Tae Kim Sounds Off, Big Companies Are Hiring Again, NVIDIA $50B Tenant | Tae Kim, Ben Zweig, Aakash Thumaty

(01:17) - Big Companies Are Hiring Again (09:58) - 𝕏 Timeline Reactions (20:19) - NVIDIA $50B Tenant (23:33) - The Three Anthropic Proposals (35:00) - Tae Kim discusses negative sentiment around AI and semiconductor stocks, arguing that sensationalized headlines obscure strong underlying demand for compute, memory, and data-center infrastructure. He remains bullish on Nvidia and the broader AI trade, predicting that agentic AI, recursive self-improvement, and growing enterprise adoption will drive substantial investment and revenue growth. (01:06:32) - Kansas Town Splits Over Nuclear Reactor (01:10:58) - Ben Zweig, a labor economist and founder of workforce data company Revelio Labs, discusses how AI is reshaping hiring, employment, and workplace tasks. He explains that AI adoption generally correlates with company growth, while creative freelancing and task-based work face greater disruption, and AI-generated job applications are making it harder for employers to identify qualified candidates. (01:29:22) - Aakash Thumaty, founder and CEO of Takeoff, discusses the company’s rapid growth from near-zero to almost eight figures in revenue and its acquisition by Sierra. He explains Takeoff’s autonomous, revenue-generating AI agents, outcome-based pricing model, deep customer integrations, and how its technology is evolving into Sierra’s new Horizon product. (01:45:30) - Zuckerberg Backs AI For All TBPN is made possible by: Ramp - https://ramp.com Public - https://public.com Cisco - https://www.cisco.com Console - https://www.console.com CrowdStrike - https://www.crowdstrike.com Figma - https://www.figma.com MongoDB - https://www.mongodb.com NYSE - https://www.nyse.com Railway - https://railway.com Shopify - https://www.shopify.com Codex - http://openAI.com/codex Follow TBPN: https://TBPN.com https://x.com/tbpn https://open.spotify.com/show/2L6WMqY3GUPCGBD0dX6p00?si=674252d53acf4231 https://podcasts.apple.com/us/podcast/tbpn/id1772360235 https://www.youtube.com/@TBPNLive

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00John Coogan?

  You're watching TVB on Tuesday, July 20, 2026. We are live from the TVB Ultra Dome, the temple of technology, the fortress of finance, the capital of capital. We've been having a lot of fun with Suno. Hope you have been enjoying it too. I'm sure we'll have a new one available soon. But first, let me tell you about ramp.com. Time is money. Save both. Easy use corporate cards, bill pay accounting, and a whole lot more all in one place. Bunch of news today. Jordy's laughing, laughing, laughing.

• 0:35Jordi Hays?

  All right, I think we get.

• 0:36John Coogan?

  Yeah, nothing like a couple of pints of Guinness.

• 0:41Jordi Hays?

  Kodex and Guinness.

• 0:42John Coogan?

  Prompt engineering, some vibe coding going on. Well, Anthropic's responded. We're going to go through that proposal, the the facts and the proposal for what next, what happens next in the open, open model debate over whether or not they should be banned, restricted, tested, limited in some ways, sued. Oh, there's a whole bunch of different possible outcomes. But we'll take you all through it. And we have Take Him joining from First Adopter at 11:30. But first, we are going to talk about the hiring market because the Wall Street Journal has a very interesting report that large, the Wall Street Journal is reporting that large companies are beginning to.

• 1:22Jordi Hays?

  They have a large white pill.

• 1:24John Coogan?

  Yes, it is a large white pill.

• 1:25Jordi Hays?

  Has hit the front page of the Journal.

• 1:27John Coogan?

  Yes, and I think people have been going back and forth on this. This is a story that's that's just getting digested by the tech folks, like the actual AI lab leaders who had predicted crazy job losses and are now not really seeing that. They're seeing productivity boosts and different diffusion taking time in certain places. And there's new capabilities, but it's not exactly a drop in replacement for a coworker, at least in in most scenarios. And that's what the Wall Street Journal is reporting. So let me set the table, and then we can debate it a little bit. First, I'm going to tell you about console. Console built AI agents that automate 70% of IT, HR, and finance support, giving employees

• 2:09John Coogan?

  instant resolution for access requests and password resets. So after roughly a year of cautious hiring, companies across technology, transportation, defense, and other industries now say they need more employees to work alongside AI systems. Total victory for both humans and AI. We're working together. Peace is possible. It's an example of Jevon's paradox, Jevon's paradox. When a technology makes something more efficient, demand often rises enough, the total use and the need for people actually increases. For roughly the past year, many companies pointed to AI while announcing layoffs. This was a huge thorn in your side. I think you hated this more than anyone else. And you were right

• 2:52John Coogan?

  to, because it did seem like it was just PR spin, etc.

• 2:56Jordi Hays?

  Yeah, it was a way for CEOs and management teams to save their own ass instead of saying, you know, hey, we over hired, or the business isn't doing as well as we would like. And we need to sort of basically settle down for a second and get our mojo back.

• 3:15John Coogan?

  Yeah.

• 3:16Jordi Hays?

  Obviously, no one wants to say that. But I think one of my favorite posts was the, and obviously, these circumstances are never great, but the new CEO of Xbox came out and just was like, very honest about the situation.

• 3:29John Coogan?

  Yep.

• 3:30Jordi Hays?

  And I think that more of that is necessary.

• 3:34John Coogan?

  Yeah. Also, there's a lot of firms where they, once they get to 10,000, 20,000 employees, they might say, look, 20,000 might be the right number. But the bottom 1,000 people are not performing. We would like to lay them off and then bring in a new 1,000 people that are better fit for the company and the current trajectory that we're on, the current skills that we need. Maybe we need more salespeople.

• 3:57Jordi Hays?

  And those bottom 1,000 people might be top 10% at another company.

• 4:02John Coogan?

  Exactly. Yeah. So the narrative appears to be shifting. Companies like CSX, Alphabet, ServiceNow, Snap-on, and consulting giant Booz Allen Hamilton have all recently signaled plans to expand hiring, particularly in areas where employees can use AI to become more productive. We have Ben Zweig from Revelio Labs coming on at 12:10 to talk about the difference in the AI driven hiring market, some very interesting data about how AI enabled firms are hiring faster than those that aren't adopting AI. But at the same time, there's a bunch of weird dynamics in the labor market where there's way more job postings than actual hirings. And so that can look like there's a falloff, and it's harder to get

• 4:44John Coogan?

  a job. But that might just be because everyone's slopping it up in the job postings. Everyone's like, put up a job posting for everything. Because I'd love if somebody, if some insane sales guy walks in the door, we might have a position. So let's keep it.

• 4:57Jordi Hays?

  Yeah, it used to be somewhat of a flex if a company was like, yeah, we put up a role and we got 2,000 applicants.

• 5:03John Coogan?

  Yeah.

• 5:04Jordi Hays?

  It's like, well.

• 5:05John Coogan?

  That's true. And then also, it's a little bit of a sign of like, oh, wow, they have 100 openings. Like, they must be like, growing so fast, you know. So but if it's just a prompt to say, oh, yeah, put up, like, look at my organizational design and and add five roles for everyone, because why not? Why not see who comes by? You know, we're not, we don't necessarily have to interview these people. So weird, weird dynamics, but we'll dig into it. So meanwhile, the latest weekly US jobless claims fell to one of the lowest levels in decades, underscoring the resilience of the labor market. The shift also reflects a more realistic understanding of AI's capabilities. Sarah Franklin, CEO of HR platform Lattice, says many companies initially assumed AI agents could replace entry level workers,

• 5:48John Coogan?

  but are now recognizing that human, human employees remain essential. Just because you have coding agents, doesn't mean you're not hiring engineers, she said, adding that Lattice is seeing renewed hiring among many of its customers, including for junior roles. Robert Half, CEO, M. Keith Waddell, said AI's effect on employment has been more benign than some had feared, adding that hiring demand continues to improve and market conditions are increasingly more supportive of business. And so I do think there was a little bit of like, a successful psyop with the, with the AI is going to be able to do everything, where I do think there are some firms that were like, yeah, maybe we shouldn't hire. Or because like, what if we get it wrong? And we hire a bunch of people, and then AI

• 6:31John Coogan?

  really does catch up, and we don't need those people. That's silly. We shouldn't go through that like, whipsaw effect. And so people are going back and forth on that. Bryce Roberts.

• 6:41Jordi Hays?

  Yeah, it's interesting. At least in, at least in our organization, which is unique and and very niche, and there's not that many organizations that are running, you know, a niche technology daily show.

• 6:55John Coogan?

  Yeah.

• 6:56Jordi Hays?

  I feel like a lot of what the the value that we get out of AI would historically been done by not super expert level freelancers, right? These sort of like, upwork style tasks that you would do historically, like, an idea for a funny song, right?

• 7:13John Coogan?

  Yeah.

• 7:14Jordi Hays?

  I've paid to get a funny song made probably a decade ago online, right? As just like a joke. And now you can just go to Suno and make something like that. Whereas, and then there's other things like, you know, make a funny website, right? I historically would maybe work with.

• 7:31John Coogan?

  So you're saying that I should, I should take down the five open roles I have for Celtic punk session musicians?

• 7:40Jordi Hays?

  Not yet.

• 7:41John Coogan?

  Because I was going to hire five Celtic punk session musicians to constantly record Dropkick Murphy's covers for us.

• 7:48Jordi Hays?

  Yes.

• 7:49John Coogan?

  Every day.

• 7:50Jordi Hays?

  Well, I'm not ready to say that you shouldn't do that.

• 7:52John Coogan?

  I'm actually closer than ever to hiring a full time Celtic punk band to to play music, to recreate songs.

• 8:00Jordi Hays?

  Kodex and Guinness.

• 8:01John Coogan?

  Yes. I'm closer than ever to doing that, where that was not even on the roadmap a few years ago. Yeah, I don't know. It's a good point. Yeah, there's a lot of things that you are doing that you would never do with a full time employee.

• 8:14Jordi Hays?

  Yeah.

• 8:15John Coogan?

  That just sort of like fills the cracks and allows you to do more different things in your organization. But the core stuff is still like, you want a person that's responsible, and then you want them using AI. I don't know.

• 8:25Jordi Hays?

  Yeah.

• 8:26John Coogan?

  The Wall Street Journal breaks it all down, but we went through most of that. So Bryce Roberts, he's, he's taking the other side of this. He says, he shares a screenshot of a text message, says, we honestly aren't hiring a ton right now. AI backfilling most roles. Backfilling, is that specifically, does that specifically refer to when someone leaves the company? You backfill them with AI? So you say, oh, someone quit, let's see if, like, if there's Steve and Jim on two different, on one team, and Steve quits, you say, hey, Jim, can you just, instead of hiring another person, just do twice as much work with AI? Is that what this person's articulating? I mean, obviously, there's some companies that are like, yeah, we're not hiring anyone. We're going for the one person, one billion dollar company. Like, I'm not going to hire anyone. I'm just going to use.

• 9:10Jordi Hays?

  Yeah, but that's rare. Usually, usually when your business is ripping, you're like, I can't hire great people fast enough.

• 9:14John Coogan?

  Yeah.

• 9:15Jordi Hays?

  And sometimes you actually, sometimes you actually don't have time to invest into various hiring processes. But yeah, I would read into this text, the company's just probably not like ripping.

• 9:27John Coogan?

  Hmm.

• 9:28Jordi Hays?

  That's my, that's my takeaway.

• 9:29John Coogan?

  Well, Bryce Roberts says, RIP new grads. Matthew Prince over Cloudflare takes the other side. He says, wrong strategy to stop hiring new grads. The right strategy, hire them and insert them into legacy teams to help them better adopt AI.

• 9:42Jordi Hays?

  And Cloudflare, of course, hired 1,000 interns.

• 9:46John Coogan?

  It wasn't, it was a crazy number, wasn't it? Up there in like, almost 1,000?

• 9:50Jordi Hays?

  Four digits.

• 9:51John Coogan?

  That's crazy. Anyway, let me tell you about the New York Stock Exchange. Want to change the world? Raise capital at the New York Stock Exchange. Pulling a crazy rare business card. I haven't seen this. I, oh, I think I know where they're coming with this. But let's play the latest good work reel. We're watching reels now.

• 10:10Jordi Hays?

  This is what we got, is Bernie Madoff. Pretty good.

• 10:16John Coogan?

  That's from the 80s, too. That's good.

• 10:19Jordi Hays?

  Yeah, yeah, I've seen a few of these around before. Of next.

• 10:24Jordi Hays?

  Solid Sam Bankman-Fried here. That's nice. That's really nice.

• 10:29John Coogan?

  I like that they actually printed these.

• 10:32Jordi Hays?

  I think he made it with balsa wood. This is balsa wood?

• 10:36John Coogan?

  The acting is so good.

• 10:40Jordi Hays?

  Wow. This is a, this is a vintage Zuckerberg. Yeah, oh five? This is a vintage oh five Zuckerberg. Let's just check the back really quick. Yeah, there it is. That is a patch from his Fruit of the Loom boxer briefs. You can tell by the smell.

• 10:56John Coogan?

  Is that real? What is that referring to?

• 10:58Jordi Hays?

  This is on athlete, you know, trading cards. I'll put a piece of the jersey.

• 11:03John Coogan?

  Piece of jersey.

• 11:04Jordi Hays?

  Yeah, yeah. All right, up next.

• 11:06John Coogan?

  Leave it.

• 11:06Jordi Hays?

  Ooh, okay. Elizabeth Holmes. We do have two, I believe. We have two, but a triple Holmes is what every good collector has in their arsenal. All right.

• 11:18John Coogan?

  Last card?

• 11:19Jordi Hays?

  One more card left. Three, two, one. Oh my God. Oh my God. Oh my God. Oh my God. Turn it off. Very funny. Very funny.

• 11:29John Coogan?

  What, it is funny how the, the like business comedy canon has really solidified around like Elizabeth Holmes, Sam Bankman-Fried, Mark Zuckerberg. There's like a few names.

• 11:40Jordi Hays?

  I'm surprised they didn't have an Adam Newman rookie card in there.

• 11:43John Coogan?

  I don't know if Adam Newman is like a big enough name relative to

• 11:47Jordi Hays?

  That's true.

• 11:48John Coogan?

  Sam Bankman-Fried and Elizabeth Holmes. It's just interesting, like the different, the different names that have broken out that you can do a comedy sketch that's like, you know, it goes as big as, as good work does because they get, you know, I think millions and millions of views in there.

