Subscribe
Share
Share
Embed
Technology's daily show (formerly the Technology Brothers Podcast). Streaming live on X and YouTube from 11 - 2 PM PST Monday - Friday. Available on X, Apple, Spotify, and YouTube.
You're on.
Speaker 2:You're watching TVPN.
Speaker 3:Today is Thursday, 06/05/2025. We are live from the TVPN Ultra Dome, the temple of technology, the fortress of finance, the capital of capital. We gotta work on that because we're working on selling the naming rights, baby. This place is gonna be branded. We're gonna sell the windshield.
Speaker 3:Sell the windshield. We're selling the windshield. So we gotta we gotta keep growing the intro. But we have a massive day to day. Little bit of an AI day today.
Speaker 3:We got folks from Google, OpenAI, Anthropic. We got the former CEO of OpenAI. Sequoia Mashir coming on. Sequoia, Sample, Google, Lights and Beast, OpenAI, and Topic. Pretty good coverage.
Speaker 3:We hit almost everything. Should go on a whirlwind tour of what's going on in artificial intelligence. I'm excited to dig into the state of affairs in the foundation model race. We're gonna go through the the the tier list of what companies in AI have the mandate of heaven. We're also gonna go through some of the deep research products and hopefully get into some of the more, cutting edge use cases for AI.
Speaker 3:So we have both some deep research folks coming on, and then we also have, some folks that are working on, video generation and video game generation and a lot of different, applications. We're gonna cover the granola story. We're gonna cover what's going on with Windsurf. And so it should be a great day, but let's run through some news just to keep everyone up to speed before Sean Maguire joins in thirteen minutes. So first off ramp, time is money.
Speaker 3:Save both.
Speaker 1:Save both.
Speaker 3:Easy to use corporate cards, bill payments, accounting, and a
Speaker 1:whole lot more all in one place.
Speaker 3:But clearing order inbound. Circle went public. The CEO is coming on the show tomorrow. That's very exciting. And, Jordy, you have the news.
Speaker 3:I will read a little bit, from Jeremy, the CEO. I'm incredibly proud and share and thrilled to share that Circle is now a public company listed on the New York Stock Exchange under Circle. Brian Armstrong post. Congrats to Jeremy and the entire Circle team on your IPO and reaching 30,000,000,000,000 in lifetime USDC volume. Let's hit that.
Speaker 3:Go.
Speaker 1:The big t.
Speaker 3:You got it.
Speaker 1:Incredible. 30,000,000,000,000. Stop. The stock is up massively. Yeah.
Speaker 1:It went it was priced at $31. It is trading at around 85 as of now. That's fantastic. We love to see it. Again, as many people would expect, Bill Gurley is is very unhappy with that.
Speaker 1:Inefficient pricing.
Speaker 3:He hates a he hates a stock. He hates a pop during after the IPO.
Speaker 1:It's good for the IPO window, which we all we we always want to be open.
Speaker 3:For sure. For sure. Andrew Roll executive chair, Trey Stevens, tells Ed Ludlow that the company has closed a new funding round of 2,500,000,000.0 in a deal that more than doubles the defense startup's valuation to 30,500,000,000.0. This is from Bloomberg TV. Congratulations to the Anderle team on the massive up round.
Speaker 3:You'll love to see it. We gotta hit the gong for Anderle. It again. We have Gong.
Speaker 1:Contact. Good contact.
Speaker 3:Very exciting. Bunch of news from, from Kevin Weil over at OpenAI. We will be digging into this with Mark Chan today when he joins, but Deep Research can now share across GitHub, Google Docs, Gmail, Google Calendar so you can integrate everything, and it can do research on your files. That's gonna be a lot of fun to talk about. This is also potentially threatening for granola.
Speaker 3:And so we're talking to a granola investor about, what the reaction will be, where the direction of that company might change or not. But if you're designing a tool for artificial intelligence or one of these products or any tool, really, go to Figma.com. Think bigger. Build faster. Figma helps design and development tell teams build great products together.
Speaker 3:Go to Figma.com.
Speaker 1:It is the backbone of the TBPN brand. It is. It is. And we would not be able to make the show without it.
Speaker 3:Yes. And so we are going to be digging into today is obviously artificial intelligence day in some ways. We're digging into artificial intelligence today. We're also very interested in talking about VR and content and augmented reality. And there's an and then there's a story in the Wall Street Journal today about meta is in talks, not advanced talks, just regular talks, regular old talks.
Speaker 1:Chitchat.
Speaker 3:But they're talking to Disney. They're talking to a 24 about content for a new VR headset. And this is my number one question about VR. When the iPhone everyone's looking for the VR, the iPhone moment of VR. When the iPhone debuted, what was it?
Speaker 3:It was first and foremost a phone. It replaced your phone.
Speaker 4:Mhmm.
Speaker 3:And so I've always thought that the path to true VR adoption was just saying, we're only going after your TV. The next generation of twenty twenty two year olds when they get to college or post college in their in their first apartment, they are just not going to buy a big flat screen TV because we have solved that specifically with with VR. Headsets. Exactly. Or or, you you know, the guy with multiple monitors, three three monitor setup?
Speaker 3:The production team back there, we can go to the production camp, show you all of the different, all the different monitors that they have. What if they could be wearing VR headsets? They could have seven monitors one day. 10 monitors. Let's give
Speaker 4:it up
Speaker 1:for the production team. Love to see it.
Speaker 3:So so the idea oh, we got the drink cam.
Speaker 1:We got the drink.
Speaker 3:Thank you to Matarina. Thank you to thank you to Andrew Huberman for inventing drinking things and caffeine.
Speaker 1:And the whole team over at Mettina.
Speaker 3:Yeah. They're crushing it. And so this idea of of of doing one thing really, really well before before going into, you know, trying to do a little bit of everything, like platform even.
Speaker 1:Exactly. You only get to be a platform.
Speaker 3:If you solve one thing really, really well. The iPhone didn't wasn't a platform. The first iPhone was not a platform. It didn't have an App Store. Right?
Speaker 3:It just had, it just had the ability to to listen to music. It was an iPod. It was a phone, and it was a Internet communicator, just web browser. And so the Meta is looking for to Hollywood for exclusive immersive video, for premium device. Now my kind of hot take here is that Husky's cooking.
Speaker 3:I know. I know. It's a all all of our boys are coming together doing we're cooking up something amazing. I'm I'm really excited. We're not gonna be able to get that much information on this soon, but, I I don't even know if they need that much immersive content.
Speaker 3:I think a lot of it is just, hey. Every single Meta headset should just ship with the Matrix preinstalled for free. Yeah. It's like, how much would that possibly cost? It's an extra $2 to rent or something.
Speaker 3:You could, like, have it preinstalled. So it's just like, you can put it on, and there's, like, 10 movies that are preloaded. You can just watch movies, and the movies are great. And you're in a really nice theater, and it just comes preinstalled, and it's all great. Because that was that was the my my Vision Pro experience was very much
Speaker 1:Yeah.
Speaker 3:Film driven. I would
Speaker 1:take it a step further, and I I think they would have to basically basically create an entire catalog. Yep. Like, I don't know if the matrix by itself is gonna be enough
Speaker 3:of a draw Totally.
Speaker 1:To say, I'm gonna spend hundreds of dollars for the average person.
Speaker 3:No. No. No. It's more about, like, the preinstalled apps. So, like, the iPhone was a really good phone.
