight-knit community of stock focused investors. Members share ideas, participate in weekly calls, and compete in portfolio competitions. To join, go to ⁠Blue Chippers and apply! Link:⁠https://bluechippersclub.com/ ********************************************************************* Disclosure: Chit Chat Stocks hosts and guests are not financial advisors, and nothing they say on this show is formal advice or a recommendation. Learn more about your ad choices. Visit megaphone.fm/adchoices

Welcome to Chit Chat Stocks. Before we get into this episode, we want to talk about our friends at Public. Heads up, folks. Interest rates are falling, but you can still lock in a 6% or higher yield with a bond account at Public.com. That's a pretty big deal because when rates drop, so can the interest you earn on your investment. A bond account allows you to lock in a 6% or higher yield with a diversified portfolio of high-yield and investment-grade corporate bonds. So while other people are watching their returns shrink, you can sit back with regular interest payments, but you might want to act fast because your yield is not locked in until you invest. The good news, it only takes a couple of minutes to sign up at public.com. Lock in a 6% or higher yield with a bond account only at public.com forward slash chitchat stocks.

welcome to chit chat stocks on this show host ryan henderson and brett shaffer analyze businesses and riff on the world of investing as a quick reminder chit chat stocks is a ccm media group podcast anything discussed on chit chat stocks by ryan brett or any other podcast guest is not formal advice or recommendation now please enjoy this episode welcome in we have another episode of the chit chat stocks podcast for you this week we have a fantastic interview coming up with sean wang from latent space uh it is a i'll say covering anything ai uh a lot of the stuff you know for me and ryan it might be going over our heads and it might be a little too hard technically for us, but that's what we brought on Sean to the show today. For a

lot of public market investors, this new AI stuff, it's hard to see what is working, what's just a narrative, all this stuff that's getting thrown at us during this boom times. So Sean, we wanted to bring you back on the show. You came on, I think almost exactly two years ago now to talk more cloud stuff. Now we're really going to talk about cloud AI and how it is impacting the startup ecosystem so sean as we kick off the show what is your relevant expertise in this booming ai field oh god that's the uh million dollar question here so i am um so for for listeners who haven't heard back the the previous episode that was on uh i was um finance and public markets guy. And I was in Bali as a hedge fund for my first career. And then I changed careers to tech

where I worked at AWS and three unicorn sort of developer tooling cloud startups. My relevant expertise is, you know, on some level, I'm just a software engineer that is building with AI now. And then on another level, I actually, when I was an options trader back in the sell side, I actually did a lot of natural language processing of the Bloomberg chats. So I fed all of the Bloomberg chats into a pricing mechanism,

then built our global pricer. So our entire options desk was running off of that thing. This was about 13 years ago. So I've always had some involvement with AI, but it was never a big part of my identity. And I think the more foundation models came into focus, and foundation models is a very special term, as opposed to traditional, maybe machine learning finance that a lot of your listeners might be familiar with, then you start to build differently. And there, the traditional software engineer skills become a lot more relevant. So the relevant expertise now is that I guess I've sort of popularized and created the term of AI engineer, which you can talk about, and created the industry such that Gartner now considers it like the peak of its hype right now.

And I consider that both a point of success and also a challenge because I have to prove Gartner wrong that it has not peaked. But, you know, they put us at the top of the hype cycle, which is kind of funny because I started it. Yeah, it's a unique challenge. But yeah, funny anecdote. Okay, so a lot has changed since we last spoke. Pretty much this whole world of AI that everyone's talking about now, or at least has become mainstream has, I believe that kind of kicked off right after the discussion, or our last discussion. So I guess the last discussion was really focused on the cloud computing industry broadly. And that was actually right around the time when AWS, Azure, GCP, all the revenue growth rates were coming down and actually now with the hindsight bottoming. So my question for you is, I guess, what has changed over the last two years and why has revenue growth at the big cloud providers reaccelerated? Yeah, again, like revenue growth at big cloud providers is due to factors that, you know, probably I don't have a full appreciation of. I also challenge the fact, the idea that everything has changed. You know, I think in some ways this is just like the next wave of something that was just a broader, maybe like 20, 30 year long trend anyway.

So, you know, we needed more cloud compute. Now we need even more cloud compute. Now we need more GPUs in the cloud instead of CPUs, right? Like what's really changed? I don't know. People still want serverless everything. People still want orchestration. People still want unlimited storage and bandwidth and all the core components of cloud. So in that sense, it hasn't really changed. I do think that if you see their plots over time of the amount of money and flops invested in machine learning models, that actually used to follow a pretty log linear Moore's law type growth chart for the last 40 years.

And then you had 2022 happen, and now everyone's like, oh, you can train foundation models now. And actually, you've seen a big inflection upwards in the amounts that people are throwing the money in there just because they see the money now. So it's obvious to everyone, including us, including me, in a way that it wasn't obvious to basically everyone but Sam Altman and Satya Nadella circa 2019. like they knew this four years uh five years ahead of everyone else and that's why they went big on open ai but now that we see this obviously everyone's throwing money into nvidia basically why why are and this is maybe a question i think i know but i'd like the answer again and it feels like it's maybe a basic question but a lot of i think listeners are going to want to kind of understand this connection. Why do these new AI companies

require so much upfront spending on NVIDIA chips, cloud computing costs, all that stuff? Yeah, I mean, so you have to split it by whether you're a foundation model lab, or you're basically everyone else that consumes foundation models. So the rough estimate for, let's say GPT-3 was like 50 million to 100 million in compute for one run. And for every one successful final training run, maybe you have between 100 to 1,000 prior runs before that, right? So just pure R&D. The estimate for GPT-4 was 500 million. We've actually had two generations of frontier models since then just for OpenAI. So that would be GPT-4.0 and 0.1.

Those are the models that, those are only the models they've released. And also not, those are only the text models that we haven't counted the video models and all the other stuff. So it's just a lot of upfront investment, right? Like I think it's like the classic capital fixed costs upfront thing

where you have a pre-training phase where you're just consuming all of the internet data. That's, there's nuances to that, but we won't go into that. Um, and at the outcomes, the other ends, you know, three to six months later, you outcomes, a model that you then spend another six months fine tuning, um, and red teaming and post-training. And then it's ready for release. Um, so like, so there's, there's opening eye doing that. And then everyone else trying to copy them.