• 12:00Jordi Hays?

  All right, pull up this image from Manhattan this morning. We got sent this.

• 12:06John Coogan?

  We've been doing on the ground reporting.

• 12:07Jordi Hays?

  From one of our on the ground reporters in Manhattan. There's a company called Black Sheep that, that got 20 trucks and they're just driving them around Google's Manhattan office.

• 12:20John Coogan?

  Yes.

• 12:20Jordi Hays?

  Saying shame on you, Google. Return our $80,000. We had to dig in. We got very curious.

• 12:27John Coogan?

  I had no idea. They make, they make sunglasses?

• 12:33Jordi Hays?

  They make $8 sunglasses that beat $350 sunglasses in NBC lab test.

• 12:40John Coogan?

  Okay. Are you wearing Black Sheep today?

• 12:42Jordi Hays?

  I wish. I wish. So, uh, Black Sheep makes direct to factory.

• 12:48John Coogan?

  Okay, factory direct prescription eyewear. Stop paying $500 for five dollars.

• 12:52Jordi Hays?

  This is from their own website. They're saying direct to factory optical disruptor, which is, which is not, this is from their website.

• 13:00John Coogan?

  Where are you getting this?

• 13:01Jordi Hays?

  Uh, on Black Sheep.

• 13:03John Coogan?

  I'm on blacksheep.io as well. It says factory direct, direct to factory. Look, I want to send some eyewear to a factory.

• 13:11Jordi Hays?

  Direct to factory.

• 13:12John Coogan?

  I'll be sending it to them.

• 13:13Jordi Hays?

  Direct to factory.

• 13:15John Coogan?

  Yeah.

• 13:15Jordi Hays?

  Uh, so this company

• 13:17John Coogan?

  That is interesting.

• 13:18Jordi Hays?

  Is fascinating. They say direct to factory optical disruptor. Black Sheep launches 25 truck guerrilla campaign against Google in Manhattan.

• 13:28John Coogan?

  Okay.

• 13:28Jordi Hays?

  And then they're, they're sort of like narrating their own guerrilla campaign. A fleet of 25 minimalist LED billboard trucks surrounds Google's Chelsea headquarters after the tech giant weaponized an organic search glitch to pocket nearly $80,000 in ad spend following Black Sheep's viral NBC Today Show debut. 25 LED trucks deployed, $77,000 drained in 30 hours. And then they're, and then they're just continuing to market their own products. So very interesting strategy here. I think every marketer has had the experience of, of having a campaign go haywire.

• 14:08John Coogan?

  Yeah.

• 14:08Jordi Hays?

  Very fascinating to take, to take this route. Let's see how it works for them. Uh, if I were, if I were Google, I would say you can have your $80,000 back, but you can never advertise on Google again. Because I just don't know how.

• 14:27John Coogan?

  I don't think Google would ban them permanently for this. This is ridiculous, but it, you know, there's, there's going to be like any other, like it's a self-serve platform.

• 14:36Jordi Hays?

  But is it a good campaign?

• 14:37John Coogan?

  But what, what actually happened? So they say, how it unfolded, NBC Today Show segment errors. Uh, they test the retail subscription against Black Sheep's factory direct pair. National search traffic spikes, hundreds of Americans search Black Sheep because they're seeing it on TV. The organic listing breaks. Google's search engine redirected organic brand traffic to a dead end third-party 404 error page. And so with the organic route broken, users were funneled into Google's paid listings. So what is their claim? How is Google responsible for this exactly?

• 15:15Jordi Hays?

  Sounds like user error.

• 15:17John Coogan?

  Because I mean, you do, you do have some control over your Google search results based on the webmaster tools. You can index certain things. And then also if you're noticing a 404 page, you could like redirect it quickly. But again, if this is happening all very fast.

• 15:31Jordi Hays?

  They can user error.

• 15:32John Coogan?

  Uh, but I mean, it is interesting because they're probably going to get more than $77,000 worth of organic just from this. I mean, I didn't see the original campaign and I'm seeing this because this is layers. But this is like, is this, they, they shared an AI image with tons of these like shame on you trucks, but those are real.

• 15:52Jordi Hays?

  These are real.

• 15:53John Coogan?

  Yes. And are those minimalist or maximalist? Those seem maximalist to me, but maybe they're minimal.

• 15:59Jordi Hays?

  Minimalist, I guess, in the, in the display of the, uh, in the way they actually are leveraging the space on the truck, black and white.

• 16:07John Coogan?

  Truly underrated surface area for stunts and advertising. Like, this message is sort of like squabbling with Google over this like sort of odd scenario, but you can imagine someone using this for something much cooler and much more positive and not like this, uh, you know, sort of unfortunate situation for them where they're dealing with the, you know, fallout of a Google error.

• 16:29Jordi Hays?

  Well, we want to interview the truck drivers. So if you're driving a Black Sheep truck around Manhattan today, reach out.

• 16:36John Coogan?

  For sure.

• 16:37Jordi Hays?

  Show. Nick, make it happen.

• 16:40John Coogan?

  Well, let me tell you about Shopify. Shopify is the commerce platform that grows with your business and lets you sell in seconds, online, in store, on mobile, on social, on marketplaces, and now with AI agents. Uh, Ilya said straight shot to SSI, so they better not be gearing up to release a work agent called Francois.

• 16:58John Coogan?

  Francois would be a very good name for a, for an AI agent. I like that. Uh, I do wonder what they're going to be releasing. Has, has the SSI is going to release? Is that complete rumor? Because all, all Ilya said.

• 17:12Jordi Hays?

  They would scale their research.

• 17:13John Coogan?

  Scale their research. So that just means they've done a bunch of research. They have some sort of architecture that they like, some sort of flywheel, and they're going to like use more compute. And so that's why they're raising money. I don't think they said like, and we're going to release it publicly.

• 17:26Jordi Hays?

  Yeah.

• 17:27John Coogan?

  But everyone's thinking like probably still LLM or something different. No one really knows.

• 17:33Jordi Hays?

  Yeah. I mean, I think still broadly, like generally LLM based.

• 17:35John Coogan?

  God, like next level. There are levels to vague posting when you live a vague life. It's just like your entire life is vagary. Anyway, uh, in other news, recursive superintelligence signs a $410 compute deal with Amazon. So funny. And it's in the TechCrunch, it's in the, it's in the, the header too. Of course, that is a typo. Uh, it says recursive superintelligence signs $410 million, million dollar compute deal with Amazon. Congratulations to recursive superintelligence.

• 18:11John Coogan?

  Throwing safety out the window. That should be the, that should be the, the tagline because there's already safe superintelligence, but we're just doing recursive superintelligence over here. Um, but of course the company is doing very well. Um, they, uh, emerged from stealth in May with 650 million in funding focused on building open-ended self-improving systems and potentially compute-intensive approach to AI research. This multi-year deal is meant to provide flexibility as the company looks to scale up those systems. Uh, recursive's $410 million outlay represents the bulk of the company's fundraising to date, but on a call with TechCrunch.

• 18:48Jordi Hays?

  Hey, hey, hey, they still have a couple hundred million left over.

• 18:50John Coogan?

  Founder and CEO, Richard Socher emphasized that he expected it to be the first of many such deals. So is this, I feel like normally when you see a, like a compute deal signed, it's always like more complicated than just like, we're buying this expensive thing. It's usually like, we're, we're paying that. I'm like, I, it's, we, we used to be so like anti circular deal that now I just have come to, I've been so normalized by them that I expect them every time. And I'm like, wait, wait, this is just, there's no circularity here. I would have expected like.

• 19:20Jordi Hays?

  Changing.

• 19:21John Coogan?

  Yeah, like Amazon's investing in you and you're buying trainium and racking it. And you guys do, they're doing new campus and they're investing in this and that, and you're investing in them. Uh, instead it just seems like it's a pretty vanilla deal. It's like, they're just buying a lot of compute from Amazon. Great.

• 19:34Jordi Hays?

  Seems like it.

• 19:35John Coogan?

  Well, good luck to them. Very excited for what they're launching. Um.

• 19:39Jordi Hays?

  Yeah. Uh, Jason, VP of Startups and VC at AWS says, part of the agreement is that we're going to co-develop him for a purpose built for these types of companies. So fingers crossed, but it seems like we could get some, some circularity.

• 19:57John Coogan?

  Yeah, let's hope so. Uh, let me tell you about Railway. Railway is the all-in-one intelligent cloud provider. Use your favorite agent to deploy web apps, services, and more, while Railway automatically takes care of scaling, monitoring, and security.

• 20:12John Coogan?

  Fingers crossed.

• 20:15John Coogan?

  Well, well, here's a deal that's somewhat circular. We got Nvidia, who revealed as a tenant for a $50 billion data center that will use its chips. So they're the tenant of the data center that uses its chips. We'll talk to Tae Kim about this. CEO Jensen Wong deploys balance sheet to backstop growth of AI computing market. Nvidia has signed leases worth up to $50 billion for a massive Texas data center. That's very, very big. That's very, very big for a single site. A previously undisclosed commitment that shines a new spotlight on the chips, on the chip group's growing role in financing AI. The nearly $5 trillion company

• 20:55John Coogan?

  is leasing the entire 1 gigawatt facility that developer Hut 8 is building, which will house hundreds of thousands of Nvidia graphics processing units. So you have to imagine that, once they have these, they serve something or wind up selling them. Like, these things change hands so many times, there's a lot of different ways that this could play out ultimately. But in the, uh, can we pull up the Nvidia chart? Yeah, there we go. Nvidia, ooh, big candle today, up 3%, 5.17 trillion. Let's take a look at Apple, 4.99. They crossed 5 today. They're down a little bit since they, since they breached that. But they're neck and neck. Google is sitting at 4.

• 21:39Jordi Hays?

  Yeah, Apple running the do-nothing-win strategy.

• 21:42John Coogan?

  Yeah.

• 21:43Jordi Hays?

  Jensen doing thousands of deals.

• 21:46John Coogan?

  They didn't even sign the open letter. There are three companies that still, I believe still haven't, uh, signed the open letter.

• 21:54Jordi Hays?

  Only three companies on the entire surface of the earth.

• 21:58John Coogan?

  No, uh, there's three major companies that are, uh, how do I actually get that to go away? I don't know. You can keep looking at Alphabet. But there are three, there are three major companies that haven't signed that Nvidia open letter about banning open source, uh, and/or not banning open source. Uh, and it's Amazon, Amazon, Apple, and Anthropic. Anthropic put out a, uh, a post yesterday, very, very clear response, sort of outlining their view on open source, their stance. Uh, we should go through it. But, um, the interesting thing that, uh, Ben Thompson, uh, was talking about today was, uh, the fact

• 22:37John Coogan?

  that Apple and Amazon haven't signed, and they both have, like, very physical elements in the world, in the sense that they're not, uh, they're sort of un-sloppable. Like, you can't vibe code in Amazon warehouse, you can't vibe code in iPhone. Uh, there are threats to those businesses, of course, and of course, Apple should benefit from open source, and so should Amazon, because they'll be able to serve open models across AWS. But it's just potentially interesting. I think the Apple standing it, standing back is more just like, look, we're not jumping on with this crazy open letter that everyone is signing. Like, we just have our own brand, we're thinking different, we're doing our thing.

• 23:13Jordi Hays?

  Well, and based on other Apple AI timelines, I would expect them to sign it in maybe a year or two.

• 23:20John Coogan?

  Potentially. If they decide in 2028, it's just like, robots everywhere.

• 23:26Jordi Hays?

  WWDC 2028.

• 23:27John Coogan?

  We're ready. Now.

• 23:29Jordi Hays?

  We're signing the open letter.

• 23:31John Coogan?

  These glasses have changed you. They turned you into a beast. Uh, let me, let me run through the, the three Anthropic proposals, because, uh, it's an important response. So Dario Amade, Anthropic CEO, uh, responded directly to that letter, summarizing, uh, supporting open weight models, uh, that circulated over the weekend. So, uh, to summarize his, uh, he makes three claims, just to sort of clarify things that I think are important. One, he says Anthropic has never advocated for a ban on open weight models. Now, that's a blanket ban. There's obviously, like, we're defining what a ban is, and what an open weight model, what a distilled model, what a foreign model is, these things all matter. But he has, he has come out and said, look, we never advocated for a total ban on open

• 24:11John Coogan?

  weight models. Uh, two, he says, uh, undergirding all of this is, the US must beat authoritarian governments in the AI race. He points to China, but he identifies any authoritarian government. If they get really powerful AI, they'll come over here and steamroll us, and you won't be free to do whatever you want to do in America. Uh, three, powerful AI models may be misused to carry out cyber attacks or biological attacks. There are, there are risks to having really, really powerful, uh, AI, open source systems just running around. So he's worried about those three things, clarifying those three points. But he makes three, uh, recommended actions. He makes three proposals. Uh, first, he says, uh, let's continue to sanction chips. Let's not sell chips to China. He says, we should

• 24:54John Coogan?

  not sell powerful chips or chip-making equipment to China. So this has been debated for years, like, going back to the Biden chip, uh, controls. Um, everyone knows every different angle on this, uh, the basics. I mean, there is a pretty good argument for chip controls, uh, even on purely geo-economic competitive grounds. Like, even if you don't believe in the, the, the risk of authoritarian governments having powerful AI, even if you just think it's like, you know, fancy autocomplete, it's like, well, it's the engine of our economy, and if you can slow down a rival economy, that's beneficial to you, right? Um, and so, and there also seems to be basically unlimited demand for chips in America. So, uh, by restricting sales to China, that shouldn't actually hurt American chip companies all that much. Uh, but yes.

• 25:37Jordi Hays?

  Well, they just, like, their argument would be, we fully lose the Chinese market.

• 25:44John Coogan?

  Yeah.

• 25:44Jordi Hays?

  Which is

• 25:45John Coogan?

  Yeah.

• 25:45Jordi Hays?

  second-largest

• 25:46John Coogan?

  Yeah.

• 25:47Jordi Hays?

  computing market in the world.

• 25:48John Coogan?

  No, no.

• 25:48Jordi Hays?

  Right? So, so I think, um, but, but you're, the counterpoint to that is, you were gonna lose it anyways.

• 25:54John Coogan?