Speaker 1:Yeah. But I'm saying
Speaker 3:Really good iPod, but it also came with, like, an like, a calculator app Yeah. That was, like, decent. And so you need a few of these things that are just, like, really easy to access, really to pull up really easy to pull off the shelf. And, ultimately, I think Meta needs to catch up to Apple in terms of the Apple TV movie store and and making sure that, like, all the streaming providers are really on there in a in a in a in a valuable way. Obviously, it's it's important to go to immersive eventually, but I think the path to immersive might be just, wow.
Speaker 3:I have a home theater in my studio apartment now. Anyway, we'll dig into this more. We'll talk to more people. Maybe I'm wrong. Who knows?
Speaker 3:But the high the the the the high points from this article in the Wall Street are as follows. Meta is seeking exclusive content from Hollywood for its upcoming premium VR headset, Loma, set to rival Apple Apple's Vision Pro. I'm super excited for this new VR headset. I think it's gonna look fantastic. I think the resolution's gonna be insane.
Speaker 3:And, obviously, there's lot of focus on augmented reality and and Orion and AI and glasses, but there's still so much work just to do just to bring VR into a into just a normal consumer experience. And what's interesting is that I think Apple really broke the seal on, like, yeah, like, you know, people are used to paying a thousand dollars for a phone and $2,000 for a computer and maybe $4,000 for a headset. Isn't that crazy? So what can Meta's been been hanging out in, like, the three, four, five hundred range if you take the reins off and say, hey. Yeah.
Speaker 3:Yeah. Yeah. It's fine to spend a thousand dollars on this thing. You could get something really, really interesting. Totally.
Speaker 3:And so Meta is offering millions for video based on well known IP aiming to attract users to its VR device launching next year. Now the big question is how long will these immersive videos be? Because, Apple did did do a bunch of these deals. They did license a bunch of interactive video products. And but they were all, like, five minute experiences.
Speaker 5:Yeah.
Speaker 3:And so you'd Not You'd you'd you'd get through them all. Yeah. And then you'd wait a full quarter, and Apple would be like, we have another one. Seven minutes.
Speaker 1:Here's five new minutes of entertainment.
Speaker 3:It's like that's not how people experience entertainment. I remember, like Yeah.
Speaker 6:Yeah. If you
Speaker 1:just think of it, a lot of people are use being entertained by their iPhone for four hours a day Yes. Often through video.
Speaker 3:And it was but but that's not even an iPhone thing. You go back a few decades to like, the original PlayStation had Final Fantasy seven on it. It it came on multiple discs. And that game, people would play it for a hundred hours. Metal Gear Solid was similar, like, dozens of hours of gaming, and and no one's really been able to deliver that in in VR and have that moment.
Speaker 3:Same thing with GTA. You know, hundreds of hours of entertainment. Anyway, very excited to dig into this new device known as Loma. It's more powerful than the Meta Quest VR headsets now available with higher fidelity video. Let's hear it.
Speaker 3:They got the screens done. They pulled them forward off the benchtop. The design is similar to the large pair of eyeglasses, more like Meta's Ray Ban AI glasses than goggles that the Quest and Vision Pro use connected to a puck that users can put in their pockets. So maybe they're going puck, which is interesting because that was very contrarian. Everyone's like, this is Steve Jobs would never let this And Palmer came out and said, no.
Speaker 3:Puck is great. Keep the you don't want heavy things on your face. That's just not a good experience. And so Met is planning to charge less than a thousand, but more than 300. And so I would say $9.99 is probably the right price.
Speaker 3:I want it to be sort of expensive so it can be a great product. Meta spokesman referred to the comments by Meta chief technology officer Andrew Bosworth about the company working on many prototypes, not of not all of which go into production. Meta is working with Avatar director James Cameron's Lightstorm Entertainment on exclusive VR content. The two companies announced a partnership last year. So I think we gotta get on this.
Speaker 3:We gotta have a VR stream. Yep. Three hours of content every day. We're gonna be the reason churn is low on the next VR headset.
Speaker 1:Yeah.
Speaker 3:Because you just throw this thing on it just like you're sitting here on the drink cam. You can you can click through. I mean Yeah.
Speaker 1:Just click all to the the different angles.
Speaker 3:It's pretty it's pretty doable. It's pretty doable. I mean, you can film valuables, like like, usable spatial video on just an iPhone now, and then you can play that back in the, in the Vision Pro. And it does look three d, which is cool. In other news, Wall Street Journal's reporting that Reddit is suing alleging unauthorized use of the site's data, an online discussion forum.
Speaker 3:Who would have expected this? A key accessed the site more than a hundred thousand times saying after saying it had stopped, Red is suing Anthropic. And and and Anthropic debates this. I'm sure we won't be able to get into this today because it's I'm sure it's it's it's caught up in the courts, and there's a whole bunch of legal restrictions. But, we'll do our best to understand how these deals come about.
Speaker 3:It seems like most of the time, it's not that the company that has much data doesn't want the AI company to use their data. They just wanna have an equitable agreement where everyone is getting the most value. And I think Reddit Reddit surged the stocks up. Right?
Speaker 1:Yeah. And then for more context, OpenAI is already paying Reddit approximately 70,000,000 per year in a content licensing agreement. Yep. So they kind of got ahead of this issue and decided to to strike up an actual deal.
Speaker 3:And I believe Google has a deal with them too. And I think this might be one of the
Speaker 1:has a deal with Reddit?
Speaker 3:Yeah. I'm pretty sure because there's that meme about, like, the best way to search Google is search, like, whatever your search term is and then space Reddit because the user generated content was better than the SEO
Speaker 1:stuff that was popular. Pays Reddit approximately 60,000,000 per year.
Speaker 3:So 60 and 70. They're getting a hundred and 30,000,000. That's pretty serious revenue. And and it's something that doesn't need to be brokered via a bunch of individual programmatic ads that might not work or anything like that or or or subscale. It's just one one one or two deals, and boom.
Speaker 3:You're up in the hundreds of millions of dollars in revenue. What is what what is Reddit's overall overall annual revenue?
Speaker 1:$20,000,000,000 market cap. Okay. I don't
Speaker 3:Not bad.
Speaker 1:Let me see here.
Speaker 3:How do they track into
Speaker 1:3,000,000,000, greenbacks in 2024.
Speaker 3:Point '3 billion. So they're getting, like, 20% of the revenue or 15% of the revenue. Yep. I wonder how big
Speaker 1:They grew 60% over over 2023.
Speaker 3:Interesting. The so they might be bigger they might be bigger than Reddit, or they might be bigger than Conde Nast, which at one point owned them. It's kind of unclear how how how valuable Conde Nast is because they're private. Anyway, so Reddit, said that the AI company unlawfully used Reddit's data for commercial purposes without paying for it and without abiding by the company's user data policy. Anthropic is, in fact, intentionally trained on the personal data of Reddit users without ever requesting their consent.
Speaker 3:The complaint said
Speaker 1:Interesting interesting saying saying that it's about the users.
Speaker 3:Yeah. Bills itself as the white knight of the AI industry. Last year, Reddit took steps to try and limit unauthorized scraping of its website, creating a public content policy for its user data that is publicly accessible, such as posts on subreddit and updating code on its back end. The user policy includes protections for users, such as ensuring that deleted posts and comments aren't included in data licensing agreements. And so, yeah, I don't think that there's a really strong precedent for agentic for the agentic for the agentic web.