Anthropics, the most successful so far, but there are others kind of on their tail. Uh, I would say Cohere, XAI, Meta, you know, all these other companies that we're, we're naming there. There's sort of like a second tier of frontier model labs that are out there. And all of them need enormous amounts of compute just for training. Once they're successful training, then they can start serving the models to people to build on top of. And that compute workload starts switching to inference. So it's a classic question of how much money goes into compute and how much goes into inference. It's typically between a two to three ratio of computer inference or a three to two ratio, depending on how you set up.

So that's what Google DeepMind runs at. It's like a three to two, two to three ratio. And I think it makes sense. The question now, though, is like OpenAI's finances are relatively public and they basically make zero margin on their business right now as it stands. which is really interesting because then you should use the crap out of it because they're giving it to you for nothing once you advertise all the costs. So they're really banking on reaching some bigger goals, some next generation. They're not trying to make profit in the current generation models. And that's something that you as a builder or an investor should exploit

because you're getting this effectively with no margin. Yeah, that's not bad. Oh, Ryan, you go Yeah. So I guess – yeah, first of all, for anyone that's listening to this and wants to learn more, Latent Space is a wonderful blog, and it has a lot of the numbers that Sean is kind of referencing here, especially with the open AI economics, which look – they don't look the prettiest when you look at it on a chart, and one of those charts you shared, and they're certainly running at cost at the moment. My question, I guess, is you mentioned the foundation models, very costly to build from the ground up. What about – so there's the foundation models, and then you said there's everyone else. If you're in the everyone else camp, how costly is it to start? Is that purely just the, as they say, GPT wrappers, I guess, or is there more included in that everyone else? Yeah, GPT wrappers are a common way to phrase it. So I have been on the side of arguing that GPT wrappers are good, actually.

This is something that was not consensus 2022, 2023. But it turned out that basically the sort of middle tier of companies that tried to not be GPT wrappers, tried to train their own custom models to compete with open AI, those were the inflections of the world. Those were the character AIs of the world. They all folded, all of them. They all got hired by the Amazons, the Microsofts, and all those guys.

um stability remember stability ai um yeah i started latent space because of stable diffusion and that's folded like it's it's really hard to be an independent lab with like the mid-tier of money like you have to you're just going to be crowded out by by like the biggest labs um there's some exceptions for different modalities like video and voice um but yeah i mean like uh for the uh for the sort of capital light startups it's never been easier never been lighter or more efficient in capital to start with a foundation model that's either open source or provided via an api and you start from there you build your customer base and eventually if you need to you can train your own custom models to serve your customers for

high volume use cases but most people don't even need to um i've i've had i've interviewed a um someone on my podcast called codium who used to develop their own custom models they've given up and they've just decided to just wrap OpenAI and Cloud and they're a unicorn now as well and they're doing very well and they considered it one of the biggest mistakes of their startup journey so far

which is trying to train their own model because they were like, there's no point which is fantastic news for the Foundation Model Labs that's exactly what they want to hear that you're completely reliant on them so I think there's a sort of uneasy ecosystem of reliance from everyone who consumes models

and the people who make the models but yeah it's it's i i would say it's very easy now to create an ai startup um and it might be the easiest possible i would say then and i think the last thing is there's another component of this which um you might not be considering which is the amount of proprietary data you have access to right like abstractly when you start a company like it's a it's a sum of like your your people your your sort of unique resources um and your maybe your technology sort of insight or whatever So maybe you don't have like special unique insight. Maybe your people are relatively commodity too.

But if you have access to some data you can't get anywhere else, which is something that I've worked on for my company, Small AI, then people will beat their way to your door just to get at that data in the same way that Bloomberg, you know, like the Bloomberg terminal,

like sure, like it's a bundle of stuff, but really like it's the feeds. It's like the pricing. It's like the journalism that you get out of the Bloomberg terminal that makes it work so much. And then obviously the network effect from the Bloomberg chat. So yeah, proprietary data, people are actually spending more and more and more money on that because I think relatively everything else is becoming more commoditized. So then the relative profits accrue to data. okay follow-up question here the so when you mentioned that a lot of people are just

kind of calling it quits on building their own foundation model and and just leaning on open ai and i believe anthropic was the other one cloud uh it makes it sound like that's where the real moat lies i guess my question to you is with the quote-unquote gpt wrappers do you think there's we'd like to talk about moats all the time here can you build a moat as a gpt wrapper and i know that there's a lot of different use cases here but if you are is it in that data is that what you're talking about is kind of the gpt wrapper is right um yeah it's it's in that this that data whether or not you acquire it um through whatever means like um you know licensing perplexity is doing a lot of licensing of e-commerce data um and you know specific articles

opening aisles also doing that of course um or you're doing it just because you built relationships with people you're the sort of um uh like a i forget like the store like the the record the and put a sort of database of record for something important within a company um the classic one would be rippling which is sort of all your employee data um so you're um like very well based you're very well suited if you have the custom data for to produce custom ai like obviously that makes a ton of sense all right let's go through you mentioned you know open ai the foundation models and then these uh other ones that as we've called the the wrappers but But there's also the, I maybe call them the four big hyperscalers, Meta, Amazon, Microsoft, and then Alphabet slash Google.

I guess Meta really only has, you know, they're not outsourcing, you know, selling the third party cloud services. But we're going to go through some of these and we're just going to play a game here with you. Given what you're seeing as, you know, what startups are going after, like what cloud provider are they using, who their relationship is with, your opinion on who's, say, you know, ahead, succeeding or falling behind in all these new AI tools. First one, we'll just give the one that kickstarted everything here, OpenAI. What's your opinion on them as we say for context here in December 2024? Yeah, cool. Um, this is going to be controversial because I have friends working there. Um, so there's a, there's a broad question now of whether or not OpenAI has peaked.