  Yeah. Um, and a lot of that ick stems from the fact that China, China has been building an indigenous chip supply for decades. We've talked about going back to, uh, a whole bunch of their, you know, state-led, state-funded chip, uh, in fab processes. They've always been, uh, a few years behind. Uh, and so maintaining that gap, all else equal, is an advantage for the United States. Uh, the second point Dario makes is, uh, he says, we should crack down on industrial-scale distillation operations. Uh, this seems totally reasonable. Customer, uh, companies can set their terms of service, and they have a right to maintain intellectual property, uh, with proper legal consequences for violations. Uh, Anthropic's been fighting distillation attacks, but according to them, it's not that

• 26:38John Coogan?

  effective. Dario proposes policy interventions to deter this behavior. And this is where I'm still not clear on what, where that goes next. Like, what is the correct policy intervention? There's a, like, policy intervention is a very, very broad thing. It can mean anything from, like, a tax, a tariff, a fine, uh, a sternly worded letter, uh, not getting invited to a golf tournament. Like, there's so many different things that policy, like, covers these days, right? Um, where, where does this actually go? He says he doesn't want a blanket ban on open weight models, but it does seem like one possible policy intervention would be to sort of, like, ban, restrict, or pressure open weights models that can be reasonably shown to have been distilled. So if there's someone who's just

• 27:23John Coogan?

  a perfect distillation, it just gets, it just doesn't quite feel right. It's hard to quantify these things. We don't have a binary where you run some sort of algorithm, and you say, yes, this was distilled. Because you can distill, you know, half on Opus 5, and then throw in a little GPT-5.6, and then mix in some, some Mistral, and like, just be distilling from all over the place, fine-tune stuff, change the flavor, change the RL environment. There's so many different pieces of it. And Tyler, you were making a point about, uh, tinker, or?

• 27:54Speaker 02

  Yeah, I mean, so, so the, the inkling model from Thing Machines, like, it used some synthetic, synthetic data that was created with, uh, I think, KIMI K2.5.

• 28:03John Coogan?

  Yes.

• 28:03Speaker 02

  So like, does that count as, like, distillation? Like, probably not, when people usually talk about it, but like, it definitely benefited from Chinese open source models.

• 28:12John Coogan?

  Yeah.

• 28:13Speaker 02

  So if you're downstream of

• 28:14John Coogan?

  Yeah, I wouldn't call that industrial-scale distillation. But it's sort of downstream of industrial scale.

• 28:21Speaker 02

  There is like some big gray area where, like, how do you actually define these things?

• 28:24John Coogan?

  Yeah. And so defining that is going to be what, that's going to be the conversation that plays out in, in DC, like, behind the scenes, uh, on the basis of this. And that's where the actual negotiation is gonna happen, between, uh, you know, the, the position of Nvidia and everyone that signed the letter, versus the position of Anthropic and everyone who didn't sign the letter. They're gonna sort of just decide, okay, well, if you can prove this, this, and this, and you can show us that your API was getting hit by these different things, and you have a really solid report of what happened, and then the model also, you know, sort of, you know, checks these boxes quantitatively, when we do this eval, then maybe we will pressure it. And then, what does that actually mean? Um, you

• 29:06John Coogan?

  could go after the lab that committed the distillation attack with lawsuits, but that seems really difficult, given the international nature of these attacks. So we're sort of back to where we started, where, you know, you're, you're, like, what, what can the government do that the lab can't? Like, the lab should be looking at every customer and saying, oh, this seems like someone who's trying to distill. They keep asking for, basically, what looks like a lot of training data. They, they're not acting like a normal user, just being like, build me a website. Okay. Um, anyway. Third, he says, all sufficiently capable models, open and closed, should go through mandatory, mandatory safety testing. So this was recently outlined by Demis Hassabis over at Google DeepMind as well. And, and it

• 29:49John Coogan?

  seems like the, the two companies are in alignment on this in particular. Uh, and it's a somewhat reasonable position, uh, although the risk is that small companies who have safe models could, that aren't distilled, could get tied up in a review queue for years before they can release. Like, that would be very, very annoying. If you're, uh, recursive superintelligence, for example, and you don't have a Washington DC office, and you're like, hey, we want to release our new model, uh, and they're like, yeah, totally. Like, you gotta go through the, the, the review process, get in line. And then it's like every, you know, every trillion-dollar company is there with a ton of lobbyists being like, we'll review our model first, because we want to get out a week before the, the small

• 30:31John Coogan?

  startup. And that's the frustration of biotech, the FDA, anything that goes through approval. We've talked about this with the nuclear stuff. Uh, it gets very tricky, and so you want to avoid that, and you don't want to wind up slowing down innovation that's happening on small scales and decreasing competition. Uh, Dario does do a good job of, like, acknowledging upfront that, he says, it would protect a US AI companies from competition, but that's never been my goal with anything that he's saying here. Um, and so, uh, it's still worth working through what happens in a really adversarial situation. Like, what if a foreign lab distills a bunch of frontier models? They're the most aggressive, they're just distilling everything. Then they jump forward a bunch in capability, uh, they get a bunch

• 31:14John Coogan?

  of smuggled chips, they take all the restrictions off of cyber, all the restrictions off of bio, uh, and then they just drop the weights on like a torrent. Where they put it up on Hugging Face and Hugging Face is like, this is really crazy. No one likes this. There's a lot of pressure to take it down. I don't know. But it's out there. Like, what does the government actually do? Like, the government probably pressures or bans like hosting the weights, maybe serving the model. You maybe you won't be able to run it in American data centers. You go to the NeoCloud and say like, hey, this thing is actually bad. And I think people are divided on this because they see the current models not as actually dangerous, which is totally reasonable to assess. Yeah, it's not that bad. But like, if there was a model that was like, yeah, it's actually just like the killing machine.

• 31:57John Coogan?

  Like, I think that most people would be like, yes, I'm democratically voting to not serve that because it's just like, it's an annoyance at best and like actually bad at worst.

• 32:07Jordi Hays?

  And the other big question is like, how much compute do you actually need for it to be dangerous?

• 32:11John Coogan?

  Yeah, totally.

• 32:12Jordi Hays?

  Is, is like, having some GPUs in the back shed gonna be enough?

• 32:17John Coogan?

  Yeah.

• 32:17Jordi Hays?

  Maybe for sufficiently advanced model, yes?

• 32:20John Coogan?

  Yeah.

• 32:20Jordi Hays?

  Or do you need access to a ton of racks?

• 32:24John Coogan?

  Yeah.

• 32:24Jordi Hays?

  Ton of power?

• 32:25John Coogan?

  Totally.

• 32:25Jordi Hays?

  And then you do need to work with a NeoCloud in that case?

• 32:28John Coogan?

  And as soon as you're a US-based company with a real data center, with a bunch of NVL72s in there, you probably have registration and, you know, all sorts of just like business registrations where the government can reach out to you and say, hey, we're actually really worried about this. Just like, there are other things you can't host in a data center. There's all sorts of stuff that's illegal, even if it's intellectual property.

• 32:51Jordi Hays?

  Yeah, exactly. Yeah, that's, that's.

• 32:53John Coogan?

  Like, you can't, even just because you have a data center, doesn't mean that you can like, take an open source, you know.

• 32:59Jordi Hays?

  You can't, you can't as a data.

• 33:00John Coogan?

  Oh, yeah, open source Marvel. Like, they'll be like, no.

• 33:03Jordi Hays?

  Even, even, you know, a CRM company can't knowingly support like a organized cartel, global cartel.

• 33:12John Coogan?

  Yeah.

• 33:12Jordi Hays?

  That is like trafficking narcotics, right? You'd have to imagine like, that they have.

• 33:19John Coogan?

  They have to vibe their own.

• 33:20Jordi Hays?

  Balances.

• 33:22Jordi Hays?

  Maybe.

• 33:23John Coogan?

  So, so, what's interesting is like, what is the next step of that? So, if there is a bad model and, and everyone agrees like, okay, yeah, we gotta not host this, not distribute this. Like, yeah, the weights are out there. People are trying to like, sort of run it a little bit. But does it go offshore? Do we wind up in like the crypto scenario where there's like these offshore things and people are using VPNs to get access to it? Like, what level of aggression do you see from the US government in that scenario? It probably should be proportionate to like the danger imposed by the model. Like, if it's just a model that's like, that's like annoying or like slightly IP infringes, but like, no one's really being like, I'm not, I'm canceling my Disney subscription

• 34:04John Coogan?

  because this new model will generate me Disney IP. Like, that's probably not like, okay, put up a crazy firewall. But if it is like the ultimate hack machine that's like stealing everyone's money from the banks, then yeah, you are gonna put up the firewall and sort of be much more aggressive. So, I think the response will be in reaction to whatever the power of the models are. But it'll be interesting to go back and forth. Anyway. All in all, the letter clarifies a lot about the Anthropic position. So I think it's good that it came out. But it's still worth working through the game theory of like what happens down the line. Policy interventions is all we got here. And I think it's still too generic at this point. I want to know like what policy looks like. I want to predict that. I want to understand what's actually being proposed, what people like, what people don't like.

• 34:47John Coogan?

  And so, I think we'll learn more about this in the coming days. Let me tell you about Figma. Agents, meet the canvas. Your AI agents can now create and modify your Figma files with design system context. We have Take Him in the waiting room. Let's bring him in to the TVP on UltraDome. Tay, how you doing?

• 35:05Tae Kim?

  Hey, guys. Doing great.

• 35:06Jordi Hays?

  What's going on?

• 35:07John Coogan?

  So, so tell me, last time you were on the show, you bottom ticked it? What's going on?

• 35:12Tae Kim?

  Uh, I think I made the bullish call on CPUs, memory, and Nvidia. Nvidia is up like 5%, 10%, but.

• 35:19John Coogan?

  Nice.

• 35:20Tae Kim?

  The CPU names have still doubled even after this big drawdown. And the HBM3 names are up 100%. So, I'm hoping that, you know, it's the same thing again. I come on here, stocks go up again.

• 35:33Jordi Hays?

  Yeah, ideally, we could have like an emergency reserve of Take Him experiences. So if the market is ever down.

• 35:39John Coogan?

  Strategic reserve.

• 35:39Jordi Hays?

  Strategic reserve. We call you up, you jump on, and then it's just up.

• 35:43Tae Kim?

  It was funny, because it was literally the exact bottom.

• 35:45John Coogan?

  That's amazing.

• 35:46Tae Kim?

  And, you know, it went exponential after that.

• 35:48Jordi Hays?

  The Take Him effect.

• 35:49John Coogan?

  So, where are we right now with the level of FUD, the level of downward pressure on the AI trade broadly, the chips, the semi-trade? Like, reset for us on like where sentiment is? And then we can work through the different pieces of counterexamples?

• 36:06Tae Kim?

  So, I think sentiment's very negative. And we kind of had this huge up, parabolic up move the last few months. And likely, a lot of retail and hedge funds piled in. And we're seeing this unwind now. I think the first, a big part of it was Iran war getting worse. Every time we had the first ceasefire negotiations, stocks started taking off right, right after that. And then when we had the actual ceasefire, we had a follow-through. And then as soon as Trump started bombing Iran again, you know, chip stocks have kind of plummeted in the last two, three weeks. And then now, we're seeing just, you know, back to the old pattern of media and the viral hot takes,

• 36:50Tae Kim?

  spreading a lot of FUD. I think we saw earlier this month, I think Reuters quoted like Zuckerberg's about agentic AI. They took it out of context. And then every media person was with a hot take that this Meta was seeing bad returns and they're gonna cut its capex. And then we had leaks right after that saying that it looks like Meta is gonna raise capex. So, we're seeing a lot of this hot take FUD. Yesterday, I think we had a flurry of stuff that scared people. The Wall Street Journal vendor financing article that we'll see, we'll see what happens with that. We had CMXT IPO in China and everyone freaked out over that. We had the information article on ASML.

• 37:35Tae Kim?

  We could go through each one. And then the Kimmy thing, it's obviously a big thing.

• 37:38John Coogan?

  Yeah, we'll definitely get there. And I want to talk about open source and Nvidia's strategy there, obviously. Starting with the Mark Zuckerberg news in Reuters. This was July 2nd. Meta's Zuckerberg says AI agent tech progressing slower than expected. Zuckerberg added that the company's reorganization that included major job cuts was not as clean as it could have been. Zuckerberg and other Meta executives have been seeking to moderate some of the organizational changes introduced this year. And they said that the trajectory of agentic development over the last four months hasn't really accelerated in the way we expected. The company's bets on new structure haven't come to fruition yet.

• 38:20John Coogan?

  And so, people were sort of reading this as maybe Meta's going to pull back. But then it felt like the response was extremely quick with, with Boz going on a podcast and Alex Wang sharing a whole bunch of progress across a few different models and data points. And then Semi Analysis wrote a whole bull case for MSL, talking about how they have compute. And also, they have more of like the internal structural alignment to sort of properly YOLO in the AI era, if I'm boiling it down as brutally as possible. Just because with Google, there's always this debate between, oh, do you sell the TPUs? Or do you sell the cloud? Do you have, vend it in the product?

• 39:04John Coogan?

  Whereas Mark Zuckerberg is able to sort of like go all in on this new idea. And so, maybe there's more, more glimmers of hope there. But what else have you been tracking downstream of Meta's ambitions?

• 39:17Tae Kim?

  Well, I mean, they've been very upfront that they're investing heavily in AI. Alexander Wang is tweeting multiple times every few weeks that they're, they're going full force. They're gonna redo open source AI models. I think he said that at the YC event over the weekend.

• 39:32John Coogan?

  Yeah.

• 39:32Tae Kim?

  And it's, I mean, if you actually look at, and then Reuters came out, I think, with an article saying that they're actually gonna raise capex dramatically this year and next year. So, all that kind of fear that, that quote about agentic AI from the town hall, that kind of like spooked the market for a few days, it kind of, it was completely false.

• 39:53John Coogan?

  Yeah, it feels like it's a, yeah, it feels like it's a comms error. Because the, the language that's been coming out of Meta has been a little bit like, AI is gonna replace our employees. And it feels like it'd be much better for them to, to come to the market with a message of, we're going on the offensive. Like, we're a hyperscaler.

• 40:13Tae Kim?

  Well, to be, to be fair, that, that was the internal town hall. They didn't mean to leak it. And Reuters leaked that one quote and put out, put out the headline before the article came out.

• 40:22Jordi Hays?

  Yeah, it's interesting, like Meta, did Meta basically go through like an eight-year period where like internal town halls didn't instantly leak?

• 40:28John Coogan?

  I think everything leaked always. I think, I think everything's been.

• 40:31Jordi Hays?