Speaker 3:Like, if I, like, if I use Google Chrome to access a website, Chrome doesn't need to pay any sort of license. But if I go to Anthropic and say, hey. Get me up to speed on this topic, and it goes out and it browses the web, all of a sudden, it feels like maybe they do have to pay, whereas Chrome wouldn't because it's just rendering the web page, and it's not transforming it at all. What is transform? What's what's fair use?
Speaker 3:And so these things will obviously play out in the court of law. And so, hopefully, they can resolve it quickly and move on.
Speaker 1:Yeah. I'm actually surprised that, they didn't already have a deal in place Yeah. Because it's very valuable data. Totally. Want that data Yep.
Speaker 1:For your models. And anyways.
Speaker 3:Well, we have Sean Maguire joining in just a minute. And the other news in the Wall Street Journal today is Thrive Holdings is betting that AI can change IT services. The company established by venture capital firm Thrive Capital, joined with ZBS to invest a hundred million dollars into an entity that will integrate AI into IT firms. It's from Josh Kushner, of course. Shield Technology Partners has already acquired four IT service companies, Clearfuse Networks, Iron Orbit, Dellvol Technology Solutions, and OneNet Global.
Speaker 3:It said Thrive Holdings called Shield Technology Partners, an AI enabled managed IT service platform. IT service companies, also called also called managed service providers or MSPs, typically provide IT support and manage tools like software and cloud computing on behalf of businesses. Founded by Josh Kushner about fifteen years ago, Thrive Capital is known for some of its high flying startup investments, including OpenAI, Databricks, and Wizz. What a what a portfolio. Not bad.
Speaker 3:Investing in its traditional services business, particularly those that rely heavily on administrative knowledge work and adding AI to supercharge them is becoming a bit of a trend. As As part of its efforts, Shield Technology Partners will embed software engineers into each of its IT portfolio businesses. Oh, they're doing the forward deployed engineer. The engineer's goal is to build an AI driven solution that all of the portfolio companies will use. We've studied all the ways in which MSPs have perhaps been on their back foot to date with customers, and and says that IT services work is incredibly well suited to what AI can streamline.
Speaker 3:And so you can imagine a whole bunch of agentic workflows for all the different things that you need to do when you're when you're deploying cloud, in managing cloud. Really quickly, before we have our next guest, let's tell you about Vanta, automate compliance, manage risk, and prove trust continuously. Vanta's trust management platform takes the manual work out of your security and compliance process and replaces it with continuous automation whether you're pursuing your first framework or managing a complex program.
Speaker 1:If you think you should be on Vanta You probably you're probably correct.
Speaker 3:Well, we have Sean Maguire from Sequoia Capital in the studio. Welcome to the show, Sean. How are you doing? Boom.
Speaker 7:What's up, team? Never a boring day on the Internet. That's for sure.
Speaker 3:Yeah. What what is keeping you
Speaker 7:Oh, man.
Speaker 3:What's keeping you up now?
Speaker 4:Well, obviously,
Speaker 7:Well, I I think there's you you got any anyone on Twitter knows what I'm
Speaker 3:talking about. Yeah. Yeah. Yeah.
Speaker 7:Or an ex.
Speaker 3:Yeah. I mean, let let let's let's skip the politics because this is purely a technology and business show.
Speaker 7:Stick to love you guys.
Speaker 4:You're the best.
Speaker 3:Stick to the technology. What, I mean, we we we've had an interesting experience with X in that, there's always been this narrative that, like, the whole the whole platform was gonna collapse. We you know, there's been rough days here and there, but overall, things have been growing. What have you seen across the x x AI merger? What are the secrets to success?
Speaker 3:How is, you know, talent tracking? Is any of is any of, like, the chaos and noise distracting? Because when I talk to x AI engineers, they're like, we're too busy. We can't come on your show. But but but what's your experience been with the the x and x AI team recently?
Speaker 7:Look. In if you go back in time, as you said, everyone said it was gonna fail. The app would crash. You know? Nothing would happen, and that didn't play out.
Speaker 7:But there was a lot of tech debt and kinda broken infrastructure, and and there was a, you know, a couple years of rebuilding the basics and foundations. I think we're starting to see, you know, real innovation happening. I love the Grok integration directly in X. It always scares me when someone you know, when I have a tweet or whatever, then someone says, at Grok, is this correct?
Speaker 3:Is this real?
Speaker 7:Like, is this is this accurate? You never know what's gonna come back. You know, usually usually, I agree with Gronk, there's been once or twice where I think some of the subtleties are a little off. But
Speaker 1:It's truth seeking. It doesn't mean that it's fully truthful every time.
Speaker 3:Yeah. Yeah. Yeah. It hasn't actually found that ground truth every single time. That's funny.
Speaker 3:Well, what about the overall horse reality.
Speaker 7:I'm probably wrong. But
Speaker 3:Yeah. Yeah. What about the overall horse race between the foundation models? It seems like every day it's going back between an OpenAI launch, an Anthropic launch, a Grok launch, a Google launch.
Speaker 5:Yeah.
Speaker 3:Are are you do you think that continues? Do you think there's, like, maybe some fragmenting and there's opportunity? I mean, we're kind of already seeing this with how much Anthropix loved by developers versus, OpenAI has been really dominant on the consumer side. And now every company is figuring out a different way to actually get to distribution. What really matters here?
Speaker 3:Is it pure scale? Is it pure cracked engineering talent? Is it distribution? Is it a combination of those things? How are you seeing it play out?
Speaker 7:Great question. I you know, honestly, my opinions have changed a lot over the last few years in in in many directions, and so don't have too much confidence in my assessment right now. But the you know, I always try to look at lessons from the past, and my current thinking is that the closest analogy are operating systems.
Speaker 3:Mhmm.
Speaker 7:And if you and I'll make a couple points on this. If you think about operating systems, first of all, there's a bunch of different ecosystem. There's the Windows ecosystem. There's the Apple, you know, OS ecosystem. Then there's, like, on mobile, there's, you know, Android.
Speaker 7:There's, you know, a whole browser environment with Chrome, and then there's open source to Linux. You know, one thing that I think is interesting about Linux, you know, there's there's more Linux servers in the world than there are Microsoft servers, but the value kind of capture of Microsoft is way greater than Linux. I personally think we're gonna see something very similar play out, whether it'll be like a you know, OpenAI will, you know, be the Apple or someone and, you know, XAI, I think, will be very successful. I think there's a good chance that Anthropic is independent and successful. I also think there'll be a big open source component, which should be, like, Linux.
Speaker 7:And I think there'll probably be 10 to a hundred times as many open source models out there, like or, like, deployments of open source models in ten years. But I think that they won't be as valuable and they won't be, like, as rich of ecosystems. And then just to make two more points on the open source analogy, like, for Microsoft, by having or Apple. By having the operating system, you know, they were able to actually win in quite a few ways on the application layer as well. You know, for Windows, they bundled in, you know, Word and Excel and, you know, then Outlook and all these other things.
Speaker 7:I think it'd be very similar for the foundation model companies. I think that the foundation models would be, like, table stakes. That'll be their kind of win, but also a very sticky moat. And even if they're not the most profitable businesses themselves, it will give them big advantages kind of on the application layer. And then one other thing that I think will happen, you know, the cloud companies have giant moats just through the CapEx dynamics of of cloud, like, to buy all this hardware and, you know, innovate with hardware and stay there is a big moat.