Um, they let the, they let the wave, um, and, but you know, that, uh, they haven't shipped GPT-5, O1 has been, uh, inspiring, but not game changing according to most industry surveys. Um, so like what, you know, where are they going next? Right. Like, they're already the most valuable private startup ever, I think, unless SpaceX, you know, is secretly worth $200 billion. So, like, where to, right? Like, how far can they go as a private company, especially when they're projected to, like, I think they lost, like, $5 billion this year. They're projected to lose $40-something billion in three years.

This is a giant pile of cash that's burning. half the senior management team has left this year all of which all of whom like individually they're all like I want to spend more time

with family I want to work on other things like but together it smells it just smells right like it never it's a good sign

when most of your management team leaves let's just put it that way no matter what they say so that's it's ugly and but like

you gotta have sympathy this is one of the most storied companies of all time already It's super hard to manage. I'm sure the politics are absolutely insane. I mean, you saw this time last year

where we had the management shakeup in the eye. So that is the question that insiders are wondering. Like, has OpenAI peaked? Like, do they have tricks up their sleeves? Obviously, you talk to people who work there, they would say, nobody knows anything. We know that we have so many tricks up our sleeves, and then people outside are just waiting

for something to happen. And nothing's really moved the needle since, effectively, GPT-4, I would say, as far as public perception is concerned. Would I say anything else there? But they're still the leaders. So they set the industry standard for everything. When they release something, everyone follows. It's still that case. So I would highlight that there.

Do you have any other questions or comments in opening before we move on? No, I think that covers it fairly well. it's it's uh yeah yeah i think i think you're right individually each one you can make the case that uh oh they you know they all left for their own reasons but as a collection uh it is smells a little fishy i guess the entire the entire safety team has left which is was a big part of their ip their uh sort of proprietary leadership like basically everyone who works on safety not doesn't only work on safety you know like i think if even if you personally don't super care about safety people who work on safety work on safety because they take ai seriously

and because they take ai seriously then they are actually like frontier like it's very sort of back to front in that in that thinking so um yeah it makes it means a lot that elias oscar has left and is now working on a different model lab called safe super intelligence because he was the original champion of scaling and opening up okay makes sense i guess question for you do you think there's a first move a lasting first mover advantage for chat gpt i guess i don't know if they were the full-on first mover but it seems like they were the first ones to really truly gain notoriety is that serve as an advantage for them um it does in a few elements and it doesn't in some others right so um i'll say the ones that that where it does um opening

now owns chat.com um there is there are informal surveys of sort of consumer market share and the number like opening eyes like far and away number one like like the number two is entropic but the average person around your thanksgiving or christmas table doesn't know what entropic is doesn't care they're just like give me the chat thing right and the chat thing is over the idea they own chat.com um so yes there is a first mover advantage to building the brand same for perplexity which you also have on the list um and um the then you also have the regulatory capture advantage and the sort of data licensing advantage right once openia has has done the deal with the wall street journal or the atlantic or whoever else they they've done they've done the

deal with the people after them can't really do the deals because these are all exclusive um and that makes sense then there's the regulatory thing that they tried to pull with um the california legislation or whatever other legislation in every other city city and country in the world like they are they're always incentivized to like because they're the leaders because they're first because they're biggest first to define the law as such that it conveniently states that everyone their size is good and everyone coming up it you know has all these regulations that they it will be too prohibitively expensive for them to comply with that that is Even without malicious intent, that happens.

It just happens. Through the best intentions of everyone, this is how the world works. So enormous first mover advantage from there. And obviously from launching ChatGPT, where you have the largest collection of RLHF data sets in the world. Where it doesn't have advantages is in places where Anthropic has sought out advantage. So, for example, for coding models, Claude's sort of 3.5 Sonnet model is currently the favorite among developers. And this model did not exist six months ago. And now if you just talk to any developer on the street, they will generally prefer Claude over OpenAI.

So the moat is pretty light in terms of specific use cases outside of consumer. But I think inside a consumer, there's a very, very standard playbook where you pursue, where you can take an initial advantage and just compound, compound, compound, compound until you screw up, which Google did. Okay. So I guess it makes sense on the sort of being the verb, I guess, especially around the Thanksgiving dinner table and anyone who's not, I guess, a developer. On the developer side, you mentioned that the most recent model from Claude is kind of the preferred one for developers. Is that something that flip-flops over time? Or when a developer gives preference to a model, do they tend to stick with that? They will stick with whoever is the best performing for their use cases. So if someone new were to come out, if Amazon were to come out with a new model that blew everyone away, they would switch immediately.

So these are the chronic early adopters. They care about not just brand, but actually, are you the best in the world at doing this thing for me? And they will ignore brands. In fact, sometimes brand can work against you. Because if you're an early adopter, you want the hipster upcoming brand, the stuff that nobody knows about yet. You want to sort of be in on the new thing. I would say Cloud has reached past that now. They are the default for developers.

But yeah, there's no loyalty there. There's no moat there in terms of ongoing advantage. And I would even say that the developer tools, the Cloud wrappers that have sprung up to put Cloud into their developer workflows, like Cursor, like Codium, like maybe Cognition Devon, the people who store your company specific code base specific context actually win more than the model labs here because the model labs are maybe prohibited or are not allowed or don't prioritize

storing your context so like that accumulation of data is actually being hoovered up by the startups in between rather than the model labs itself interesting and it seems like at least for the time being, and maybe besides the optimizations that Alphabet and Meta have gotten from their advertising stuff, which is kind of a whole different story, the nearest term opportunity for providing the most value with these models is from the developer efficiency. That's the one we're seeing constantly right away where all these companies are saying, hey, look, we started using these things and we're just seeing so much improvement in just velocity and whatever metric you want to use. So it's quite interesting for Anthropic there.