  I know, but there was, there was a period where like, the, the, the sort of attention of the media was way, way, way less on like what Meta was doing internally relative to the 2010s. And all of that attention just went to the labs, right?

• 40:46John Coogan?

  Yeah, yeah, yeah. No, that makes sense.

• 40:49Jordi Hays?

  Yeah, I guess the question is like.

• 40:53Jordi Hays?

  The question that I keep coming back to is like, where is their revenue ramp? Where is their AI revenue going to ramp? And when, right? Because as you just keep-

• 41:03John Coogan?

  Zufar would say ads, like the ads, like the AI has-

• 41:06Jordi Hays?

  Yeah, and that's-

• 41:07John Coogan?

  They're accelerating, everything's good there.

• 41:09Jordi Hays?

  That's always been my view too, but when you're, when you're continuing to ramp CapEx-

• 41:14John Coogan?

  Yeah.

• 41:15Jordi Hays?

  With, and saying like, we're going all in on Agentic and we're building a harness and we're also going to do open source. And it's like, well, what is the strategy?

• 41:24John Coogan?

  Sure.

• 41:24Jordi Hays?

  Like, where, where, what is going to take you to a billion dollars of like pure AI product revenue?

• 41:31John Coogan?

  Yep.

• 41:31Jordi Hays?

  Or, or, or just API revenue? What's going to, and then to five and 10? And what's going to allow you to like justify the spend other than? I think the market would love if they just said, yeah, we actually need all these GPUs because we can actually be 10 times, we're already good at ads. We can be 10 times better. And that's where we're going to get the ROI on all of this CapEx.

• 41:53Tae Kim?

  Well, they're definitely getting ROI on that. The market is worried about, you know, all this extra CapEx on the, they're going for the frontier AI model race again. And they had to reset, like they had to, a lot of people left and now Wang hired a ton of people and uh, we'll see what happens over the next, it's going to take time. It's going to take six to 12 months before we can see any more progress. But that first model that came out a few weeks ago was a lot better than people expected. It wasn't, you know, the frontier, but it was much better than what people expected.

• 42:23John Coogan?

  Yeah, yeah. So how have you been processing the Nvidia letter around open source? And all the back and forth, all the people jumping on, the companies that have been staying back?

• 42:38Tae Kim?

  I think it's been very impressive what they've been, they've basically united the entire tech industry against Anthropic in the last like three, four days.

• 42:48John Coogan?

  18 trillion in market cap has signed on, last time I checked, across uh-

• 42:53Tae Kim?

  I mean, Google took a little time, Amazon signed on eventually, I mean only, yeah. They signed on yesterday, they tweeted out.

• 43:00John Coogan?

  Interesting. I think Apple is still the holdout, but that feels-

• 43:03Tae Kim?

  Which is kind of strange because they're the one that would most benefit from open source, open weight models being, you know, more available, I would think. But I don't know what Apple's thinking. But I mean, they pretty much got the whole tech industry to uh, kind of corner Anthropic in their position.

• 43:21John Coogan?

  Yeah.

• 43:23Tae Kim?

  OpenAI signed on.

• 43:24John Coogan?

  Yeah. What did you think of Dario's-

• 43:26Tae Kim?

  Obviously Nvidia is afraid.

• 43:27Jordi Hays?

  Yeah, I don't know, I don't know how, how, if, I don't know if they're really-

• 43:34Jordi Hays?

  I don't read it, I don't read it as being like cornered by any means, right?

• 43:39Tae Kim?

  Well, Jensen is on the record that, you know, he said, I think to Bloomberg, that there was a rising sentiment that something was going to happen on the, on the regulation front.

• 43:48John Coogan?

  Oh.

• 43:49Tae Kim?

  At the White House or whatever.

• 43:50John Coogan?

  Sure.

• 43:51Jordi Hays?

  So this was kind of- Yeah, we saw that, this was, this was last week, you had at least four people in the admin say, we're not against open weights, we're against distillation. And at least I was reading into that of some type of regulatory action around open weights. And, and then positioning it as we're targeting, this is like, we're hitting-

• 44:11Tae Kim?

  I already wrote yesterday, yeah. Yeah. About, you know, pushback and restrictions and he's doing it under the safety umbrella, but uh, definitely Microsoft and Nvidia are worried that uh, what the White House or Congress is going to do something on this front. And that's why they uh-

• 44:30John Coogan?

  It seems very reasonable that he would have no problem with like Gemma or Lama or any of the open source from like American hyperscalers where if you find out that they're distilling, you just walk across the street and sue them. Uh, and also these big companies have huge, huge, I mean, they have safety teams, but also just like huge incentives to not have a safety incident happen on their watch because you're trying to like catch up to the frontier and then all of a sudden you have a safety incident. That's going to be really bad for your overall brand. Uh, and you have a different business to protect, whether it's social networking or uh, Google search. If all of a sudden the Gemma model winds up being a thorn in someone's side for a cybersecurity reason or a bio reason, uh, that would be uh, really, really bad.

• 45:13John Coogan?

  But a, a, a foreign company that is just like hurling it over here can kind of just be like, you guys deal with the consequences potentially. So I think that's what, what Dario is worried about. Uh, what about um, the overall idea of like where, where it feels like we're sort of replaying the DeepSeek moment? Open source is going to reduce cost and so uh, that's a reason to pull back on the AI trade overall. How have you processed that?

• 45:38Tae Kim?

  It's almost, it's almost a perfect catalog. Uh, people are worried about Kimi.

• 45:42John Coogan?

  Yeah.

• 45:43Tae Kim?

  But when you actually read the technical paper and their blog posts, uh, this is not a tiny efficient model. This is 2.8 trillion parameters. It's going to require a ton of compute to serve. I mean, we saw it the first day they put it out, but their servers got slammed. Um, even in the blog post, they say it's best run on a, a kind of a, a server with 64 GPUs. So big super clusters that are network well and, and that's perfectly runs great on Nvidia. And if you remember, you know, during the whole Deepsea thing about a year, a year or so ago, uh, the market freaked out that, you know, DeepSeek was going, was so efficient that it will lead to a compute glut. But DeepSeek was the example of the reasoning model that actually, it was the opposite. It created a ton of demand.

• 46:27Tae Kim?

  And I think the same thing is going to happen with Kimi, where uh, when you have more capable models that come out, people find uses for them. And right now, just like last year when reasoning models took off, Agentic AI and agents are taking off right now. And the market is kind of like not realizing that because right now, just like last year when reasoning models were taking off, right now Agentic AI is taking off. And the next six, nine months are going to be bigger than anyone believes. And Sam is on the record, Sam is on the record over the weekend saying, at the YC thing again, like people don't, I don't know why people don't like listen to you, it's on YouTube.

• 47:07John Coogan?

  Yeah.

• 47:08Tae Kim?

  That the next six months, it's going to be much more dramatically better for AI than the last two years in terms of advanced capabilities. And I, I heard you say RSI before, I think it's going to be RSI.

• 47:18John Coogan?

  Yeah.

• 47:19Tae Kim?

  People inside OpenAI and definitely Anthropic, Anthropic put a blog post on this. RSI I think is a lot closer than people think. And if RSI actually happens in the next three, six, nine months, that's going to soak up insane amount of compute. I mean, we have this exponential ramp for reasoning, exponential ramp for Agentic. And then if RSI actually happens, and I think it sounds like both frontier labs think it's going to happen very soon. That's going to soak up an unbelievable amount of compute as the AI models, you know, use more compute to, to self-develop and improve. And I, I think that's one thing that people are missing, that both Anthropic and OpenAI are kind of winking that, oh, it's happening.

• 48:00Tae Kim?

  Anytime I tweet something on RSI, all these frontier AI speakers like my tweet. So I think that's good.

• 48:06Jordi Hays?

  Uh, what, what is your, what is your sort of framework around compute hoarding? Because certainly, certainly it is, it has been happening when you look at, when you look at, you know, uh, like going back to the Meta example, right? They're not selling compute yet, they're maybe curious about it or, or exploring some deals. But they have all this compute and they're betting on their own ability to create the capability that will have enough demand to uh, justify, justify that. Do you just think it, there, there, there's so much demand overall that it just, you know, even if there's, there's hoarding, it just will leak out? And it's okay?

• 48:47Tae Kim?

  Well, there's so much demand overall. I mean, the SK Hynix executives said uh, during their IPO run, um, that their customers are asking five to six times more than they're able to serve. And they're going to double capacity over the next five years, they said. And their customers, and I, I'm going to assume, it sounded like Jensen, um, are asking for five to six times more than they're, they're able to build. So there's overwhelming demand. You guys were at the Advanced AI uh, AMD event. Lisa Su raised her CPU, Agentic CPU forecast. Just three months ago, it was 120 billion for 2030. Three months later, they raised it to 220 billion.

• 49:28John Coogan?

  Yeah.

• 49:28Tae Kim?

  Like, she doesn't do that. She doesn't do that.

• 49:31John Coogan?

  You have that just on the record?

• 49:33Jordi Hays?

  I've got that ready, I can do whatever.

• 49:35Tae Kim?

  I mean, like-

• 49:36Jordi Hays?

  Well, I just love this chart because he called it perfectly.

• 49:40John Coogan?

  He actually did, it's crazy.

• 49:42Tae Kim?

  CEOs don't, don't, you know, raise their TAMs by like these multiples in a few months if they're not seeing insane demand coming in.

• 49:51John Coogan?

  Especially not public CEOs who are serious business leaders who've been running like non-meme stocks for decades and are like serious people.

• 49:59Tae Kim?

  Everyone's freaking out that this has let the dot com bubble all over again, the finance worries. But what if these hyperscaler GPU cloud businesses are amazing businesses? Like Morgan Stanley says, if you do inference, it's 60 to 80% profit margins, right? These are amazingly profitable businesses as long as we keep growing the next few years. And, and again, just like last year, we're, we're on this exponential run right now over the next two quarters. And the market isn't seeing that. Everyone's freaking out that, oh no, we're spending too much. And even uh, Sam Altman, uh, podcast came out today and another podcast is-

• 50:36John Coogan?

  He's flooding us out.

• 50:37Tae Kim?

  Sam is out there. He said that he regretted, you know, pulling back on the compute purchases. They made a mistake by uh, not putting the pedal to the metal because now things are taking off again. So like, I, I, like, Amazon, the CEO in April, if you, everyone read his annual letter, Andy Jassy wrote. He talks about how free cash flow works. We're not betting $200 billion on a hunch. We see the demand, we know it's going to be insanely profitable and free cash flow positive in the medium to long term. So that's why you're investing $200 billion now. And in a year or two, we're going to see insane amounts of free cash flow, the thing that people are worried about right now.

• 51:19Tae Kim?

  It takes time to build out these data centers and fabs and you bet now to bring that in in a couple of years.

• 51:25John Coogan?

  Like if you, if you see uh, you see free cash flow, that is, that assumes that uh, like, like the revenues have to catch up and then the CapEx can't grow more exponentially. And so that means you have to see some sort of plateauing. Maybe it's at the end of the chart, maybe it's this 2030 range, but there is a different world of just like continued growth forever and then we sort of like run out of money?

• 51:49Tae Kim?

  The pushback I have there, that's a static view, right? If they don't grow revenue for the next three years, yes, you can't do that.

• 51:58John Coogan?

  Yeah.

• 51:58Tae Kim?

  But they're growing, Azure is growing 40%, Google Cloud is going 80%.

• 52:03John Coogan?

  Yeah.

• 52:04Tae Kim?

  You know, Amazon's growing high, uh, double digits. So if, if-

• 52:10Tae Kim?

  If revenue is growing forty to eighty percent this year, next year, and the year after, that's more revenue you have, that's more operating cash flow you have to invest, right?

• 52:20John Coogan?

  Yeah.

• 52:21Tae Kim?

  So that's, that's what people are missing.

• 52:23John Coogan?

  Yeah.

• 52:23Tae Kim?

  And if this stuff, if, if the data center you're building now, you're spending all this now, generates an unbelievable free cash flow in twelve to eighteen months because, you know, this agentic AI is actually aging and re-architecting all the workflows inside companies. And you need to do the agentic AI coding agents to make your product better. Because if you don't, if you don't iterate a hundred different iterations for your product in R and D, if you don't do AI, just like AT&T is doing at the Gentic, uh, advancing AI at AMD, he's talking about they're putting a hundred gen AI models into production. They're burning a trillion tokens a, a month, and then that's growing double digits. The reason why they're doing that is because, you know, by using agentic AI,

• 53:08Tae Kim?

  you're providing a better customer service, a better, you have a better product R and D, and you're helping your companies make better products and services. And if you don't incorporate AI into your company, Verizon, your other company is gonna do, it's gonna incorporate AI and then disrupt you, and then you lose all your revenue.

• 53:27John Coogan?

  Yeah.

• 53:27Tae Kim?

  So everyone's worried about ROI, ROI is important, but you also need return on revenue because, uh, if you don't use AI, your, your rival is gonna use AI to beat you in the market.

• 53:38John Coogan?

  Yeah, yeah, no, I, I, I think the diffusion story is still, e- even though we got like sort of jitters by the token maxing thing, just the actual usage of AI across companies is still pretty limited in terms of the amount of people that are using it, the time that those people are using it, like, there definitely is a San Francisco bubble of startups where everyone is using AI a lot, but if you just walk into a normal business, a lot of people are like, yeah, I gotta check that out, which is a crazy place to be.

• 54:07Tae Kim?

  Let me give you some, some context here, some key context.

• 54:09Jordi Hays?

  Yeah, Ara Karasian, Jared Sleeper, everyone I know, who's saying, uh, enterprise adoption disparity remains enormous, and he cited Ara saying, uh, usage would a hundred X if every company adopted AI to the degree of the most advanced companies.

• 54:24John Coogan?

  Yeah.

• 54:25Jordi Hays?

  There's like this, there's this small group of companies that are...

• 54:27John Coogan?

  People forget in the ramp, i- i- in the ramp data, like, adopting AI can mean like having a ChatGPT Pro account for someone, which is like, not exactly the same as like using Codex and like coding agents and stuff, like, it's important, I think that, you know, if I have someone on my team, I want them to be able to go and do a deep research report, but, uh, that's like table stakes. The question is like, are you actually speeding up anything that's repetitive in your job? And, uh, that diffusion is just starting to take hold.

• 54:56Tae Kim?

  So, so the total market size in terms of IT and knowledge management in corporations, it's about six trillion dollars, right, a year? The two main frontier AI model companies, OpenAI and Anthropic, I'm gonna say, I think this is roughly accurate, are doing a hundred twenty billion dollars combined in ARR.