Speaker 7:I think these foundation model companies are gonna be I think there's gonna be way more value that occurs and would be way bigger moats than people realized. I think they'll all basically have hardware moats like cloud style hardware moats. They will have the, like, the operating system style, you know, very, very detailed research that's hard for anyone to replicate. And then I think they'll probably make a lot of their profit from applications on top of it. That's that's my current thinking.
Speaker 7:Thinking can change inside Yeah. Fingers blocking.
Speaker 3:So I know, obviously, you weren't, investing during the original operating system boom, but your firm Sequoia Capital was. And so have you had any discussions with the with with the kind of the lineage of the firm or the history, and and seen how is the revenue ramp or the business scale different this time than, say, in the .com era or in the previous era? It feels like it's ramping faster than ever. It feels like we're seeing more companies that are hitting a billion in revenue or a hundred million faster than ever. But is that real based or or anyone that you've talked to that was, investing in that era, does it feel different this time around, do you think?
Speaker 3:Yep.
Speaker 7:I mean, one of the beautiful things about being at Sequoia is we do have this long history, and we get to tap into the kinda institutional knowledge.
Speaker 3:Mhmm.
Speaker 7:You that said, sadly, Don Valentine died Yeah. Four or five years ago, like, early into my time. Rip, what a what a absolute legend. You know? And led the original Apple investments.
Speaker 7:Yeah. But there's still a lot of Google institutional knowledge in the in the firm, which is, you know, not directly operating system, but but they create an operating system later. I mean, first of
Speaker 4:all,
Speaker 7:the revenue of these companies is scaling insanely just faster than any products in history before, for Starlink. So obviously not a foundation model company. Mhmm. But I basically made, like internally, I I I made a Excel spreadsheet of AWS's revenue growth, like, in the first twenty years of AWS compared to Starlink. And, you know, Starlink has, in five years, got into where what took AWS ten years to get to.
Speaker 7:And and I and now, like, with these foundation model companies, we're seeing as fast or even faster revenue growth. You know, that said, these are very I think the business models like, the initial business model is more clear, and the profitability of these companies is in you know, the more of the in profitability is insanely high, and so you gotta discount the revenue growth. But I I would just say the biggest lesson, I think, from the past is you have to capture, like, territory early on, and and the doors will kind of close behind you because of these CapEx dynamics and, and just, like, lock in with users.
Speaker 3:Yeah. I mean, you mentioned Starlink. Do you think there's obviously yeah. It's it's such a weird company because it's, like, a space launch company that now is an Internet company, ISP. But there's actually a little bit I'm starting to hear of an AI narrative just that having Starlink potentially unlocks edge compute or inference in areas that would typically have kind of stranded energy resources.
Speaker 3:So all of a sudden, if there's some super remote area that has really cheap energy, you can go and and set up a data center there and then do inference and stream those tokens over Starlink. Do you think that's an underrated narrative? Do you think that that's developing on course? Do you think there's any bottlenecks that people should be thinking of within that story?
Speaker 7:So when we first invested in SpaceX, the part of the core thesis was Internet everywhere.
Speaker 4:Mhmm.
Speaker 7:And and I would say, like, it goes way beyond AI. But I I think the Internet everywhere thesis is is huge, and that will be, you know, everything from oil rigs, you know, to airplanes, to boats, to, yeah, edge AI devices. But, like, the I think the bigger thing for Starlink is Starlink just has, like, a 10 x plus cost advantage for moving data compared to, you know, building new Transatlantic or Transpacific fiber lines. And in the world of AI, we're gonna be moving these models are gonna be moving so much data around themselves. And I just I think Starlink is incredibly well positioned to be the pipes to move all this data for AI.
Speaker 7:And so I that I actually am I care more about that just because of the volume than some of the kind of edge applications for AI specifically, but those will be big. And then one other thing, I just gotta give a plug to to Bitcoin.
Speaker 3:Plug to Bitcoin.
Speaker 7:Let's go. Basically, three three years ago, I visited the biggest Bitcoin mine in the world. Genesis Digital, their mine is near Midland, Texas. It's actually backed by SPF, which is, you know, wild. He he got both
Speaker 1:Anthropic good bets. You cannot
Speaker 3:deny that.
Speaker 7:Exactly. He got both Anthropic and and Genesis. But these guys had a gigawatt scale Bitcoin mine operating three years ago, and already for them, like, having it it taught it taught me a lot. And, you know, bit like, Bitcoin mining is the absolute tip of the spear where you need the least amount of data movement, like data in and out Mhmm. To dollar, generated or Sure.
Speaker 7:Or, like, power consumed. And so I I actually think that was like, Bitcoin mining is underrated in terms of how much it's pushed, like, frontier power generation, turned into compute. And I I don't think it's a coincidence that Crusoe, you know, which is now powering Stargate, started off as a Bitcoin mining company or that Coreweave Coreweave. Which which is, like, $80,000,000,000 stock as of yesterday is now, you know, is now an AI data center company. And I just I think I think that's honestly the bigger the the bigger theme.
Speaker 1:Yeah. What's your updated thinking around nuclear? We have these new executive orders and it was announced this week that Meta announced a partnership with Constellation to power some of their AI power needs. What's your kind of updated outlook over the near term to medium term?
Speaker 7:I'm an all of the above guy for energy. Like, we need all of it. We need all of it as quickly as possible. I, as an individual, invested in a few nuclear companies going back, like, nine, ten years ago way too early. And and I like, to put a little bit more meat on these statements, nuclear is incredible, but deploying large amounts of nuclear is slow.
Speaker 7:Like, even if you deregulated it to zero, I think it would be, like, a, you know, a more than a decade, well beyond a decade to deploy, like, a terawatt of new nuclear, call it ten years if you did it as fast as possible for America starting now. Solar is just a way faster way to deploy a lot of energy. Nat gas is a way faster way to deploy, you know, to deploy a lot of energy. We have been producing insane amounts of natural gas, which we didn't have the pipelines to actually use, so we're just flaring it a lot of times because kinda like the the dollar value per like, when you have an oil well or you're fracking, it's producing it's emitting natural gas and and oil, and you just made so much more money from the oil than the natural gas that we didn't really care about it. And that started to flip.
Speaker 7:And so, anyways, I I think we have to do all these things. I think we need more natural gas, more oil, way more solar, and then kind of have nuclear coming as the reinforcement juggernaut coming online, like, ten to fifteen years from now.
Speaker 3:That's a good framework. Fantastic. I mean, we have to have you back for, you know, an energy deep dive. We we we we know a fair amount of the nuclear and solar entrepreneurs, and there's a bunch of people doing really cool stuff. So, have a safe trip.
Speaker 7:Per personal plug. I had a seat in the New York merc mercantile exchange. Yeah. When I was like 22 years old. It was it was insane.
Speaker 3:That's wow. Cool.
Speaker 1:Hey. Good luck on the timeline today. I know you're gonna go in there. Put on your put on your hazmat suit and just get in there.
Speaker 3:Good luck. Good luck.
Speaker 7:Peace, guys.
Speaker 3:Safe travels. Cheers, guys. Fantastic. Let let me tell you about Linear. Linear is a purpose built tool for planning and building products, meet the system for modern software development, streamline issues, projects, and product road maps.
Speaker 3:Go to linear.app. Next up, we have Jack in the studio. We have an in person guest. Let's bring him in. Play some soundboard for me.