I think you kind of hit on how some of these other startups besides OpenAI, the one that turned into a bit of a verb, just given they're the first one and how they're trying to succeed. I want to talk about some of the larger ones. And maybe the biggest elephant in the room is Alphabet and Google, the one that seems to be going vertically integrated here, trying to essentially copy everything that open ai is doing what's your opinion on um what they have been able to accomplish over the last couple of years and how they've tried to catch up with you know these new startups trying to encroach on their turf smartest thing google ever did was acquire deep mind um for demis azabis and then dumbest thing that google ever did was every year they continue to

employ Sundar Pichai um unfortunately like he's run out everyone's patience I think for for how he's managed this whole this whole transition um and uh what yeah like what specifically like like why like what is the frustration with Sundar Pichai that he's uh obviously I think I'm making very very broad statements and you know he's he's still like a thousand times better uh business person and technologist that i am but um i think to squander google's initial advantages uh that were set up for the for him like 10 years ago um like they invented the transformer they invented the tpu uh they invent they they have the largest data sets of basically everything in the world like youtube email search what name it they have it um vision for like freaking self-driving cars

and um like they have it they have everything and i mean don't forget maps yeah exactly right so to lose that to a startup is like you know bad um and then to to have like two of everything so you know the classic example was um google does have too many resources that has the gold that has the cash cow of ads um and then like waste a lot of it in like experiments that don't pan out that was google x for a long time then it was google brain versus deep mind then they consolidated brain and deep mind under davis isabas um and like the most recent thing that this is why i think it's nice to talk to me instead of people who actually work at google because they can't say this but i can through me they can say it uh the people who i talk to within google are not very

happy with how deep mind is still being run because there's still two of everything um and specifically there's the gcp version of deep minds offerings and then there's deep minds offerings or DeepMind's offerings. So this would be Vertex AI and AI Studio from DeepMind. So there's a lot of still, I would say, inefficiency, mismanagement. People still leave. You have Notebook LM here, which is one of the breakthrough products for Google. I called it Google's chat to BT moment this year, in a sense of it was something that was not expected it to be a big hit was a huge huge huge hit um and uh you know uh could have could have been doubled down on the creators of notebook lm have just left google like they can't retain this talent

because they can't do anything inside of google without a lot of politics politics um so it's really rough i would say like i i'm very sympathetic obviously it's it's hard to run a huge organization like that but you know like um it's it's a it's a reality that like they have to deal with it and see their way through and i really hope they do because like you know obviously they they serve it like two billion people crazy so like whatever they do hasn't been so you know they have to take that seriously what would you you know we're on the out on the outside of google so we can't say here someone say oh all right we're going to completely change our strategy we totally got this wrong what would you be looking at as someone from the outside as okay

hey, they're fixing these issues. I'm liking what I'm seeing here in regards to these AI labs and the research stuff. All right, listeners, we've got a new sponsor here on Chit Chat Stocks. The name is Blue Chippers Club. Blue Chippers Club was recently started by two friends of ours with the goal of building a tight-knit community of stock-focused investors. Inside this awesome community, everyone gets to share and break down their portfolios with each other, participate in weekly calls and compete in fun portfolio performance competitions. I really like this idea and it's why we're promoting it here on the show. In fact, Brett and I are in the community ourselves and enjoy just how much value we get collaborating with other investors.

When I first got into investing, a role model of mine recommended that I build a network of, or community of friends to bounce investing ideas off of, and Blue Chippers does just that. If you're interested in joining, head on over to bluechippersclub.com and hit apply, Right now, it is completely free to join. The link will also be in the description. All right, folks. If you're a regular listener to Chit Chat Stocks, then you know that we use FinChat.io daily.

FinChat is the complete stock research platform for fundamental investors. They have robust financial data on more than 100,000 stocks globally, including company-specific segment and KPI data. Want to see Google's cloud's quarterly revenue growth? FinChat's got it. Like to track PayPal's take rate? They've got that too. and they recently added sell side conferences and insider transactions to the platform best part is

it's completely free to sign up however if you want to get unlimited data you can use our link finchat.io slash chitchat to get 15 off any paid plans the link will be in the show notes um yeah so like uh they are paying attention so they've improved a lot since like the early days of BARD. Remember BARD, RIP BARD? They've improved a lot in the sense of paying attention to what people want out of them. NoPokeLM was a really, really good start. I would say that the frontier models as well, Gemini Experimental has consistently beat OpenAI now two months in a row, which I would say is an achievement for them. This is coming from the start of the year when we actually um did not know what the gemini strategy was at all uh now it's pretty clear

they can keep up with open ai and that's that's that's a that's a huge improvement from where they were um creating ai studio i think was a plus because vertex ai was definitely weighing google down like they're a distant fifth or sixth in in in the sort of market rankings i i that i see um so like yeah so like i mean they're doing they're fixing a lot oh then i think the last thing i'll also shout out is they're basically they rehired noam shazir who uh who was the lead author on the transformers paper um and he's building he's building his new team around him and i would say that's a unmitigated positive like whatever comes out of him will not will not see the effect of for another four to five years but it will be good just because um he's continued

to make hits like every single time he comes up with something it's it's been a huge banger um for those who don't know it's it's not just the transformer like he had this he had the same year in 2017 that albert einstein did in 1905 um in a sense every year he published every paper he published is basically now the industry standard um and then he did that again last year uh so i we expect him to do more um you know when when he lands and ramps up at google okay shifting gears a little bit i want to talk about meta they have i guess i'll let you kind of summarize it but uh what has their overall strategy been uh and from the outside looking in i would think they are one of the

going from spending a lot of money on ai development ai applications to show having that show up in revenue it seems like meta would be one of the closest like that gap would be the closest given that they can see the targeting efficiencies with their advertising revenue do you think that's true do you think they're going to see some of the highest returns on the advertising or on their ai span and then i guess maybe what are your thoughts overall on meta's ai initiatives um it clearly has done very well for the stock price i would say um reinventing themselves from a vr company to an ai company in the last one and a half years maybe two years has been fantastic

for the stock price i think it's added like a trillion and a half of value to them uh it's incredible in two years yeah quite quite a turnaround yeah on that narrative we're an company now forget the name forget the meta rebrand like we're in a company um exactly actually it's even more gaslighting it's like we've always been in ai company look at all these things we did with with uh facebook ai research and hiring yan lukun so we've always been there just like we've been in everything like they worked on a phone like someday they're going to come out and say we've always been a phone company i don't know um yeah so like uh meta's done super well um like they they are they've adopted the android strategy to open ai's iphone strategy