• 55:14John Coogan?

  Yeah.

• 55:15Tae Kim?

  You know, why can't that go to two hundred, three hundred, four hundred billion in the next year or two? I mean, they're, they're, they're growing at exponential rates.

• 55:22John Coogan?

  Yeah.

• 55:23Tae Kim?

  And we're taking off, and if the market is six trillion dollars, right? Why can't they grow to two hundred, three hundred, four hundred billion in the next couple of years?

• 55:31John Coogan?

  Yeah.

• 55:31Tae Kim?

  I mean, it's, it's like, just do a little logic and rational deduction.

• 55:36John Coogan?

  Yeah.

• 55:36Tae Kim?

  Um, this is definitely possible, and it's happening right now, and it's accelerating, and people aren't, you know, they just taking, you know, these big headlines where we had this, uh, you know, fifty billion dollars for the Financial Times, and we find out it's over thirty years, it's like on the homepage.

• 55:53John Coogan?

  Wait, wait, yeah, yeah, okay, I wanted to ask you about this. Nvidia revealed as tenant for fifty billion dollar data center that will use its chips. Explain what, what is actually going on here?

• 56:02Tae Kim?

  So, so the Financial Times, you know, put on their homepage today.

• 56:05John Coogan?

  Yeah.

• 56:06Tae Kim?

  Nvidia is gonna backstop a, a lease for a data center in Texas.

• 56:10John Coogan?

  Yeah.

• 56:10Tae Kim?

  Fifty billion dollars, and I saw that, I was like, oh my gosh, oh, that doesn't sound good.

• 56:15John Coogan?

  Yeah, no, it literally sounds like they're buying their own chips, like, it sounds like the most bad thing you could do.

• 56:22Tae Kim?

  Yeah, then they actually read the article, like halfway down the article, it's like, a fifteen year lease, uh, and it's only fifty billion dollars if they renew the lease in, after fifteen years, so it's like over thirty years if they renew it. Then if you think about that, you're like, wait a minute, fifty billion divided by thirty, if they renew it, that's like a billion dollars a year.

• 56:42John Coogan?

  Nvidia's fifteen year lease commitment for the Texas site is worth basically twenty billion, and renewal options would take the total value to fifty billion over thirty years, according to Hut eight.

• 56:54Jordi Hays?

  Okay, but what do you think their, what, what are their plans for the site? Is this, they are gonna have some, like, what, what do you expect them?

• 57:04Tae Kim?

  So, so my, my point is, this is a billion, you know, whatever, a billion or two billion dollars a year, right?

• 57:09John Coogan?

  Mm-hmm.

• 57:10Tae Kim?

  It's a non-story, but it's a, it's a big headline, sensationalized headline on the homepage.

• 57:15John Coogan?

  Yeah, and also, and also, it's not that it's, i- i- it's not like you're taking a two billion dollar loss every year, it's, you are the tenant, and then you are also renting that out, so hopefully you're making profit?

• 57:26Tae Kim?

  It, it, it's a rounding error, it's like, you know, they're doing three hundred twenty billion run rate a year now, that's gonna go to four hundred, five hundred billion next year, and we're talking about something that might be a billion, you know, like, this is not a story, but, you know, this is how people run with the sensationalized headlines, and people panic and freak out.

• 57:47Jordi Hays?

  I think they just wanted to say the biggest number.

• 57:49John Coogan?

  Yeah.

• 57:50Tae Kim?

  That, that's exactly the point, and, and we're gonna see what happens with this Wall Street Journal article, um, both OpenAI and Nvidia are not commenting so far. Uh, we'll, we'll see.

• 57:59John Coogan?

  But take us through the rumor.

• 58:00Tae Kim?

  We have to wait.

• 58:01John Coogan?

  Rumor mill time.

• 58:02Tae Kim?

  Well, it's not rumor, it's the Wall Street Journal and other, uh, people reporting that, yeah, Nvidia is in talks with OpenAI to backstop SoftBank up to two hundred fifty billion, you know, to be, we don't know the details, and I don't wanna speculate and comment, but let's actually see the details before we, I, I think the market had a really big negative reaction yesterday.

• 58:22John Coogan?

  Oh, sure.

• 58:23Tae Kim?

  To this story, uh, because everyone, I mean, Jim Cramer was telling his audience, like, sell everything at the open today, because AI and datacenters.com, oh, you know, it was insane, it's just, let, let, let's see the actual deal.

• 58:38John Coogan?

  Yeah.

• 58:38Tae Kim?

  And then that tricks and the numbers before we panic and freak out.

• 58:42John Coogan?

  Yeah, yeah, yeah, yeah, that makes sense.

• 58:45Jordi Hays?

  Yeah, honestly, when you say freak out and sell everything, sell your dollars, sell your house, sell your house, sell your stocks, then I'll freak out, but until then, Tay, I feel, I feel okay.

• 58:58Tae Kim?

  I, I mean, I just see the fundamentals, I see the CEO of AMD expanding her TAM, you know, dramatically over the last three months. I see RSI under Horizon, like, every AI researcher's like, oh my God, this is gonna happen, we have to get there sooner, and then I see, you know, the obvious use case of, uh, agentic AI where you have to re-architect your workflows internally, every company has to do this. So, everything is taking off, you see like, um, you see, uh, when the president of Korea came to San Francisco area last week, you know, they had like a day in the valley.

• 59:36Tae Kim?

  Instantly, Nvidia CEO Jensen Huang, Broadcom CEO Hockton, Dario, you know, Sam Altman are there, right? Yeah, do a little logic deduction, why, why are they there like crazy? Because they need HBM memory, and they're dying to have it. So, if you think about that, that means there's insane demand, and HBM memory is in, in shortage, there's tremendous demand for it, right?

• 1:00:03John Coogan?

  Talk about the, oh, sorry, uh, talk about the Nvidia CUDA mode, it feels like, uh, a big piece of AMD's advanced AI event was, uh, maybe the CUDA mode isn't as much of an issue anymore in the age of agentic AI, you can have an AI agent write you the software that you need to use any chip, and that creates, uh, less pricing power for Nvidia, but there's another world where you're not really, like, Nvidia doesn't necessarily need a mode because everything's just growing so fast that they're still growing, but how, how have you interpreted the processing of like, the potential death of the CUDA mode?

• 1:00:43Tae Kim?

  So, AMD is on it, Kimi wrote like a couple paragraphs in their blog post about how they created a GPU kernel, all that, so everyone, you know.

• 1:00:52John Coogan?

  Sure.

• 1:00:52Tae Kim?

  Gets scared or whatever. It, it's like, we'll see what it's like in real life, you know, this is just, you know, AMD is incentivized to say, oh, CUDA's not a problem anymore. CUDA's been a tremendous mode, and I think it continues to be a mode, and the reason why is, uh, it's super reliable.

• 1:01:10John Coogan?

  Yeah.

• 1:01:11Tae Kim?

  All the bugs have been optimized and fixed.

• 1:01:13John Coogan?

  Sure.

• 1:01:13Tae Kim?

  And that comes from hitting the software, you know, millions of times and billions of times, right?

• 1:01:18John Coogan?

  Yeah.

• 1:01:18Tae Kim?

  Like, you don't know if, you know, if, if you use Claude Code or Kimi, that, what they figure out using their training data is gonna work in the real world, right? They could talk about one little piece that does well, let, let's see how it actually works.

• 1:01:32John Coogan?

  Yeah.

• 1:01:33Tae Kim?

  But Nvidia's big mode is, uh, it's scale, it's co-design of actually working through the networking, the CPU, the GPU, and how everything works together, and the other big thing is their balance sheet and their ability to, uh, get supply commitments from, you know, I, I think I said this before, um, optical startups are like upset because Nvidia secured all the supply for all the optical components. Same thing with TSMC wafer, same thing with HBM memory, so Nvidia is using their size and Gorilla and being able to prepay and get components that are in shortage, so they've become the dominant, you know, over the next year or two, you're gonna see Nvidia able

• 1:02:16Tae Kim?

  to add tons of revenue because they were able to lock up all the supply components. That's another thing that people don't really talk about, is their supply chain and their ability to work with partners and secure, secure component inventory.

• 1:02:29John Coogan?

  Is there still energy FUD that we would run into an energy bottleneck before we run into a chip bottleneck?

• 1:02:35Tae Kim?

  So, Jensen said this last week on a Bloomberg interview, that there are a lot of bottlenecks, including, um, data center shell, power, and all those things, uh, components, um, energy, whatever, so all those things, it sounds really bad, right? And then, right after that, he said, I think we have, the chip industry has enough supply to double their revenue every year, basically implying Nvidia has enough supply, uh, for, for energy and all that stuff. No one is pricing that in, so everyone talks about bottlenecks, Nvidia CEO just basically told you on Friday that they have enough, uh, enough supply chain and all the bottleneck stuff to double revenue every year, and no one, you know, Nvidia's revenue, uh, estimates

• 1:03:20Tae Kim?

  for next year are a lot lower than double, I'll tell you that.

• 1:03:23Jordi Hays?

  Do you think the market prices in just how much of almost every important AI company in every category Nvidia actually owns? Like, it feels like every single, like, we're constantly focused on who's going to raise CapEx next and where is this quarter coming in? And it feels like in two or three years, people will look at Nvidia's balance sheet and be like, wait, they have what I imagine then will be, you know, could end up being a trillion dollar plus of just like ownership in all of these great companies, which again just goes back to the advantages of that early scale while they're, you know, while

• 1:04:06Jordi Hays?

  all these companies are trying to compete away Nvidia's margins and all these different things, they've been able to accumulate, again, positions in all of these incredible companies. I mean, we saw the SSI news yesterday is a great example of that, but how do you see it?

• 1:04:24Tae Kim?

  So, I think, look at Jensen's history in investing in these companies and CoreWeave and see how much money they made. They just bought stake in the optical companies, Lumentum and Coherent. Jensen is enabling the future because he sees this overwhelming tidal of demand and he needs these companies to be able to build up their supply chain and to give supplies and chips to Nvidia, so they actually ramp very hard. You know, everyone's freaking out that this is vendor financing. What if hyperscale GPU cloud is so profitable and these companies need capital to build up that supply so they can serve the GPU cloud services over the next year or two?

• 1:05:09Tae Kim?

  Maybe Jensen sees that coming like he did with all these other companies like CoreWeave, and that's why he's investing in these companies to be able to expand their ability to make the components the industry needs. So I think you're exactly right. In a year, two, three years, Nvidia is going to have like all these stakes in these companies, and it's going to look like he was a good investor, because he has been in the past.

• 1:05:34John Coogan?

  I mean, the man can buy a leather jacket for like five grand and sell it for a million dollars. I don't know what else you need to see.

• 1:05:42Tae Kim?

  I mean, think about.

• 1:05:43John Coogan?

  Are you the secret bidder? Did you win that?

• 1:05:46Tae Kim?

  No.

• 1:05:47John Coogan?

  I gotta get you a jacket. The real question is, how long until someone distills a jacket and open sources it? You can get a dupe of a Jensen jacket for two bucks. That's what I want. Anyway, thank you so much for coming on the show. Jordy, you got anything else?

• 1:06:02Jordi Hays?

  This was great.

• 1:06:04John Coogan?

  Yeah, this was great.

• 1:06:05Jordi Hays?

  Always a pleasure. Thanks for putting up with all of our jokes.

• 1:06:08Tae Kim?

  Hopefully, this becomes the lucky charm for the markets.

• 1:06:12John Coogan?

  Yes, I agree.

• 1:06:14Jordi Hays?

  I agree. I agree. The bottom is in. Great to see you, Tay.

• 1:06:18John Coogan?

  Have a good rest of the week. Goodbye. Let me tell you about CrowdStrike. Your business is AI, their business is securing it. CrowdStrike secures AI and stops breaches. I wanted to run through on the power issue, there's an interesting article in the Journal, an underground nuclear reactor is coming to this Kansas town and it's dividing locals. This is something that I had tweeted about years ago, like, why don't we just put the nuclear reactors underground, put solar panels on top, best of both worlds, optimal use of energy, but there's a lot of fear and uncertainty and doubt about this one. They say no one has tried operating a commercial one a mile down until now. It's great that it's here, it's kind of bad that we're the guinea pigs, says the residents of Parsons, Kansas.

• 1:07:00John Coogan?

  Residents of the sleepy farming outpost agree on many things, but whether to put an experimental nuclear reactor a mile deep in the granite beneath their town isn't one of them. Elected officials and some others see a chance to create jobs and lure data centers and manufacturers to a rural patch whose economy has been flatter than the surrounding cornfields. Another group is effectively saying, not under my backyard, it's newbie, not nimby, because it's not under my backyard. Noombi or something like that. I put $125,000 into my house and now a nuclear reactor is coming to town, said Gerald Johnson, an IT professional who planned to retire in Parsons. I can't think of a worse idea.

• 1:07:44John Coogan?

  No one has tried operating a commercial nuclear reactor deep underground until now. I'm surprised no, even like the Soviets in like 1950 didn't try it. I feel like they were trying everything. I'm surprised.

• 1:07:55Jordi Hays?

  Yeah.

• 1:07:56John Coogan?

  But the so-called gravity reactor is the creation of Liz Mueller and her father, Richard Mueller, emeritus professor of physics at the University of California, Berkeley, it's Berkeley people again, and an inventor. They founded DeepFission, a three-year-old California startup that raised $150 million in the past year, including $40 million last month through an IPO, largely to fund the work in Parsons. Parsons, with a population of 9,600 people, sits about midway between Kansas City and Tulsa, Oklahoma. DeepFission drilled a first test hole this spring on 100 acres.

• 1:08:28Jordi Hays?

  Office chairs like that have really fallen off, which tells you now might be the time to bring them back, right?

• 1:08:35John Coogan?