Speaker 3:Jordy, welcome to the stream. How you doing,
Speaker 1:Good. There he is.
Speaker 3:Second time on the show. Good to have you here. What are
Speaker 1:you wearing today, Jack?
Speaker 2:I'm wearing the jacket. The t d p n jacket in the in the capital of capital. Fantastic.
Speaker 3:Thanks for coming. Thanks for hanging out. Here, you can adjust
Speaker 1:your mic a little bit there as well. Cool.
Speaker 3:I I wanted to kick this off with, like, a little bit of a rundown on the different Foundation Labs. We're talking to a lot of them today. And I and I noticed that Jordan Schneider from China China Talk and Dylan Patel ran through their AI mandate of heaven tier list. And so I wanted to read through that and kind of get your reaction and then just kinda do, a vibe check and let it and and talk to you about what we should be expecting from different labs over the next year. It's a little bit of a horse race.
Speaker 3:So, up first at s tier, they have OpenAI. It's the only foundation lab that made s tier. Does that feel right to you? What are you watching from OpenAI?
Speaker 2:Yeah. I think that's exactly right. OpenAI executing both on the product level, getting the distribution Yep. Getting into hundreds of millions of people's phones. Yep.
Speaker 2:But also also on the research level, you have people like Nome Brown, people like Aden
Speaker 3:Yep.
Speaker 2:Just doing this incredible frontier research. O three, I think Mhmm. Just as a model impresses me the most of any model that's come out so far. Mhmm. You know, Brad Lightcap said in the Yeah.
Speaker 2:Wall Street Journal recently. They had 2,000,000 workplace users in February. And they're at 3,000,000 now. Wow. So so just just really exceptional growth.
Speaker 2:I think
Speaker 1:I I was I was thinking earlier, it'll be funny. Our kids in in twenty years will be like, dad, they're making me use OpenAI teams at work. Yeah.
Speaker 8:I don't if it's gonna like
Speaker 1:the default like the Microsoft Teams default.
Speaker 3:Yeah. I mean, there's a little bit of narrative that, that maybe and and we can move on to Anthropics in the a tier alongside DeepSeek and Google. There's a little bit of a meme that, like, Anthropic is crushing it with developers. They're the default choice for Windsurf cursor users, but then OpenAI is more dominant with consumers. But I I feel like recently, I've heard that it's maybe even more skewed than people think.
Speaker 3:Like, it it's it's maybe not like the the vibe on x might be, yeah, like, you know, seventy thirty OpenAI clawed for day to day grab grab a random person on the street, but it might be even more skewed. Does that feel right to
Speaker 4:you?
Speaker 2:Yeah. I think I think Anthropic's really solidified with developers, but it's it's, like, totally given up on consumers. But I think OpenAI wants to take that on. I mean, there there's rumors about some sort of Windsurf acquisition. They're releasing 4.1 and Codex.
Speaker 2:Yeah. They're pushing hard on coding, and I think that's something to watch from them this summer Yeah. And going into 2026 is is can they can they secure that?
Speaker 3:Do you understand the model names at this point? 4.1? I I have access to 4.5. Why would I wanna go backwards? Is that is it are the models fragmenting to, like, where I'm gonna have to learn a new a new taxonomy for, okay, if I wanna write code, use this one.
Speaker 3:If I wanna write poetry, I use this one. If I wanna do, math or reasoning or build a chart, I use that one. Because it's putting more work on me, I feel like.
Speaker 2:I think Sam said that they're gonna try to fix the model naming scheme this summer. So that's the real thing to watch.
Speaker 4:Okay.
Speaker 2:They're gonna keep s tier is can they can they get coherent model names? Yeah. But, yeah, 4.1, it's cheaper. Yeah. It's specialized towards coding.
Speaker 2:It's kind of their 3.7 Sure. Type of driver.
Speaker 3:At the same time, I I I know you're not super up to speed on the Alibaba, like, models, but I I saw some I saw some release where Alibaba, Quinn released, like, a hundred different models. And Will Brown was kind of saying, like, this is awesome from a research perspective because they have, like, they have, like, one model that's just good at bio. And it's kind of like this hyper fragmentation. It's the opposite of going in the unification direction. It's actually it's actually going more specialization, and then maybe you unify that at the end.
Speaker 3:But I don't know. It seems like if you're a consumer company, you can't you don't really have that affordance. Right?
Speaker 2:Yeah. I think in terms of research, Alibaba's a bit underrated. I I mean, compared to Deepsea gets all this press, all this coverage.
Speaker 3:Totally.
Speaker 2:But the Quinn models are really good. People are doing you see from lots of people these really cool RL experiments, these really cool kinds of things. They're they're lagging behind The US models. They're not they're not a tier. They're not b tier.
Speaker 2:You know? But they're doing some interesting stuff, and I I I think that's super cool.
Speaker 3:Yeah. I really wonder if they're if they have a distribution advantage in China. Obviously, we wouldn't feel it here, but I I really haven't gotten up to speed on what is the chatty bitty of China. In terms of distribution. Obviously, DeepSeek had that moment, but have they actually executed properly on the on the product side?
Speaker 3:I don't know.
Speaker 2:I'm surprised that Google hasn't been able to turn their their general distribution advantage into an AI distribution advantage. They have these really good models. The new Gemini came out today. It's it's got really good benchmarks on a lot of things. Yep.
Speaker 2:But they're yet to I I think they're yet to crack distribution. We we sometimes say.
Speaker 3:Did you see that did you see that mock up that was just the Google search box, but a Gemini prompt? Yeah. It was like, if they wanted to go full send, if they really if they were really AGI pilled, they would just say, hey. We're done with Google search. I mean, it would destroy their economics
Speaker 4:and everything.
Speaker 2:I'd commit to it. I'd commit to it. I think, it looks like the model that they used to power those search pumps right now Mhmm. It look it seems really lightweight to me. It gives a lot of wrong answers.
Speaker 2:When asked 2.5 something, it's always right.
Speaker 3:You know? Sure. That's interesting. You think that's just AI
Speaker 1:overview box is like, we're just gonna hallucinate.
Speaker 4:Just gonna
Speaker 3:hallucination box. Yeah. Well, they they are launching, like, the, like, advanced AI search, but it's like a toggle, so you have to find it, which is, like, always the problem with Google. Well, I mean, they still wound up in the a tier according to Dylan Patel and, and Jordan Schneider over at Chinatalk. Obviously, v o three was, like, a huge one, and then they also have all those, like, priced performance things.
Speaker 3:But I I've heard this narrative that, like, maybe some of the hyperscalers are super focused on benchmarking and and not even hacking the benchmarks necessarily, but just, like, just thinking about them. And a lot of the frontier labs, the independent labs have just kind of moved on philosophically from caring about benchmarks. Is that the right move? What's driving that? Like, is it is it are are we in, like, the post benchmark era, essentially?
Speaker 2:Yeah. When I when I think about models and benchmarks a lot, I I think, like, which models outperform the benchmarks?
Speaker 3:You know?
Speaker 2:When you see o three's benchmarks, they're good. They're kind of what you expect. Yeah. Then when you watch o three Think, you see this model is actually reasoning.
Speaker 4:Sure.
Speaker 2:When you watch Cloud four Opus or Cloud four Sonnet Think, it's like, woah. This is really good. Same with GPT 4.5. I think the Gemini models are good, but they're exactly as good as the Benchmarks let on. You know?