right like and like someone had to fill that role and probably meta was the right person to do it was the right people to do it because um this is basically the exact role in the tech ecosystem they've always played like they've never they're like out of the list that you have here they're the one company that doesn't provide cloud services um they only do consumer and they're really really damn good at it uh and in the tech world they specialize in basically open sourcing tech uh for everyone else that is proprietary to others so where um you know like maybe google would have like you know the angular platform which is very very owned by google uh meta open sources react um where uh where amazon would have dynamo db is like a kind of very proprietary

platform meta open sources like the what what eventually is now cassandra um like it's it's very, very interesting that Meta basically just always open sources the stuff that other people close source, and that is their tech strategy. Part of it is recruiting. Part of it is a somewhat unvalidated strategy that people will build on top of Meta's Lama models, and that will create an ecosystem that benefits Meta. This is what they say in earnings calls when you ask Zuck, when you ask Jan LeCun, hey, why do you spend $100 million trading Lama3? What does that do for meta one it's a rounding error that's like a day of earnings for them um two um they they will say they'll feed you some line about how like people you know are building thousands and

millions of of derivative tooling and models for llama and that benefits meta um i don't know i you know i haven't seen like a huge amount of evidence for that but like that that doesn't mean it doesn't exist i just i haven't seen it um but it is true that it builds a huge amount of branding for them it helps them hire for sure uh it makes them super relevant because everyone loves them because they give you know um nine digit models away for free and uh what's not to love like that's that's great okay we got other questions about you know startup spending um the invention of the ai engineer which you think uh invented if i'm not if i'm remembering that correctly and h100 oversupplying kind of the nvidia semiconductor stuff but i want to hit

quickly amazon and microsoft yeah where do you where do they stand uh as some of the big cloud providers and backers here yeah it's interesting it's a it's a very sort of chinese uh you know i see you have some listeners who are like i'm still interested in chinese market in chinese market this is like bat dynamic where like if you're backed by one of the b a or t then you're not backed by the others and like everyone's trying to draw sort of feudal territories around them and for a while it's kind of looked that way so amazon's kind of uh very close to anthropic like anything that anthropic releases um the amazon will also have on bedrock right um anything that open ai releases obviously microsoft will have right and it's just kind of you know every big

cloud has its has its little sock puppet uh ai lab that that it can secretly funding um i would say like that's that's not entirely through true for anthropic because anthropic is also funded by google but anyway but like that that is mostly true that is the strategy there amazon has also bought Adept, which also had the former co-authors of the Transformer paper. And now Adept is now the sort of Amazon AGI division, which is kind of interesting. And I would say the last thing about Amazon is they've always invested in their own hardware and their own silicon. And Anthropic is now a major part of their strategy for proving that out. So in the race to replace NVIDIA, every cloud is basically investing in building their own silicon. Google

is obviously ahead with the TPUs, but Amazon is probably next. And I'll probably put OpenAI on Microsoft after that. So that's Amazon. I would say like in terms of just generally cloud services wise, I think what I said last time

might stand in the sense that Amazon Cloud Services, AWS, was this leader, it was this early leader in cloud. But now because of the AI wave, they are behind Google and Microsoft just because Amazon in general is behind on AI.

Okay. So I think that touches on the list of companies we had here. We could seemingly go through tons and tons of AI companies. What's your thoughts? But I want to shift gears a little bit. Um, so it's when I, when I listen to you talk and I read your write-ups, it just sounds like so much money. And I guess you see this in NVIDIA's earnings reports is just being poured into training these models. And I'm going to read you a quote that you wrote, um, in one of your articles, it says, this is a little bit of a, I think it might've been a year ago, maybe not quite as long, But it says, it is now consensus that the CapEx on foundation model training is the

fastest depreciating asset in history, but the jury on GPU infrastructure spending is still out and the GPU rich wars are raging. Meanwhile, we know now that Frontier Labs are spending more on training plus inference than they make in revenue, raising 6.6 billion in the largest venture round of all time, while also projecting losses of 14 billion in 2026. The financial logic requires AGI to parse. I love that quote. um i guess my question is that was me being sarcastic a little bit yeah um i guess my

question is like i can't help but wonder about the roi here so much money is being spent i guess why why can this continue um like especially in the startups like excluding the ones that have existing business models like alphabet or meta if you were rationalizing this from from the startup's perspective this this level of spending what is the return what's the outcome here right okay so so we should split the startups uh do you consider opening a startup sure yeah i guess you know the largest that's the largest one okay all right so like uh you're you're you're definitely very public markets and i've been in private markets for like seven years now so uh it's very it's very different my frame of reference for me like a you know 200 million

dollar company is a big company um anyway so um i want to start so yeah like um you one you can justify on growth right like um the revenue ramp is unheard of right like um now what they like name me a company that's gone zero to two billion in revenue in two years like uh and and like release realistically they can reach 100 billion like within the decade like it's not impossible and like what's your revenue multiple on that yeah i guess that checks out what's the pre-mortem what's the what's the potential downside where yeah where does this stop where does this fantastic question yeah well so okay first of all right um i do think when i talk to finance guys you guys are very different than tech

guys because you look for the pre-mortem you look like where does this go wrong and then the tech guys and the vcs go like where does this go right and i think um it's very important in tech investing you get paid orders of magnitude more when you when things are right than when things are wrong right your downside is 1x your upside is 100x right so um i think finance guys tend to overemphasize downsides sometimes but obviously it's different when it's public markets okay rent over um pre-mortem wise uh yeah it's going to it's going to be that it's there's a lot of burning of cash and there have been ai winters before and that we have we seem to have maxed out our current direction if in terms of like whatever gpt5 was going to be and now we're sort

exploring other things and we've released two other models that were not gpt5 um and and uh maybe you know that was it for this wave and we're going to go into winter for another like five to ten years and then the next generation whatever is you know the next open ai comes and whatever money you threw into this one you know wasn't it uh and that's them's the shakes like it happens before uh you know if you weren't around in like the sort of 2014 ai wave you weren't around in the 2070 AI wave, like it's happened before and, and, you know, people famed out and they were all really smart and really well-intentioned and had all the right things and all the right PowerPoint charts.