  No, I need a new office chair. I might go for one of those, the high-backed leather. It's good, good. DeepFission drilled a first test hole this spring on 100 acres at a mostly overgrown industrial park dotted with old munitions bunkers just outside town. On a recent day, Maurice LaFountain, DeepFission's senior engineering director, showed off a pink-flecked granite retrieved from the company's first test hole and joked that the billion-year-old rock would make a nice countertop. An empty steel container canister sat on a cleared drilled pad, waiting to go down a second hole this year. The plan is to send another one loaded with nuclear fuel into a third hole to heat water

• 1:09:15John Coogan?

  a mile underground and generate electricity on the surface in 2027, 2028, an astonishingly short timeframe by industry standards. Interesting. Anytime you're putting a nuclear reactor in the hole, it's kind of scary, he said in his office. It's great that it's here, but it's kind of bad that we're the guinea pigs. Very interesting. I'm surprised we haven't heard more about this company, this idea, everything that's actually being planned. There's something a little, I understand where they're coming from, there's something a little bit nerve-wracking about, like, even though you would think a mile deep, if something goes wrong, it's less of an issue, it feels like, well, people, it's harder to get to and like, just go and solve the problem. Deal with it, as opposed to like, oh yeah, it's a building over there, I see people coming

• 1:09:59John Coogan?

  in and out all the time, the experts are in control. I don't know. What do you think? Are you pro-nuclear underground, a mile underground? Could be the future.

• 1:10:08Jordi Hays?

  Could they not find maybe a place to do that that wasn't right under a town?

• 1:10:14John Coogan?

  It's not right out of town, it's outside of a town. You need some infrastructure, you know, like, I bet what they would say is like, look, it's a 10,000-person town, we went miles away, we're on 100 acres of land, like, we are outside of the town, but yeah, there aren't that many places that are truly like uninhabited for like hundreds and hundreds of miles just because of the nature of America. There's towns all over the place, every street, and you need roads to be able to deliver equipment and whatnot. Anyway, let me tell you about Codex. Codex is a powerful workspace for getting work done with AI agents, whether you're writing code, analyzing data, creating content, or automating business workflows, Codex helps you move projects forward from start to finish. What is this? How did we get here?

• 1:10:59John Coogan?

  Anyway, we have Ben Zweig from Revelio Labs coming on the show. How are you doing, Ben?

• 1:11:05Speaker 04

  Good, good. I love that intro.

• 1:11:07Jordi Hays?

  That was just for you. We're testing that out for the first time.

• 1:11:11John Coogan?

  Yeah.

• 1:11:12Speaker 04

  Nice, and that's just the vibe.

• 1:11:13John Coogan?

  Yeah, the vibe of the labor market. Take us through a little bit on your background, how you work, and then some of what you're tracking in the labor market and how it ties to your actual business.

• 1:11:24Speaker 04

  Yeah, for sure. So I'm a labor economist, been tracking labor market data for a long time, and started Revelio Labs. So Revelio Labs is a workforce data company. We're collecting, curating, synthesizing all labor market-related data that's out there in the world, and of course, you know, a big question is how is AI affecting the labor market? Though, you know, we're uniquely positioned to answer that question, and it's on everyone's mind. So we started putting out this labor market, this is kind of AI labor market tracker, which is really about answering like, how is AI affecting the labor market today? So not really getting into the speculation of what might happen.

• 1:12:01John Coogan?

  Yeah, just today.

• 1:12:02Speaker 04

  But really, like, what do we know?

• 1:12:03John Coogan?

  Really quickly, what is your business model? Who gets value out of this data? And then I also would love to know, how do you go about getting more accurate data? Because I see like, you know, obviously, the Census Bureau, the government has access to do polling, ADP is a very logical place to get data because they run payroll, so they can see the data, but what's been your strategy there? And then who's the customer?

• 1:12:27Speaker 04

  Yeah, yeah, so I'll start with the customer. So a lot of it is hedge funds, so they're speculating on the performance of companies. Yeah, nice.

• 1:12:36Jordi Hays?

  Thank you for your service.

• 1:12:37Speaker 04

  Here for hedge funds. Yeah, they don't get a lot of love these days, but yeah, they are speculating on the performance of a company that they have no affiliation to. You have to understand the, you know, what's happening in the company, the workforce dynamics, HR departments for benchmarking also, so strategic workforce planning, people analytics, talent intelligence, these are all like kind of segments of analytical HR and academic research. So, you know, they, of course, want to know what's going on. So these are, so basically, we get the data not through surveys, not through payroll, but really from the internet. So, you know, LinkedIn profiles, job postings, Glasser reviews, layoff notices, immigration filings, freelance platforms, like, anything and everything that is in the public domain.

• 1:13:21Speaker 04

  And that information has to be, you know, enriched and synthesized in a smart way. Like, there's all sorts of sampling biases, there's lag some reporting, there's like raw texts, you know, so we have to classify that to occupations, to skills, seniority levels, and, you know, importantly, work activities, which is more of a recent thing for us, but important these days.

• 1:13:41John Coogan?

  And then, sort of backtest against like the historical actuals to see that the model's working and then you can be more up to date?

• 1:13:47Speaker 04

  So, we don't backtest against financials, because if we did, no one would trust it anyway.

• 1:13:53John Coogan?

  I mean about like, like, if you ran your model on like, what was the employment rate in 2021, you could look at the actual employment rate to sort of calibrate that your system is predicting employment correctly? Is that roughly correct?

• 1:14:06Speaker 04

  Yes and no. I mean, for some models, we can see what was retroactively revealed. So, you know, when someone changes their job, they don't necessarily update that right away, and we can see that, you know, every timestamp has like more information than it did before. So that's like a solvable problem, but in terms of, you know, saying what's happening in the labor market at large, we can kind of use BLS data, so BLS is Bureau of Labor Statistics, we can use that data to kind of like proxy for it, but that's got issues itself. So I don't know if we want to use that as ground truth. So, you know, I think BLS has, you know, a view on what's going on from survey data, ADP has from payroll data, and we have from internet data, and they're all kind of independent in their own way, kind of uncorrelated errors.

• 1:14:46Jordi Hays?

  Okay, so I want to understand how

• 1:14:52Jordi Hays?

  How you look at your data in the context of AI diffusion, right? So a company, an individual company or an industry might have fluctuating like labor data, right? Maybe they're adding a lot of people. But then individually, if you look at those companies, maybe some companies are adopting AI quickly. Some companies in that sector aren't really adopting AI at all, or they're doing it in a very minimal way. Let's say they just have a basic, you know, chat GPT $20 a month subscription. So like, how are you? I was, I was talking to John, maybe was it six months ago? I was saying like, I really want there to be a firm that is just studying AI diffusion

• 1:15:34Jordi Hays?

  in specific industries and getting into the nitty gritty, probably doing surveys to actually understand how. Cause every company says they're adopting AI, but we all know that there's like such a broad spectrum. And then of course, some people are saying that just because they want to be in, feel like they're a part of the club.

• 1:15:55Speaker 04

  Yeah, yeah, I think it's probably mostly those that want to be part of the club. But I agree, I mean, so, so there's a few ways to get adoption data. So, so I think adoption is the hardest part of all of this because that's really a firm level, you know, piece of information. Whereas AI exposure is like more of a person level piece of information.

• 1:16:12John Coogan?

  Sure.

• 1:16:12Speaker 04

  So, so I'll tell you the way, we do it in a couple of ways. So one is that we, we had a partnership with, we still have a partnership with Ramp. So I know, friend of the pod.

• 1:16:22John Coogan?

  Let's go.

• 1:16:23Speaker 04

  So they, they can track adoption just using like AI spend.

• 1:16:29John Coogan?

  Yeah.

• 1:16:29Speaker 04

  So you can see like dollars spent on tokens, et cetera. So, so that's like a pretty good way to get adoption. The problem with that is that, first of all, it's like a self-selected sample. You know, Ramp skews toward more like tech, which is fine, like that's overcomable. The other issue is that it's anonymized, so they can't release information at the firm level. So, you know, when we collaborate with them, like we have, you know, the labor market data and they have the adoption data. So, you know, it's like complicated, you know, we have to send data, they have to like run something, we have to do some matching. So it's like a little bit, it's got some friction. The other way to do it is through, we use this measure, which is used in a paper, a recent paper

• 1:17:13Speaker 04

  that measures adoption by like, sort of hiring AI integration teams. So the thought is that, you know, if someone's like hiring AI integrators, you know, beyond some threshold, that they're like taking it seriously.

• 1:17:27John Coogan?

  Yeah.

• 1:17:27Speaker 04

  And they're embedding it into their business processes. And by that metric, we see about nine percent of firms like getting very serious about AI. It's a very conservative way to measure AI adoption.

• 1:17:37John Coogan?

  Yeah.

• 1:17:38Speaker 04

  But seems to be pretty good, like it's correlated with all sorts of other things.

• 1:17:42John Coogan?

  Yeah, I sort of hate that idea as a metric, but it probably makes so much sense in larger organizations that that is a great signal, but it just feels like completely the wrong way to go about actually changing a business. Like, I, I feel like adoption should be so much more ground up than like, oh, we're hiring a special team to do this. But that's the way businesses work and I think you're correct to identify that. It probably is very indicative of a change in the stance of the business.

• 1:18:09Jordi Hays?

  Where's, where's an area that, that AI is really good and you're seeing job loss? Cause like AI is pretty good at software engineering now or generating code. And, and you know, the companies that are adopting it the most are hiring a lot of engineers. Whereas I've heard in LA specifically, apparently the models that do product photography. So men and women that, you know, wear a bunch of clothes for like an old Navy when they're releasing a new collection. Like that work has been very impacted because that talent, they don't have a brand yet, right? And so maybe certain companies will just say like, yeah, let's just take this shirt that we have

• 1:18:53Jordi Hays?

  and just generate it on 20 different AI models and you're, and we're good to go, right? It just doesn't, doesn't really matter that much if they're using real talent or not. And so they choose the easier, cheaper route.

• 1:19:07Speaker 04

  Yeah, I think that's a great example. I mean, for the most part, you know, across the board, adoption is generally correlated with growth. But where we're seeing reductions, I mean, I think, I think the creative fields are a great example. So, you know, if you need like video B-roll or just like, you know, stock images or just, you know, podcast intro music, you know, that, that is like very easy to get from these kind of AI-generated, you know, creative elements.

• 1:19:34John Coogan?

  Sure.

• 1:19:34Jordi Hays?

  Yeah, it's so, it's so fascinating because how many people? Yeah, it's like, it's just quite interesting because when you, when some of these things,

• 1:19:47Jordi Hays?

  how many people were actually in those roles? Like would it actually, does it show up in labor data at a, at a large scale at all, right? People that are just doing, you know, stock photography and making their living that way or?

• 1:20:01John Coogan?

  Yeah, and a lot of these people might've sort of sloshed around, like, I mean, I see, I see Instagram reels from people who years ago were posting like After Effects tutorials, Premiere Pro, DaVinci Resolve, like little video editing tutorials. And now they're posting like AI-enabled workflows. And instead of showing you how to deal with a green screen, the old fashioned way, they're just doing it the new way. And they're probably still doing it for clients and the client, the, like the client is like spec is just like, I need ads that convert and they're just doing more of the work. But then there's other stuff that's bleeding out, all sorts of different stuff.

• 1:20:38Speaker 04

  Yeah, I mean, one, one kind of framing, I, I would put this in is that, you know, the, we're seeing a lot of kind of automation of things that are very task-based, things that are like really micro jobs. They aren't like full jobs at all. So we're seeing like declines in freelancing across the board. So freelancing is hit pretty hard, but that's really an environment where people transact in tasks. They're not, yeah.

• 1:21:00Jordi Hays?

  Yeah, this is why we were just talking about this earlier. The, you know, historically, like if you need a really specialized website, like it's not your main site. But let's say in our case, we're doing a drop, three years ago, we would have gone

• 1:21:14John Coogan?

  Yeah.

• 1:21:14Jordi Hays?

  and maybe gone to Upwork and, and said like, hey, I need a simple website made and just find somebody to do that one off. And now AI is just so good.

• 1:21:22John Coogan?

  Or like a basic logo for a first draft, that would be like a 99 designs. Before you bring in like a real branding firm, you might just get a freelancer to mock something up for you. Now image generation can do that for sure. What do you make of the computer science shifting? Because there are so many opportunities for entrepreneurs, startups are growing. There's some tech layoffs, but at the same time, it feels like just in general, if you're, if you have a computer science degree, you're probably going to be a bit better at using AI broadly. And so there's lots of opportunity and yet the, the number you have here is computer science enrollment is down 28% from its 2022 peak.

• 1:22:03Speaker 04

  Yeah, yeah, so I have mixed feelings on it. First of all, it's very dramatic.

• 1:22:08John Coogan?

  Yeah.

• 1:22:08Speaker 04

  So one thing that, that kind of, one optimistic take is that the supply side of labor markets is actually quite responsive to changes in technology. And that wasn't obvious before. And you know, if people can reorient themselves flexibly, that's great. That means, you know, we can, we can be adaptive, we can have more of a dynamic economy and worry less. So I'm encouraged by that responsiveness. I think it's an overreaction for two reasons. One is that we are not seeing declines in employment, you know, based on the firms that are adopting a lot. And that's true in engineering, it's true in tech. We're not seeing mass layoffs despite the narrative. So I think it's premature for that reason.

• 1:22:50Speaker 04

  Another reason is that I think even just a couple of years ago, maybe even less, I mean, time is like elusive to me. But I think, you know, not so long ago, you know, we thought of AI as chatbots and code assistants. And now it's more agentic tools. So it used to be such a low barrier to entry type of technology where, you know, anyone's grandma can use it. And, you know, coders, you know, engineers were really just like, you know, replacing their work at high rates. Now, you know, we're seeing, you know, complicated tools, like, you know, agentic systems are hard to use.

• 1:23:29John Coogan?

  Yeah.

• 1:23:29Speaker 04

  They, they, they kind of favor the digitally native and people who have experience with engineering.

• 1:23:34John Coogan?

  And even when they're easy to use, there's, there are a whole bunch of like, from a business, from an enterprise perspective, like cost trade-offs, privacy, security, how, how deep is this system? Like even just firing up a coding agent today, you're hit with prompts like, do you want this to be, have access to your documents folder? And that's like a question. And, and a lot of consumers are like, I don't know. And a lot of it, businesses are like, I don't know. So there is some sort of like capability overhang.

• 1:24:08Speaker 04

  Yeah, yeah, and I think, you know, it's, it's a different job than it was before. You know, like people are, you know, engineers are spending less time, you know?

• 1:24:17Speaker 04

  You know, doing the front-end engineering for a website, but they're doing more of kind of that DevOps. So I think it's premature and I think we'll, I mean, you know, I suspect we might have a shortage of engineers in the way that now we have a shortage of radiologists. Everyone was nervous that like radiologists were going to be a thing of the past and, and now there's a shortage and you know, wages are super high.

• 1:24:40John Coogan?