Speaker 2:Sure. And I I I think they don't have the vibes yet. What I wanna see is Gemini 2.5 Ultra. If Google releases something with some big model smell, something really cool, maybe maybe that's them.
Speaker 3:What is what is the big model smell? I just don't like the idea of smell at all.
Speaker 2:Oh, yeah.
Speaker 3:It's just a weird it's a weird sense. My view is It's just it's the vibe check. Wait. Who who who coined it?
Speaker 2:I think it was Aiden McLaughlin.
Speaker 1:I don't know.
Speaker 3:That's great. But but but basically, we're we're in, like, the intangible period. Is that is that the
Speaker 2:idea that's unquantifiable? I I think Anthropics given up on really training on the benchmarks, and I think it's gone really well for them. You know? You see that they're really good at sweep bench. They're not crushing it on MMLU.
Speaker 2:You know?
Speaker 3:Yeah. But
Speaker 2:if you tried Force on it, it's great. You know? Other labs that are lower down on this tier list seem
Speaker 4:to have
Speaker 2:seem to have not given up on doing really well on the benchmarks.
Speaker 3:Yes. Yes. That makes sense. I mean, it's it's possible that, like, you you must defeat the final boss to, like, play the endgame. Yeah.
Speaker 3:And so maybe the endgame is this vibe check, this big model smell, but the the in the interim, like, yes. Like, if you're not you only earn the right to go into big model smell if you can dominate in all the benchmarks.
Speaker 2:There was an interesting moment where three point seven was beaten on every benchmark. Yeah. So now now for state of the art on stuff again. Yeah. Three point seven was losing on everything.
Speaker 2:There was a better model for everything hypothetically. But then if you looked at what what you might call, like, revealed preferences bench, which is just like, what do people use on cursor?
Speaker 4:Sure.
Speaker 9:What do people
Speaker 2:what what's going on Vamp?
Speaker 1:Yeah. Revealed preferencesbench.com.
Speaker 5:Yeah. Yeah. Should see someone's buy that.
Speaker 8:Was great.
Speaker 2:Yeah. Yeah. 3.7 was Yeah. Was pretty high up there. You know?
Speaker 2:So it seemed like they had something that that wasn't captured there.
Speaker 3:What about cornered resources? Data is the new oil. That seemed like a very silly concept in the moment when everyone had scraped the web entirely, and there was it really felt like data was fully commoditized. Then we see v o three. And for the first time, it feels like, okay.
Speaker 3:There is at least one dataset that is so large that you can't copy it onto a single hard drive or compress it, and it's YouTube and Google owns it. And and, yes, people might scrape it here and there, but Google has a durable advantage there. But is that is that the wrong way of thinking about it?
Speaker 2:Yeah. I mean, I'm not sure about the video models. I I I think it's true that data is like both super super important, but also has just become like tremendously overrated. Because the first people thing people learned about AI is like, oh, it's a result of the data that goes in. But now that we're unlocking things like RL, embedded post training, it seems to me like you can you can have some non data solutions to some of these problems.
Speaker 3:Yeah. Mean, that was the original, what, generative adversarial network for image generation was like synthetic data generation and then and then, and then testing it. And so, like, I'm I I it just v o three feels so so much like a beneficiary of YouTube.
Speaker 2:Yeah.
Speaker 3:But I don't know if that's just if we're just waiting and we'll see the next Sora, and we'll be like, oh, OpenAI figured it out. And, like, yeah, maybe they found some, like, you know, kinda workaround to the data, but, really, like, the the vast majority of the consistency and the innovation there was algorithmic progress, not just, you know, core to resource and data.
Speaker 2:Yeah. One thing about video models is it's been so secondary, but they've become so impressive. I think that if you showed them both to me a couple years ago, I would be more impressed by v o three than even, like, Cloud Force on it or something.
Speaker 4:You know?
Speaker 2:It's just it's not what I it's it's really, really just incredible.
Speaker 3:Well, yeah. I mean, I I think a lot of it just comes down to, like, the cost of instantiating the thing.
Speaker 4:And
Speaker 3:so if I go to if I go to deep research and I use o three and I have it pulled together, some, you know, twenty minute research paper, it's like, that's a few hours of work. Maybe it's a few thousand dollars of, like, a researcher's time. Maybe we're getting up into, like, PhD level. Do it on your own.
Speaker 4:I could
Speaker 3:do it on my own. But but if I actually want to crash a Ferrari through the Hollywood sign with champagne bottles flying
Speaker 1:brand name.
Speaker 3:A custom Hollywood sign that's huge, like, unless I'm See if it's I'm either doing actors chasing us. I'm either I'm either renting all that, shooting it practically, and it's a multimillion dollar Michael Bay shoot, or I'm doing it all in CGI. And even to do it in CGI is millions of dollars of rendering. And so even for an eight second clip, it just looks like, wow. I got something that normally would cost a million dollars to make happen.
Speaker 3:And there's no there's no real, like, textual asset that feels like, wow. This is a million bucks worth of assets. Anyway, interesting. X AI, they are cooking. They've been obviously GPU rich, scaling up.
Speaker 3:People seem like they're in the b tier here, according to this chart, but everyone's kind of excited about what's coming next. What is your take on Grok, XAI? Are they close to the big model smell? Is that feels like a natural beneficiary of Elon's strategy of just go big. But how are you thinking about Grok generally?
Speaker 2:Yeah. I I'm not the most impressed yet. Mhmm. I mean, Grok three is good. It's a good model.
Speaker 3:Sure.
Speaker 2:It's like a funny thing, like, Grock's whole thing or or something that people who really like Grock often say is like, oh, it's it's trained on this real time x data. It's this x availability. One thing I've tried a few times, because I saw it in a tweet Sure.
Speaker 7:Is it if you
Speaker 2:have a tweet you can describe, maybe I say John Coogan's tweet about bringing media back to Hollywood. Yes. And you ask Rock to find it, it can't find it. You ask to go tweet to find it, it can find
Speaker 4:it.
Speaker 3:Wait. Really? Yeah. L three can find it. That's so interesting because I feel like I feel like x is pretty locked down at the at, like, just the w w w layer.
Speaker 3:Right? It's pretty hard to find. In fact, a lot of times I'll post in a post from x, and it will have to go to, like, thread reader unroll and find an archive off of x because it clearly can't access it directly. Here's here's But that is fascinating. Is fascinating.
Speaker 3:So That feels that that feels solvable.
Speaker 1:Adam. Adam ships t b p n guest dot com. Yep. Like, last week Yep. I had a friend find it Monday.
Speaker 1:We hadn't announced it anywhere. It's not even visible on the Google search. Really? And o three found it.
Speaker 3:Wow.
Speaker 1:And it's like, how did how did you find this? And he was just looking he asked o three, can you pull together a list of all the guests that we can And
Speaker 4:it found
Speaker 1:that link Oh. Randomly. And Google doesn't even find it.
Speaker 3:Interesting. On
Speaker 2:top good at search. And I think I think that might have been OL. They they mentioned OL ing on tool use Sure. In the blog.
Speaker 3:Very, very interesting. Also,
Speaker 2:like x AI, it's like not really much revenue, nearly no revenue yet. You know, at some point, you need to start pulling that out. I'm glad they're they're pushing on the distribution, you know, but
Speaker 3:Yeah.
Speaker 2:Things come around.