And it just didn't work just because the tech didn't scale. And that's, that's not, that's not the fault of anyone. It just, that's the risk that you take. I would say that's the flippant answer. I pre-mortem wise. I would also say that there's a huge amount of challenge for AI specifically in terms of incumbents versus disruptors. AI is a very powerful force that helps incumbents. So in the sense that like Google has lost a huge amount of advantage to open

AI and perplexity, but it is not, it is still the default search for, you know, 5 billion people on planet earth. Like they can get it back. And the sheer amount of incumbent dominance that Google has can still win. The sheer amount of force of nature that the meta is exerting just in terms of its social networks can make it still win on its AI stuff. So it's absolutely not clear that one of these AI foundation lab startups will win here. I tend to view these questions as not that useful because to me, it's going to be clear that some mix of incumbents and startups will win. So it's not an either or, they both will win. It's just

which of them will win. Some of them will lose, some of them will win. It's not useful to discuss these categories this way. I'd rather have it the other way and go like, okay, what are the determinants of success? And some of the incumbents will have it, some of the startups will have it. Okay. Let's talk inference training and costs. So it seems like from our point of view, and again, we focus on costs where things can go wrong. What's all the spending going to turn into ROI? It seems like the biggest bottleneck is the fact that it's very, very expensive to train these things and then run them. Let's imagine a world where all these smart people just come up with all these innovations that decrease the cost of these things by

a thousand X a decade from now. How would that change everything in the industry? Do you think that's likely? Do we see that helping or hurting the cloud providers? That's kind of what my frame of reference is because it seems like all this stuff is moving so quickly. I just, I don't know, I get confused on whether like the state of NVIDIA, the state of all this spending can whether these innovations can break through and make things less costly. So there are a number of things

that I want to sort of comment on here. So you said decrease 1,000x within a decade. I just want to give an FYI that the current depreciation or the current sort of scaling or efficiency curve of AI is faster than this.

It is 10x per year. um, per year. I'm not kidding. Um, so, so a thousand X is three years, not 10. Okay. Wow. Sounds pretty nice. Um, that is the current trend line and obviously trend lines break. Um, but that is the current trend line and that is a fact. Um, okay. The other thing I think you need to also be updated on is that, um, you are still in a very sort of GPT three and four mode of having a very big mental model of high upfront training and then low low subsequent inference that is in the process of changing right now because of one um it is very clear and i mean yes only i would completely nail this when they hired no known brown um you know a year

ago um and uh this will this will be the story of the next decade is figuring this out which is the inference time scaling is now a thing as opposed to training like some pre-training scaling, which means that the cost of training and inference is going to rebalance such that inference will be higher. That generally is a good thing because you can charge very good markups on inference as opposed to charging amortized cost on training. Inference cost is directly attributable in a way that training is not. And the performance is fantastic as well. So the rebalance, basically, like exactly when the big governments of the world, including the US, decided that 10 to the power 25 flops for a model was too big to let you responsibly train yourself. You have to notify us when you're doing a large 10E26 run.

That's when we figured, that's when we stopped pre-training anyway. Like we were going to stop anyway. So like we let the bureaucrats in Washington think they're doing something. But actually now all the focus is moving to inference, which is really, really funny. Like you're now focusing on the thing that no longer matters, which is kind of interesting. That all aside, I would say combining this training inference shift and the overall cost reduction, I don't think it benefits or hurts NVIDIA either way. I think as long as they continue to deliver hardware improvements on the pace and scale that they have, it's good. the problem with nvidia now is expectations are so high that they have a new generation every

single year and all it takes is just one little misstep and they're they're going to be knocked way down um just just because hardware is hard and mistakes happen and you know look when look we had some delays in apple what what you know people were like proclaiming the death of apple um so it's going to happen to nvidia and you know that would be the perfect time to buy nvidia to be honest. And I mean, who does that benefit? I mean, it benefits consumers. I think you also need to think about a spectrum of intelligence, right? So the large labs are going to be focused on the closest to AGI that we can get. For people who don't know, there are five levels to AGI that OpenAI has defined. We're currently at level two. They consider us to be reaching level three

soon. So you can read up more about that. The rest of us have progressively dumber levels of intelligence that we can use on progressively dumber and cheaper machines right so um like this year on my phone i have apple intelligence um doing like simple tasks for me like summarizing my messages and you know i can i can sort of ask it for for visual search of my videos and photos this year as well by the end of this year chrome is shipping gemini nano inside of chrome so you can sort of query models without um without without calling a server so like it all runs on your machine so those those ai models are free like they run on your device like there is no cost to them therefore like basically nobody makes money in them um they're this is just google and

apple um trying to serve their customers well and trying to say like you don't need to call open ai for summarizing your emails like you can just call your local model it's fine right so um it benefits the consumer uh it benefits incumbents who can bundle stuff for free where other people would have to charge separately for them like like an anthropic is never going to get on your operating system like unless they build an operating system which they're not going to um so they're always going to be third party they're always going to uh to suffer relative to the incumbents okay the last time we spoke you said the big this was mostly pertaining to the hyperscalers but you said the big get bigger in tech and it's kind of a broad classification

um for i'm trying to remember what my question was but i think it was basically like what are the embedded advantages of hyperscalers and you mentioned they get bigger in tech do you think this now we're seeing like you said companies go from zero to two billion in revenue in two years do you think this rise and i guess recent proliferation of ai development changes that in any way or do or i guess amplifies it ah does it amplify that if does does it amplify the hyperscaler effect sure economies of scale the big get bigger yeah yeah yeah um they it should but currently it hasn't which is cool um so why is that so the the hyperscale the hyperscalers should benefit

from um uh from having that scale having that resources having the data centers the the customer relationships what have you um not even talk about data but they're also yes also to data they should have all the advantages in the world and the only reason they don't is basically this economy is a scale right like the the problems with managing so many people and having so many egos and so many divisions and so many conflicting priorities that you can't really figure out which one to focus on. Whereas OpenAI had one goal and they did really well at it.

So yeah, there are diseconomies of scale. They do exist. And I will say that so far the evidence has been that hyperscalers have not been able to benefit from their scale here in AI. I don't know how true that is going forward. You're always subject to change, always subject to change in management, to be honest.