  It's like the final boss of AI automation. AI researchers like one day I'm coming for you radiologist. You imagine that it all started with like a radiologist just bullying an AI researcher and be like, what you're doing is so useless and the AI researchers like, I'll show you radiologist, I'm going to put you out of a job and the radiologist just like, I'd like to see you try. And then years and years go by. Talk to me about hires to posting ratio. It's down 38.6% since late 2022. I can imagine that there's a lot of slop posts. We were debating this before, but how do you tease that out?

• 1:25:22John Coogan?

  What do you make of the hires to posting ratio dropping?

• 1:25:26Speaker 04

  So this is the thing that I get the most nervous about. So, you know, we're seeing some slop posts, some slop job postings, but we're also seeing a lot of slop applications. And when a job goes up, you know, you get, I don't know if you guys have posted a job recently, but I just did last week and I got, you know, a thousand applications in the first like five minutes. There are all these like job boards that are kind of helping people auto apply. Even Indeed is doing this, which I think is a bad move for the record, but they'll do what they want. It's, you know, so, so basically employers are getting completely signal jammed. They're getting overrun with these applications that look strong, but they really have no way of verifying. So the utility of each job posting is going down.

• 1:26:08Speaker 04

  It's not, it's not as good of a way to find candidates anymore. So employers are relying on networks. It's getting harder to hire. And in the economy at large, we have this kind of low hire, low fire environment where there's just not a lot of movement in the economy. And I think that is the result of, you know, AI usage in the search and match process.

• 1:26:29John Coogan?

  Yeah, you would think that, like, I, I've been surprised that social media has not been that overrun with slop. Like there's, there's definitely some slop problems here and there, but the, in general, the algorithmic feeds have been sort of set up to deal with this where the bad slop gets filtered out pretty quickly.

• 1:26:54Speaker 04

  Except for LinkedIn, but yeah.

• 1:26:56John Coogan?

  Sure. But I've been, I've been surprised that, that there hasn't been as much of an intermediary where you put up a job post. Yeah, you get hammered with a thousand applications, but the filtering is really, really good so that you're really only looking at the top 10. Maybe you dip into the top 100, but you're not at all annoyed by the bottom 900 because I guarantee you that there are millions and millions of sloppy Instagram videos out there that would annoy me if I saw them, but the algorithm will just never show them to me. And then maybe there's one that uses AI, but it's good and it will show it to me because I still enjoy it. So it feels like, hopefully there's people working on this. I'm sure that people are, but that feels like the, the next iteration to unclog this

• 1:27:41John Coogan?

  because that seems like a major problem. Like you need the matching in the, in the US economy to be really, really strong.

• 1:27:46Speaker 04

  Yeah. I mean, there's been some regulatory challenges there too. So a few years ago it became illegal for employers to, you know, sift through candidates using AI.

• 1:27:55John Coogan?

  Wow.

• 1:27:56Speaker 04

  And I don't know how enforced that is.

• 1:27:58John Coogan?

  Yeah.

• 1:27:59Speaker 04

  But it's, it's a liability for employers and not a liability for candidates. So there's some asymmetry in how, in who can use AI.

• 1:28:07John Coogan?

  That's very interesting. I had no idea. When did that? Yeah, I remember, I remember that, that you can't use AI to, to, to, to filter out candidates. I think of it as like, I, I, I understand where that came from on like bias based into models and like very preliminary, uh, deep, barely deep learning algorithms to sort of like, look at the person's name and look at their graduation date and like try and filter for that. Like, I'm just thinking about like, it like, did, is the resume a complete slop? You know, like a complete, like a Pangram level that doesn't seem to impose like bias in the same ways that they were trying to avoid. So we're in this weird like knock on effect world, but that's the way these things go. Jordy, anything else?

• 1:28:50Jordi Hays?

  No. Come back on as there's more, come back on as there's, as there's, uh, yeah, more, more data that's notable.

• 1:28:57John Coogan?

  Yeah.

• 1:28:58Jordi Hays?

  You can tease the hedge funds a little bit.

• 1:29:00John Coogan?

  Yeah.

• 1:29:01Speaker 04

  Yeah, for sure.

• 1:29:02John Coogan?

  And congrats on the progress. Thanks so much for coming on.

• 1:29:04Jordi Hays?

  Yeah, great to meet you Ben.

• 1:29:05Speaker 04

  Thanks guys.

• 1:29:06John Coogan?

  Have a good one. Let me tell you about MongoDB. What's the only thing faster than the AI market? Your business on MongoDB. Don't just build AI, own the data platform that powers it. And let me also tell you about Cisco, critical infrastructure for the AI era. Unlock seamless, real-time experiences and new value with Cisco. Up next, we have Aakash from Takeoff. He's the founder and CEO. He's been on the show before, but this is the first time.

• 1:29:29Jordi Hays?

  Here he is.

• 1:29:30John Coogan?

  A new flag. How's it going? Give us the news.

• 1:29:33Speaker 05

  Uh, it's going well. It's great to see you guys. The news is that Sierra just bought us. We announced it last Thursday.

• 1:29:40John Coogan?

  Jordy completely missed.

• 1:29:43Jordi Hays?

  Let's go.

• 1:29:45John Coogan?

  That's the first, that's the first miss.

• 1:29:51Jordi Hays?

  First miss.

• 1:29:52John Coogan?

  Brutal.

• 1:29:53Jordi Hays?

  Hard to do that to you.

• 1:29:54John Coogan?

  Congratulations.

• 1:29:55Jordi Hays?

  I gotta go back, go back to practice. I'll work on this.

• 1:29:58Speaker 05

  So, no, I'm excited. I'm excited. Thank you guys.

• 1:30:02John Coogan?

  So tell us the story of the company. I mean, we got, we got enough time for, I think for you to tell the entire story from start to finish because all of this sort of happened pretty quickly. Where were you before you started the company? When did you start the company? What was the growth like? Take us through the, the, the journey.

• 1:30:17Speaker 05

  Yeah, absolutely. So we started the company a little over a year ago. We started to build basically like agents that would be slightly more capable than what we're seeing today.

• 1:30:27John Coogan?

  Sure.

• 1:30:28Speaker 05

  Right? We fundamentally founded the company on this hypothesis that there were two different kinds of agents. There's human in the loop agents and then there's truly autonomous agents. Human in the loop agents are the agents that we all love to talk about. Like we're talking about plot code, Codex, things that you prompt, they do things, they can do them for a very long time. It could be minutes, dozens of minutes, hours, even some case, in some cases, days. But fundamentally you are the person that kicks them off and evaluates their work. And then like you were talking about in the previous interview, clicks, accept viewing my downloads folder.

• 1:30:55John Coogan?

  Yep.

• 1:30:56Speaker 05

  Autonomous agents are not that. Autonomous agents, when you think about it from the perspective of a buyer, it should feel like you are multiplying your labor force. And when I say feel like, I mean, it should be a one-to-one translation. I should feel like when I buy Takeoff, I'm buying a hundred thousand agents that can do what I might have a team of disparate humans, software, et cetera, doing today, but at a much more massive scale. That was like the foundational thesis of like, how do we actually build agents that can do this? We call them long horizon agents. We call them fully autonomous agents. And then we'd go after explicitly revenue-aligned use cases. And the reason we went after revenue-aligned use cases is, well, there's a lot of different reasons. But the most obvious reason is like, why are you going to buy mission-critical AI software from a kid with crazy hair? Like the only thing that's going to get you to do that is if I can prove to you, I'm going to make you more money. Right?

• 1:31:41Speaker 05

  And the way I get, and the way I get to prove to you, I'm going to make you more money is I make it very zero risk for you. I'm like, give me your lowest quality leads. I'm talking to a lending company. Give me your patients that are kind of like churn if I'm talking to a healthcare company. And like, let's see what I can do with the agents that I build for your company. Let's see if I can recuperate that lost revenue. Let's see if I can increase your top line. And if we are all successful here at the end of the day, I'm going to be in your board deck at the end of the year because you bought a piece of software and revenue is up double digits. That was the foundational pitch thesis, the whole idea of what the company was going to be. We tried building agents in different, in different ways. We started actually with browser agents because we figured if we can use software, we can do what humans do.

• 1:32:16John Coogan?

  Yeah.

• 1:32:17Speaker 05

  But we thought that that would actually, we realized, not thought, we realized because we get spit, eaten up, chewed up, spit out by the market over and over again. We realized that that's like not actually.

• 1:32:25John Coogan?

  Over and over again for like, for like six months.

• 1:32:26Speaker 05

  For six months, for six months, that's fair.

• 1:32:29John Coogan?

  It's not like you're like, yeah, we were getting chewed up.

• 1:32:31Jordi Hays?

  One simple trade.

• 1:32:32Speaker 05

  For the viewers, what Jordy and John are referring to, if you haven't read our literature, which I don't expect you to, is that we entered this calendar year at effectively $0 in committed revenue. And by the time we got acquired by Brett and Sierra, we were at near biggers in revenue. So what they're referring to is that very short and vertical kind of, no pun intended, takeoff in revenue ramp.

• 1:32:53John Coogan?

  Let's go.

• 1:32:54Jordi Hays?

  And so again, like. More people should name their company Takeoff.

• 1:32:57John Coogan?

  Great nominee, that's a lesson. It's amazing. Yeah.

• 1:33:01Speaker 05

  I mean, it's got its own SEO things. Former member of Migos, rest in peace. Now we're, we're honored to like, carry the name with a positive light. That being said, like, you're only going to make money if your agents are trying to sell or like, you know, trying to be sold upon the value of adding revenue, if you actually add revenue. And so we de-risk it because I, you know, I'm not Brett Taylor. I can't walk into a room, or at least I couldn't walk into a room previously and get someone to pay for something that's not already driving results. So we go in, we do pilots. They're not necessarily free, but they're paid on outcome. And so if I drive this outcome that we're talking about, usually directly revenue or something tied to revenue, for example, for a lending company, loans funded, loans originated, I get, you're going to pay Takeoff the way you pay a human being on commission and some

• 1:33:44Speaker 05

  sort of base units based on the amount of tokens, voice, SMS that are used. So the customer thinks I'm only paying when I create X thousands of dollars in margin, I'm paying hundreds of dollars of cost of goods sold. They love that trade-off. And they're like, if it fails, it fails. And if it succeeds, we're making more money at the end of the year. That's how a kid with crazy hair walks into a room and ends up selling multimillion dollar contracts over and over and over again. Cause like, once it starts actually working, even what I did not expect really is like the compounding nature of, of, of exponential growth.

• 1:34:12John Coogan?

  I love it. Walk me through how deeply you're integrating or you were integrating with some of those first customers, because there's a world where you're just like, give me the stale leads, I will go off and I will do the email, I will do the SMS, I'll do, I'll do the, whatever happens and I'll sort of like, either build my, build those systems or maybe you'll set up your own like Mailchimp account or whatever you want. Or there's another version where you're like, I just, I want to live within your CRM, within all your tools. I will do API integrations into whatever legacy systems you have to actually collect all the knowledge to make the correct move and actually drive revenue.

• 1:34:50Speaker 05

  No, it's a fantastic question. And it was, it's categorically the latter.

• 1:34:54John Coogan?

  Okay.

• 1:34:54Speaker 05

  There was this, there's this lecture, for lack of better words, I give for every potential candidate and existing employee of Takeoff, which is, we are given the privilege, right? Not the right, but the sheer privilege of sitting between our customer and their revenue. Like, I could explain this in a million ways, why it's so important. But the most important thing to explain is that our buyer was always the CEO or a C-suite member. It wasn't some VP of something that reports into something that reports into the CEO. When you are selling revenue, you are selling to the CEO. That's what he or she is getting graded on at the end of the year. Whether they're a public company, of which some of our customers are, or whether they're a massively multi-billion dollar private company, they're getting graded on revenue.

• 1:35:32John Coogan?

  Shots fired at the chief revenue officers in the audience, but.

• 1:35:36Speaker 05

  No, no.

• 1:35:39Speaker 05

  It's, it's, I mean, it's one, it's one of the things Brett pointed out.

• 1:35:41John Coogan?

  You're 100% right.

• 1:35:42Speaker 05

  It's kind of amazing that like every one of your customers, your contact, like the person who's in my iMessage, top five iMessage is the CEO.

• 1:35:48John Coogan?

  Yep.

• 1:35:48Speaker 05

  And so, to answer your question, I would give this lecture that when we are referenced by our customers, they have to think of us as their best employee. When I say us, I mean myself, like Aakash, Spencer, my teammates' names. They have to think of us as their best employee. And the way we get there is we have to understand their business as well as any individual that works for them. They could be a 10,000 employee company. They should be able to ask us about anything that is even remotely related to the line of work that our agents are doing for their business, and we should be able to answer it. I'm talking about gross margins, I'm talking about conversion rates, I'm talking about time to fund. I'm talking about time between first contact to revenue generated, literally everything. And we know we've succeeded when the CEO starts asking us questions about their business. That's when you're in like, you know, the promised land.

• 1:36:32Speaker 05

  That's when you are literally their friend, when they're texting you 5:30 in the morning, 10 o'clock at night. And like, none of your family or friends are in your top five iMessage anymore. It's just your customer CEOs. And so, it's very much understanding, like, the intricacies of that business in order to build an agent that could actually do what's going to drive that company's revenue. And so, we have to understand every, every piece of software that they're using, every single thing that somebody might do. Because we are trying to genuinely scale the workforce. And you can only scale the workforce if you can do it end to end. And that's like a really important thing that I think most agent companies don't get. If you're going to sell an agent into some work stream, but you're only going to take like a horizontal slice, it's virtually useless, because then you have to educate everything below and above it, how to drive the end-to-end results. So if you want to actually drive the business outcomes, own the whole thing.

• 1:37:16Speaker 05

  If you want to own the whole thing, you have to be capable of owning the whole thing, which means you have to understand the business well enough to build the agent to do so. You guys had Marky, who's been a friend and incredibly, incredible founder, who I've learned a lot from on the show, I think a month or two ago.

• 1:37:28John Coogan?

  Yeah.

• 1:37:28Speaker 05

  And Marky talked about how her entire company and product is rooted in this foundational philosophy that we have to translate whatever language the company is speaking to what the agent is going to do.

• 1:37:39John Coogan?

  Yeah.

• 1:37:39Speaker 05

  But we think very, very similarly. Our culture is entirely predicated on that assumption that if we don't understand the business better than our customer or as well as our customer, our agents aren't going to do it as well as we need them to.

• 1:37:49John Coogan?