Speaker 3:Yeah. Makes a lot of sense. Last one will end with
Speaker 1:the highest revenue multiple of any company in history. Yeah?
Speaker 3:Yeah. Last one will end on Meta Llama sitting in d tier, but maybe not out of the game yet. The two interesting bull cases I've been discussing have been, one, is there a world where, open source American AI becomes geopolitically important for countries that are slight allies, and they're either choosing between deep seek or an open source, American model. And the and OpenAI would not be in the conversation. And then also just, you know, why would you ever bet against Zuck?
Speaker 3:He has a capital cannon that can fire 10,000,000,000 at random projects forever. And the question is, is that enough? What are you looking for from Meta and Llama in the future?
Speaker 2:Yeah. It seems like you hit some some issues recently, but I I I'm not betting Gensuk. He's got the capital. He's got some GPUs. Mhmm.
Speaker 2:They can get together some really great research. I would love to see better American open source models. I mean, I'm not betting on open source in the long term as maybe the cornerstone of AI. Mhmm. But the fact that all of our American research groups, lots of really smart RL researchers are doing experiments on Quen and not on Lambda is not great.
Speaker 2:You know?
Speaker 3:Yeah. Yeah. Yeah. So so should there there's one interesting twist there, which is Quen has so many different models. Lama has a few.
Speaker 3:They're still working on rolling out Behemoth. But, would it would it be, like, almost more of an olive branch to the developer community to fragment the models and and and really focus on hitting researchers? Is that kind of a potential path that they should take?
Speaker 2:Yeah. I mean, I think it would be really cool if they did that. It would be somewhat charitable.
Speaker 3:Yeah. Yeah. Yeah. Exactly. Developers love handout, but, know?
Speaker 2:I don't know. I I think I'm curious about what they do on the product level and how they can build stuff in better. Mhmm. On the product level, people aren't incredibly sensitive to whether o three can search 50,000 websites like we are, you know? They they they care more about just having something that's really good, something that's really good to talk to.
Speaker 2:Maybe meta shifts focus that anymore. I'm not feeling it right now in terms of like when will a meta model grab number one on Amorino or something? It seems like it's it's gonna be some time, you know, but I'm not counting them out at all either.
Speaker 3:Yeah. I mean, if they can just, yeah, stay on the the lagging edge, that could still be valuable in a lot of their product rollouts. I mean, we forgot Apple in the L tier. We we do have another guest hopping on in just a minute. But Apple in the l tier, how do they dig themselves out?
Speaker 3:Is it build? Is it buy? What do you think is gonna happen? They could maybe they have a lot of cash. They could maybe buy someone.
Speaker 2:They could buy someone.
Speaker 3:Yeah.
Speaker 2:You can get buy buy lab, and then then you gotta Yeah. You gotta upgrade. I I there was a there was some report that they had some internal models. Yep. I wouldn't be surprised if they could train stuff.
Speaker 2:It's just, look, we haven't seen anything at all.
Speaker 3:You know? Like Do you think they're really training on Apple Silicon? Like, you you've seen those photos of, like, all the Mac minis wired together. Does that seem like something that's really just like AI general? Okay.
Speaker 2:Yeah. I think non GPU training ones are gonna be bigger next year. Really? I think the GPUs for Google too. Sure.
Speaker 3:Yeah. So they already have a long time with TSMC. They could go do something like a trainium or an infra chip from Amazon or TPU.
Speaker 2:Yeah. I mean, with the TPUs, Google has by far the most compute.
Speaker 3:Yeah. I mean, I guess Apple's pretty good at chip development and design. So, like Yeah.
Speaker 2:They could do it. Chip's pretty good.
Speaker 3:That would be their yeah. That would be their their their advantage if they could build a really strong chip and cut that cost.
Speaker 5:I wouldn't
Speaker 1:bet on it. But
Speaker 5:Yeah. Maybe.
Speaker 3:Yeah. Yeah. I I like the idea of just opening it up and really partnering.
Speaker 1:The the thing over the last twenty four hours is one account sharing. It's so over for Google. And and then immediately sharing, wow, Google's gonna destroy everyone in AI and just like seeing seeing how the the post rank.
Speaker 3:Yeah. Yeah. Anyway, anyone else on here? They got Mistral in f tier. Poor for the French.
Speaker 2:Yeah. They're not trying Le Chat.
Speaker 3:Yeah. You gotta be I I I do wonder about Mr. All because, you know, the the models are real, but none have, like, broken out in capability. But there there's this question of, like, if you want a national champion in your country, it might not be enough to just have the foundation model layer. You also might have to go and win in the free market in the application layer.
Speaker 3:And so, yeah, you could have even if you had a comparable model, if you're not if that's not if people aren't if people are going to do chat.com instead of laychat.com, like, you have not won and you don't have your national champion.
Speaker 2:Yeah. And there's a I think there's some truth to this, but there's also the the regulatory stuff in the EU. I mean, lot of releases I think v o three is not in the EU. Mhmm.
Speaker 3:A lot
Speaker 2:of releases don't come there. Mhmm. Maybe Mistral just uses uses regulatory motes to monopolize. Not not a fun way to win, but maybe that's the bull case at this point.
Speaker 3:Yeah. What what what was your reaction to, the conversation back and forth with, Dwarkash and Shalto all about, about, the the the the this debate over over over, I forget. It was like spiky intelligence and how you actually, train someone. There's so many different things. We see that the models are really good at one thing and then they fail RKGI.
Speaker 3:What's your overall timeline right now? How are you looking?
Speaker 2:Yeah. Dole, Kash raised the point that you can't kind of do this continuous learning, this like short run continuous learning. Like, you can tell me, Jack, I want you to
Speaker 5:do something different as she can
Speaker 2:get an intern. Yeah. I figure that out. And and context is a weaker tool than that. And I think that's absolutely true and that's an unsolved problem.
Speaker 2:I don't know how much that moves my needles on timelines. Like, one thing that could be true is just that OpenAI or Anthropic makes some, like, sweet agent and they Mhmm. Starts accelerating their AI research and they just get, like, really efficient algorithms really quickly. Some Yeah. Architecture that just destroys the transformer.
Speaker 2:Yeah. But I do think it's a meaningful unlock if you could if that could be solved. And I think that sort of like mid level memory type of stuff is really interesting. Yeah. Great.
Speaker 2:Or or solutions around context around VapR.
Speaker 3:Well, is fantastic. We have our next guest. So much for hopping on. Good to
Speaker 4:come on.
Speaker 3:We an in person guest. For sure. Thank you so much. Next up, we're we're we're heading over to Google World. We have Arush from Google.
Speaker 3:He worked on the deep research project that dropped from Google in 2024. It was a full year ago. It was in December, technically. But, very excited to to talk to him about, that product, all the things that go into deep research. So we'll welcome him to the studio if he's available.
Speaker 3:How you doing? Good to
Speaker 4:have you here.
Speaker 9:What's up, guys?
Speaker 10:Thanks
Speaker 9:for having me Not
Speaker 3:too much. We're having a great day. We got a great lineup, and, excited to dig into it. Would you mind kicking us off with just an introduction on yourself and, and a little bit of the history? I I wanna hear about the history of of the products that you've built at Google, what the interaction between, research and product looks like, and what you're excited about.
Speaker 9:Yeah. For sure. First off, a team. That's pretty good.
Speaker 3:Pretty good. Yeah.
Speaker 4:Hell yeah. Let's hear it from here.