Like, like literally put, promote Demis Hassabis to CEO of Google or like, you know, get someone else in there. Like, you know, the Google story might change very significantly if you can use a different, if you can use the existing assets really well. But it does take, I mean, superhuman levels of organizational management to, to organize a hundred thousand people to, to like pivot. This episode is brought to you by our friends at Yellow Brick Investing. Yellowbrick is an aggregator of the best stock pitches across the internet. By tracking thousands of blogs, newsletters, fund letters, podcasts, and more, they collect and summarize the best stock pitches and bring them to you in a single place. If you're a regular listener, you know that we use Yellowbrick every single week here on the podcast to discover new investments or just find reports on companies we've already heard of. Try it for yourself.

Simply go to joinyellowbrick.com slash chitchat and search a company or ticker you are interested in. you are bound to find a great report on just about any company that's join yellowbrick.com slash chit chat heads up folks interest rates are falling but you can still lock in a six percent or higher yield with a diversified portfolio of high yield and investment grade corporate bonds on public.com you might want to act fast because your yield isn't locked in until the time of purchase. Lock in a 6% or higher yield with a bond account only at public.com forward slash chitchat stocks. So we've been jumping all over a bit on at least this topic, but one thing that you focus on a lot in your writing discussions and stuff over the last few years is what you've

called the rise of the AI engineer. Can you go through that? What makes it different than other maybe adjacent engineers um just general software machine learning or stuff or data sciences and why are they going to be important over the next decade and beyond interesting okay um i typically talk to i appreciate this or phase this to a technical audience so this will be interesting talking to public investors okay so you know how snowflake ipo and everyone's like what the hell is snowflake you know how that was the thing i that was me that was me that was us that is happening to ai right now so snowflake was the darling of data engineers and data engineers were a thing basically because uh facebook had crapped on the data and they were

like we need people who specialize in moving data around and making sense of data they invented a data engineer it proliferated out of facebook to everyone else uh it became a whole industry snowflake became data warehouse and and was the one of the biggest ips of all time okay so that thing is happening now to AI. We have seen this play, we've seen this story play out many times in my career. So I've seen this for DevOps, data, front-end, and let's call it sort of

serverless or cloud engineering. So the same thing is happening to AI in the sense that there's a special sub-field of software engineering that is specializing in building with LLMs, and they will define the stack.

Whatever new tools that they like will become the preferred stack and the preferred way to build AIs for everyone else. And I saw this early and called it and promoted it as a trend. And now Gartner thinks it's peak hype. But I do think it's basically the sort of defining engineering trend

for the next 10 years, at least. There will be some day that it's over. When we hit AGI, whenever that AGI is, then AI will do everything for us. They will never have to work a day in our lives. But until that day, I often say that AI engineer will be the last job. Like if you're scared about AI taking away jobs, then you should think about the mechanics of how jobs are taken away. It's by someone sitting down and encoding what it is you do for work into an AI.

And that is the job of the AI engineer. So ultimately, it is the job to take away the other jobs. So if you want to build AI lawyer, you're going to build AI customer support rep, you're going to build AI financial analyst, you're going to hire an AI engineer to study what it is. you need done to put together the systems for you to do it and to do it reliably and to scale it up to hundreds of thousands of AIs working for you. So that will be the job. You talked about something that I thought was extremely interesting. Or maybe it's not a new idea, but it was at least to me. And that's the distinction between

or just the idea of autonomous applications, autonomous agents, and why they are the most valuable. It made sense for me as a generalist, just the fact that you said, hey, look, it can save anyone a ton of time. So what do you mean by autonomous agents? And why do you think they can be so helpful for companies, people, just anything? Well, I mean, wow, it's kind of like hard to define because I live in a world where this is assumed knowledge. So it's really interesting to step out of it for a sec. Autonomous agents, I would just kind of think about it to the common person as having an assistant, like a secretary or a virtual assistant or executive assistant. If you ever worked with one of those, particularly if they're remote,

if you've never seen their face, they work in the Philippines or something, you just email them and stuff happens, right? That's an agent. Right now, that role is performed by a human, just like humans used to perform the computer role in the 1950s for nasa like right now the humans are performing the virtual assistant role because they live in the philippines and they're 10x cheaper than than people in the u.s but that will go to ais as well um that is guaranteed um it is already partially going right like we're the problem right now is we live in this liminal stage where not everything works well yet so we have to like scope it down to the to the subset the things that work well and don't use ai for the stuff that doesn't work well but like you know i

always say like you don't even need ai to have a good agent like my perfect agent for me is my scheduling link that you use to book me for this call um it's a it's a it's a calendar link right like i sent you a link you booked yourself you could have rescheduled you could have canceled i could have canceled i could reschedule and like we didn't have to talk to each other we just like dealt with this link um and that's great like it manages my calendar for me um i have a bunch of these out to different people that need to work with me asynchronously. And that's great. So I think that is the promise of the agent. And I think that is one agent working for you. And I think the scale here is, the promise here is defining one sort of AI employee. This is what

Jensen Huang loves to call it. He has this vision of AI employees for NVIDIA as well. Defining that and then scaling that, it's just copy paste to have the next hundred and the next thousand working for you. And this is way better than hiring humans because they don't need holidays. They don't need silly things like human rights. And they work exactly the same. So there's no training time too. Like once you train one, you've trained all of them. So there's just like an enormous amount of advantage to having these things. And then finally, I think on the humanity level, like we're basically topping out in human population, like based on projected growth rates, we're going to top out at 10 billion people.

I think this is like really nice that we're figuring out how to scale intelligences just as human intelligences top out, right? So ultimately we're going to need tens and hundreds of AIs working for each individual human for us to keep scaling the way that we are scaling. And I think that's beautiful.

Yeah, that is a, that does sound quite nice. And even moving beyond the virtual agents, which I've seen some ideas from, I think it was an investor talking about this, about the future where there's millions upon millions, maybe billions upon billions of virtual agents talking to each other online and figuring things out. But what about, and I know you're more in the startup land, so you see some of this stuff. What about that in robotics? We've seen a lot of ideas and you can kind of connect the two together. Well, we put an autonomous or AI agent or something that's been trained, an AI employee within a robot,

and they can replace some jobs that aren't that great for some people and provide a lot of value for society. Are you seeing that at all? Or is that something that is maybe way far out into the future? No, no, it's here. They're walking around. I mean, like, you know, half my rides are Waymos, right? They're all robot drivers.