  So what does that translation look like for you? Is it a bunch of markdown files and then your agent can interpret those? Because you could go all the way to like, we pre-trained a model just for you. And then we could be like, we fine-tuned a model for you, sort of the thinking machines model. And then you could be like, well, we're using, you know, the frontier models, but we have a custom harness for you, or we customize our harness for you. Or we write a special integration, or it's just, or the agent just shows up and it figures it out.

• 1:38:17Speaker 05

  I love that you're giving multiple choice, because if you didn't, I would just like, ramble. It is the second half of answers that you just said.

• 1:38:23John Coogan?

  Okay.

• 1:38:23Speaker 05

  It's like, another foundational philosophy that we kind of built this company on is that the inference API is a commodity. That's a soundbite, you can clip me, that'd be great. And that's a crazy thing to say, right? It's a crazy thing to say that inference API is a commodity, because we think about Claude and Anthropic. Sorry, Anthropic and OpenAI at tens of billions in revenue.

• 1:38:39John Coogan?

  Yeah, yeah.

• 1:38:39Speaker 05

  People are going to be like, that's all inference.

• 1:38:40John Coogan?

  Yeah.

• 1:38:41Speaker 05

  I would disagree. I would say it's a function of the things built on top of inference. Talking about ChatGPT, Claude, Codex, Claude Code.

• 1:38:47John Coogan?

  Yeah.

• 1:38:48Speaker 05

  And what we need to be able to do is, you can call these, most people would call these harnesses, right? Like harnesses with like a great GUI with a great command line interface. But it's functionally the thing that's delivering end-to-end value. And like coding was a great first, coding and chat agents were a great first product, because that was the end-to-end value. Again, it's a human in the loop type agent, so they are giving the value to the person that's using the product.

• 1:39:10John Coogan?

  Yeah.

• 1:39:10Speaker 05

  The second, the subsequent type of like wins in enterprise AI. And when I say wins, I don't mean like hundreds of millions in revenue or even billions. I'm talking about like the next wave of tens of billions of revenue. It's going to come from the fact that your harness, which is just a fancy way of saying, agents that can do multiple things and operate across multiple different services, surfaces as opposed to a single, a call and response API, should be as capable as someone that is like, is on a job listing, or like something that you are hiring to drive an outcome or a result for the business. And so, it's a combination of harnesses, and specifically Takeoff, what we built was, what we call it as a DSL, a domain-specific language, right? So you should be able to educate, direct, and build the agents on Takeoff

• 1:39:51Speaker 05

  using our domain-specific language that is built around the idea of we're trying to handle something end-to-end. The other kind of like, unique thing about this is like, your agents have to be answerable to the outside world. Like, you can't just say, go do a thing. The thing that you are, that our customers are calling APIs for is, go fund this loan, or an API call to go onboard this patient, or go get this patient's prior authorization. That requires multiple actions by the agent that then in turn require input from the outside world. Let's use the borrower example, the loan borrower. If you're getting an API call to your agent that says, go fund this loan for this borrower, this lead, you have to call that lead, contact them, help them with that initial like rate quoting, understanding what options they have, whether it's a HELOC, a refi, a home equity loan. Then you have to have a second call, after whatever happened in between the first call and the second call.

• 1:40:35Speaker 05

  You have to go reach out to third parties, the e-notaries, the underwriters, everything else involved, document collection. Then you have to have a third call saying, hey, I noticed you got stuck here, because you have this like weird Iowa borrower question about co-borrower co-signing. Then you have a fourth call is getting it over the line for funded.

• 1:40:50John Coogan?

  Yep.

• 1:40:50Speaker 05

  This is something that for an agent to actually handle end-to-end, your harness is like, is, is transcending just like tool calls, right? It's like, it's an always-on harness with a heartbeat that's answering anything that could happen by, on behalf of, or in relation to this like central entity in this example of the borrower.

• 1:41:07John Coogan?

  Tell me a little bit about post-merger integration. I could see Sierra having a product called Takeoff. I could see Sierra just being a service or a company that you work with for a bunch of different things. And I don't even know if I need different products, because AI is so broad that everything sort of merges together into like, one product that can do multiple things. And I just flip on a switch and say, okay, I want you to handle this, I want you to handle this. But how are you thinking about integration? I know it's, I know it's like really, really early, but.

• 1:41:37Speaker 05

  No, it's, it's.

• 1:41:38John Coogan?

  I imagine that this was what you were talking to Brett about, was like, with a vision of what these two companies can do together. So like, take us through a little bit of it.

• 1:41:46Speaker 05

  Well, I mean, that's actually great. I'm going to go in reverse order of your questions here. Like, when Brett and I first chatted, we basically realized that we have a similar vision of what the world was headed towards. And what we realized was that, by virtue of just, again, being a kid with crazy hair, like, there's no chance that I was going to like, compete and win and customer support. There's a dozen companies, three of which that are like.

• 1:42:08Jordi Hays?

  Selling yourself short. You seem, you seem, to me, I'm getting young Brett Taylor.

• 1:42:15John Coogan?

  Yeah, let me pull up a picture of Brett Taylor's hair, please, and see if you can.

• 1:42:18Jordi Hays?

  When he was your age.

• 1:42:19John Coogan?

  Hair. I think you got a, I think you got a shot, but yes, okay.

• 1:42:23Speaker 05

  The point being that like, we had to come from a different angle, right? We had to sell a thing that only the early adopters were ready for.

• 1:42:29John Coogan?

  Yeah.

• 1:42:30Speaker 05

  And like, like, it's not like we, and so like, when I, when I say we had a similar vision of the, of the direction the world was headed in.

• 1:42:34John Coogan?

  Sure.

• 1:42:35Speaker 05

  Like, we were further along on that timeline.

• 1:42:36John Coogan?

  Yeah.

• 1:42:37Speaker 05

  And like, we had done this thing that I don't think most of the world realized was possible yet. And it wasn't until we proved it was, right? That's what was really exciting to, I think, Brett and the company is like, hey, we'd love to get to where you are, but we're realizing that you're already there. And why not get there together and then scale it times a million?

• 1:42:52John Coogan?

  Sure.

• 1:42:52Speaker 05

  Right? And so, that's kind of how the original conversation started. We then kind of came to this like realization that, and I think what was actually really interesting for you guys to understand, or for anyone who's listening and watching, is that we realized we were on to something when our first like three, seven-figure customers were, like, they already had customer support vendors, right? They had like a Sierra or a Decadon or something else there. They were spending on average between a few hundred grand and maybe like, maybe a million dollars with them. With us, they were spending at least three times more.

• 1:43:24John Coogan?

  Wow.

• 1:43:25Speaker 05

  Right? So like, they had, they had an AI support vendor, and they also had Takeoff. They were spending three times, in one case, eight times more than they were spending with their support vendor.

• 1:43:32John Coogan?

  And that makes sense, because you're driving revenue, and you're talking to the CEO.

• 1:43:35Speaker 05

  And if you're driving revenue, there's three kinds of software, right? There's revenue-driving software, there is functional software, and then there's must-have software.

• 1:43:40John Coogan?

  Yep.

• 1:43:40Speaker 05

  And if you're the first category, Google Ads, Facebook Ads, $1 in equals more than $1 out, I will keep spending until I flat that line.

• 1:43:46John Coogan?

  Yep.

• 1:43:46Speaker 05

  And that's what we were going for. It's exactly how we want to be thought of by our customers.

• 1:43:49John Coogan?

  Yeah.

• 1:43:49Speaker 05

  And so, we realized, you know, again, all the things that Brett and I were talking about, that we were excited about, a lot of the same, similar, shared ideas around where the world was headed. We're like, hey, together, this can be 1 plus 20 equals 1,000. And so, we, we, I don't know if you guys saw, probably not, because you have a lot going in your mind. We, we as in Sierra and Takeoff, Sierra together, launched this product called Horizon, which is this new thing. And the reason it has to be this net new thing is because we want people to realize this is a step-function jump in capability.

• 1:44:17John Coogan?

  Yeah.

• 1:44:17Speaker 05

  A step-function jump in capability, which is going to drive revenue for your business. It's not just agents in like one second, cost savings, but it's agents that your CEO is buying. And that's really freaking exciting.

• 1:44:28John Coogan?

  Yeah.

• 1:44:29Speaker 05

  I don't know if I can swear, I'm sorry. But that's really exciting.

• 1:44:32John Coogan?

  That's awesome. Yeah, no, it makes, it makes so much sense.

• 1:44:34Jordi Hays?

  You're great at, you're great at naming.

• 1:44:36John Coogan?

  Yeah, you know, these are all, every, every name is great. Like, these are all good.

• 1:44:40Jordi Hays?

  Yeah.

• 1:44:40John Coogan?

  I love them. Well, thank you so much, Jordi.

• 1:44:43Jordi Hays?

  It was great, great to meet you. I found the, I found the, the whole pitch very compelling. I was just imagining myself as a, as a CEO or enterprise buyer, just being like, I'm sold. Just send the contract, send the contract.

• 1:44:55Speaker 05

  I mean, I will say, like, I've been on a few sales calls with Brett now, and Brett, and it's very flattering to hear this from Brett Taylor, right? Like, one of the best salesmen that's probably ever lived in software's history. We've had a few sales calls together. It is magic in that room. Like, people get really freaking excited when we show them Horizon. And it's like, it is, like, people start imagining what they're going to be doing for their business, all the awards they're going to get, the fact that in the board deck, it's going to be plus double-digit percentages at the, at the end of the year. And that is very exciting for us as a company.

• 1:45:20John Coogan?

  It's very exciting.

• 1:45:21Jordi Hays?

  Amazing.

• 1:45:22John Coogan?

  Thank you so much.

• 1:45:23Jordi Hays?

  Now, I can see why you guys did the deal. Great, great to hang, dude.

• 1:45:25John Coogan?

  We'll talk to you soon. Let me tell you about public investing, for those who take it seriously. They got stocks, options, bonds, crypto, treasuries, and more, with great customer service.

• 1:45:36Jordi Hays?

  Mark Zuckerberg is in the Wall Street Journal opinion section with a new piece, "The AI Future is for Everyone." He says, "The history of democracy and economics has proved that centralized power stifles human potential." And it's quite long, I'm going to go let you guys read it, but let's head into the comment section. Let's get a, let's get a quick reaction, let's get a quick reaction.

• 1:46:01John Coogan?

  This is the Wall Street Journal, I think it'll be pretty.

• 1:46:03Jordi Hays?

  No, it, it looks relatively tame.

• 1:46:05John Coogan?

  It'll be tame.

• 1:46:06Jordi Hays?

  But yeah, in, you have. Making a, you know, a clear effort to position to be the overtly, there was a white space for a guy investing hundreds of billions of dollars a year in AI that is like, says, hey, this is going to be really great for everyone.

• 1:46:24John Coogan?

  Yeah.

• 1:46:26Jordi Hays?

  And I'm going to help us get there.

• 1:46:27John Coogan?

  It is, it is interesting. Facebook does have some monopolies, but like the competition for attention is constant and there are always sources outside. Like, they've never had a full monopoly on, on social media even with TikTok and Snapchat and LinkedIn and Twitch and YouTube and Netflix and the podcast feed and SMS and iMessage. Like, there are so many other platforms for transmitting information. Like, I, I don't know, I, it's hard to jump straight to a critique here, but the, the key quote that Andrew Curran pulled out was

• 1:47:10John Coogan?

  that he said, "In most cases, like cybersecurity, the history of open source software has shown that giving everyone full access to powerful systems will be the best way to protect safety and security over time." So, he's firmly on the side of democratizing powerful AI. And he is yet another one. I imagine that they, that they signed the letter. I, I've lost track at this point, but you can imagine that he did. Anyway, thank you so much for tuning in. The other piece of news is that Apple is launching Apple Upgrade next week. Then iPhone, iPad, Mac, and Apple Watch leasing slash subscription program. They said,

• 1:47:52Jordi Hays?

  They're saying you will own nothing.

• 1:47:54John Coogan?

  And you will be happy.

• 1:47:56Jordi Hays?

  We're launching our new program. You will own nothing and be happy.

• 1:48:00John Coogan?

  It's partnering with Klarna to launch in the United States at online and retail stores. It's now official. Leasing prices start as low as $20 or $17.99 per month for iPhone, $11.99 for Apple Watch, $24.99 for Mac, and $11.99 for iPad. So, interesting. I mean, a lot of people are saying this is a direct reaction to increased prices for memory, increased prices for products. There was a time when an iPhone was a couple hundred dollars and there were incentives to jump on a Verizon plan and you sort of amortize the cost over that. Those days are gone. Like, we're in the world of like a $2,000 iPhone. It's a significant.

• 1:48:40Jordi Hays?

  Same thing with our gongs.

• 1:48:42John Coogan?

  For people. You want a subscription gong?

• 1:48:44Jordi Hays?

  No, I'm just saying there was a time when a TVPN gong was $200.

• 1:48:49John Coogan?

  Yeah.

• 1:48:49Jordi Hays?

  Now it's in the tens of thousands of dollars.

• 1:48:52John Coogan?

  It's actually so expensive. Somebody, a friend of mine texted me and was like, "Where do we get the gongs? I need a gong." And I was like, "I think you should start small." And this is not like you're, you can't handle the big gong. I was more saying that like, there is a joy to being on the hedonic treadmill of larger gongs. Like, you don't want to jump straight to the biggest gong. You want to start with a small gong.

• 1:49:12Jordi Hays?

  And work your way up.

• 1:49:12John Coogan?

  Work your way up because every gong that we've added has been so electric when we

• 1:49:17Jordi Hays?

  I think it's time for a new one.

• 1:49:18John Coogan?

  You want an even bigger gong?

• 1:49:20Jordi Hays?

  Yeah, I want, I want one that's hanging from the rafters.

• 1:49:23John Coogan?

  A giant gong. Maybe. But, and also every gong has a different flavor, different sound, you know, different amount. You got to warm them up, all sorts of things. We always warm up the gong.

• 1:49:37Jordi Hays?

  Anyways, folks, that's our show for today.

• 1:49:39John Coogan?

  Thank you so much for tuning in.

• 1:49:40Jordi Hays?

  Enjoy the rest of your July 28th.

• 1:49:43John Coogan?

  Leave us five stars on Apple Podcasts and Spotify.

• 1:49:47Jordi Hays?

  Money never sleeps. You shouldn't either. Call me, bud.

• 1:49:51John Coogan?

  Sign up for the newsletter at tbpn.com and we will see you tomorrow at 11 a.m. Pacific. Goodbye.

• 1:49:57Jordi Hays?

  Cheers.