It's really like if you have the means and you haven't driven in a driverless car yet, like get get on that like they are in teslas they're in waymos um this is this is coming it's going to reshape our cities there'll be no more parking it's great okay um the yeah so they are coming physical intelligences are a thing um and there's always the question of why humanoid forms because like you know it's just a quirk of nature that we that we have two arms and two legs um but the the simple answer is that we built the world to fit us therefore now we must built machines to fit the world that we that we built for ourselves so they probably should take

humanoid forms um and yeah like they can take jobs that are downright dangerous for people that's fantastic um and also you know potentially if uh if we if we don't watch where we're going they can also fight for us which is terrifying um but yeah it is absolutely uh the case that they should take warehouse jobs they should uh they should take the the hard menial labor so that we can do the higher value add stuff so um i think it's i think it's coming um did you see do you guys see the um videos of the optimus robots shopping for kim kardashian i did not no no they're walking around out there um the problem with optimus and the problem with elon in general is that you don't know how much of it is real um so a lot of these like are tele-operated by

humans like sitting behind a camera somewhere but honestly even that is fine like even even that is training data even that will be will be learned by a model someday and and then you can take the human out of it um and and by the way teleoperation is still cheaper than having a human there physically in person so like it's fine these are just cars that walk uh you know like we and we're comfortable self-driving cars like we're you know these are these are just slightly different cars so um it's going to happen i would say obviously like put this one or two generations behind how production ready it is for for the for the everyday consumer like it's so much easier to move things in the world of bits than in the world of atoms um so like let's solve the bits stuff

first like the software first then then we can move on to the hardware because the software clearly isn't done yet yeah i saw or you had some analogy within one of your pieces where you said this current iteration or kind of the new products out there the stuff that a lot of startups are working on all these new innovations seem very similar to what self-driving was maybe 2015 through 2017 and i remember back then everyone was saying oh you know it's right around the corner and technically it was you know it just took kind of eight to ten years to get there uh to where it started proliferating throughout society but it might take maybe that long for some of these new things as well which can seem like a long time but in the grand

scheme of things it really isn't it's nothing like it's it's like it's like we had we took like 30 years to roll out the car like we had the we had you know the model t and it took a long time to like you know roll that out all right i think we've covered a lot of the questions we had expected to ask beforehand i guess one question that i'm thinking of now we do we do have primarily public market listeners that listen to the show on a regular basis it sounds like a lot of the development is happening in the private markets at the startup level um because people uh we've seen the exodus of uh executives once they go into big tech companies, they don't like the politics, they want to be able to operate in a smaller company.

For the public market investor, instead of where can they get the most exposure to AI, what industries in, I guess, the medium term, we can call it, I don't know, maybe next five years, will be the most adversely impacted by the advancements in AI? um high margin low mps software businesses okay so the legacy i guess software businesses like uh is that like an oracle is that what you would yeah yeah because like the the cost of making new software the cost of porting software over has just vanished um overnight like it's a lot cheaper to write software now um if you If you haven't tried it, I feel like I need to throw out things.

There's a lot of theoretical talk in this podcast. Go right in the way mode, like the job one. Job two, go to bolt.new and type in make a Spotify clone, make an Airbnb clone, whatever. You'll be shocked at how far along I'm doing these things. Bolt.new is the one that I happen to have just interviewed on my podcast, so it's top of mind. um and and uh yes these are these are sort of toy examples but in the real engineering world in in

with like cursor windsurf uh cognition all these things and not so much condition but you know that's a different story but like they are all real examples of the cost of software going a lot lower right so if you listen to the all-in podcast like chamath is working on 80 90 which is like this this like grand vision of like rebuilding all legacy things uh with ai and i i think like that is directionally correct i'm not sure that chamath is the guy to do it but directionally someone is going to crack it someone is going to crack it like he he's right on the thesis i don't know if he's the guy but whatever it doesn't matter like someone's going to do it so like like the hunt for service now is is on the hunt for oracle is on i would say oracle is a little

bit. Anything databases, databases are very hard tech and take a lot of time to mature. So maybe not specifically the databases, but the things built on top of databases. The reason that Salesforce is so threatened by AI that they are, you know, Mark Benioff's doing his agent force thing is because Salesforce knows that it is like a prime target for new Salesforce, whatever new Salesforce is. Interesting. All right. Well, I have bolt.new loaded in my browser right now. I'm to try it out after this episode we appreciate you taking the time here sean you have quite a few say for anyone that wants to go even more advanced and all this stuff i mean you have so much over at latent space so why don't you tell the listeners i think you'll find your work

what you're doing and that uh the conference you did and maybe you're doing yeah so so the conference is not really for this audience conference is more for engineers uh learning the latest and greatest in engineering stuff it will have zero business uh sort of angle to it but maybe you like that i don't know um latent space is more where i let the sort of venn diagrams of business and and tech uh happen and i i basically do what a research firm would do if i were reporting on these as public companies folks um so you know i kind of perform the analyst role which is kind of fun for me like because i used to do that for my hedge fund um and yeah that's that's where you can check out all the all the the interviews and the essays that we have on

on tracking the market and then i think the last piece that we have right now is small ai news so small.ai news is the newsletter that is on that gives you a daily pulse um so these are this is a completely uh ai generated newsletter or like a 99 ai generated newsletter there's a bit of human curation um that surveys all twitter already all discord to keep you super up to date so this is the one that like the andre karpathis of the world uh super chintales of the world read to keep up with AI if you want to get that level of granularity. So basically, I have the six-month granularity. That's the conference. Then I have the weekly granularity, which is latent space. And then I have the daily granularity, which is AI news. So plug in wherever you want.

Beautiful. All right. Well, that's going to do it for this episode. Thank you again for joining. Let's hit the disclosure. We are not financial advisors. Anything we say on the show is not formal advice or recommendation. Ryan, I, or any podcast guests may hold securities discussed in this podcast, may have held them in the past and may buy, sell, or hold them in the future. Thank you everyone for tuning into this wonderful interview. Thank you, Sean, for taking the time to join us and we'll see you next time.