Speaker 1:Let's go. Let's go. Let's let's
Speaker 10:hit it.
Speaker 4:Yeah. John's gonna have to go.
Speaker 8:Good work.
Speaker 3:Cool.
Speaker 9:Yeah. No. Love love to be here there. A lot of yeah. It's been fun.
Speaker 9:It's been a fun ride. I so I'm a product manager on the Gemini team.
Speaker 3:Cool.
Speaker 9:I've been here since a little while back when it was called Bard.
Speaker 1:The Bard days.
Speaker 9:And Bard days. Yeah. And so, yeah, about, I don't know, maybe like this time last year, we started kicking around this idea of deep research. Mhmm. Where one of the things we noticed is a ton of people come to the product and ask, like seeking to learn something or asking questions and kind of doing research y type type things.
Speaker 9:But if you ask really hard questions, one thing we noticed is the model would just give you like an outline of an answer. It wouldn't actually tell you Yeah. Something very comprehensive. So we kind of just ran with a hypothesis of like, let's take off the constraints of like, it has to respond within a few seconds. It has to use this much compute.
Speaker 9:Like, let's let it let's just see how far we can push what the model can do. And this was before thinking models or anything, and then, like, kind of and any of that that good stuff. And so we kind of worked on this idea for a bit, and then we launched in December back on Gemini 1.5 Pro. It was the model that we were using back then. We launched Deep Research as kind of as a bet to just see, like, would people be into something that makes you wait fifteen minutes but gives you something comprehensive.
Speaker 3:I'm happy to wait, although I do want to speed up. Questions about, context window size. How important is that million token context window? That feels like it's been a unique Google feature for even longer than I expected that. The advantages in AI seem to last days, maybe weeks, before another model comes out that that, you know, meets or is roughly around the same capability.
Speaker 3:How important is large token context windows in, in deep research like products?
Speaker 9:Yeah. It's huge. It's it's it's, like, really what enabled us and kind of gave us the confidence that this was even worth trying. Yeah. I'd say that the long context enabled us to do basically, be very recall forward and really cost a very wide net as we research the web
Speaker 4:Mhmm.
Speaker 9:And try and find gems of information that that we then stitch together. And so that that was, like, I think our biggest differentiator and and really allowed us to build this product. The other thing that long context allows us is, like, once you've finished your research, not just the report, but everything it read along the way is in context. So you can keep asking questions going deeper with the with within Gemini. And even if it's, like, a tidbit of a fact that's not in your report, if if it's in if it's been read at some point Yeah.
Speaker 9:It'll be able to retrieve that and and give you that answer. So so it also helps sort of beyond that first turn, keeping a good experience. Yeah. And then reasoning models was, like, the next big big step jump for us Yeah. Allowing it to then do more critical analysis.
Speaker 3:So in terms of, like, actual product design, I'm interested in in the direction this goes. Here, you could see one world where the models are baked down into silicon. Everything's running even faster. You're distilling the models, and all of a sudden, I'm getting a deep a a twenty minute product in two minutes or even twenty seconds. You could also imagine a world where what's possible if the economics work such that I could request a two hour research report or a two day research report.
Speaker 3:How are you evaluating those? What would you personally be more excited about, and what do you think users actually want? Because stated preferences and revealed preferences are are often different, or do we wind up with both?
Speaker 9:Yeah. So one of the things that we noticed, one, when we launched this, we had no idea people would be willing to wait. Like, every metric at Google from the day it started is reduced latency and, like, all metrics go up.
Speaker 3:Yep.
Speaker 9:Right? So this was definitely a bet where we were like, a lot
Speaker 8:of people thought we were
Speaker 9:crazy, where we're like, it's we're just gonna take a ton of time and people will wait. Yeah. One thing we noticed is that, like, after about a minute or something like that, people are fine. Like, people will go away, do other things, come back. We'll send them a notification when it's ready.
Speaker 9:So the big pleasant surprise for us is, like, people don't mind waiting. I'd say so in terms of, like, efficiencies gains, one of the things that we're more excited about is, like, okay. If we can make models more efficient, instead of reducing down the research time, can I give you just a way better output? Like, can I use can I bank that savings and give you something way more insightful, way higher quality? Yeah.
Speaker 9:I'd say the other thing is, like, even if I could give you, like, a a deep research answer in fifteen seconds, it's gonna take you fifteen minutes to read. So there's also an aspect of just, like, how much do you want to consume this. Right?
Speaker 4:Yeah.
Speaker 9:So so for us, we're not as stressed about, like, can we make this faster? Can we make this quicker? I do think there are probably other points in the, like, latency comprehensiveness spectrum that people might like. Right? We picked, like, one extreme of, like, let's just go super hard and and build the most comprehensive long, you know, thing that takes a while.
Speaker 9:Yeah. But there might be totally other points people are interested.
Speaker 3:Yeah. Yeah. Sometimes I noticed I've generated, like, so many various deep research reports across all the different apps that I'll, like, I'll follow it up with a prompt like, okay. Like, yeah. Boil that down for, like, 10 bullet points because, like, I don't have time to read that.
Speaker 3:And then I'm like, wait. Like, maybe I should have just asked it to give me 10 bullet points in the and I just, like, burned a bunch of GPU cycles. But I get I guess the question
Speaker 1:is and forth between the two until you kind of understand the Exactly.
Speaker 3:But but I guess the question is, like, is is there is there a product or is the natural evolution of just general prompts that as as algorithms get faster, as these models run faster, that there is a deep research amount of work that happens, within a few seconds, between every response? And and, basically, the question is, like, how much can you port from the deep research product and strategy and design back into just your average LLM interaction?
Speaker 9:Yeah. I think there's definitely a lot of learnings that we can kind of start upstreaming really around, like, being able to form a plan, follow that plan to do that sort of multi hop steps of, like, search iterating, like, finding insights, changing your strategy possible before going back to the user. And so you're you're kind of starting to see this in, like, 2.5 pro and stuff like that. And I you can imagine that that will continue where you will see more, like, mini deep research or more sort of, like, planning and and sort of, like, iterative reasoning before, like, giving you an answer. And, yeah, as that could just start getting faster and faster and faster, then you start just getting, like, way more insightful or comprehensive answers.
Speaker 3:There any other interesting areas? I mean, deep research feels like one of the first, like, really solid product market fit experiences in, I guess, like, agents broadly. Are there any other areas that you're excited to think about knocking down with either different products or just maybe un, just like cool uses that you've or developed or as a user, patterns that you're leveraging that's maybe go beyond just the average, like, need a I need a a research report.
Speaker 9:Yeah. Totally. So I think there's, like, a few different angles that that, like, I think a lot of people are exploring. Mhmm. One is you kind of point out, like, what does a two hour deep research look like?
Speaker 9:What does an overnight deep research look like? Yeah. If you can have, like, a a very well defined problem Mhmm. Where, like you know, we have early experiments at Google, like AI coscientists and stuff. Right?
Speaker 9:Like, you could run that overnight and it can come up with, like, novel scientific hypotheses. Right? So there definitely is an angle of, like, if you can define a problem and an outcome really well
Speaker 6:Mhmm.
Speaker 9:Applying more compute can actually get you, like, better and better answers. Right? So there's definitely an angle of, like, other whole new classes of problems where you can even go even further with deep research. There's a second aspect of, like, you you know, we had the chance to go, like, meet a bunch of people who are, like, researchers at the Fed. Right?
Speaker 9:And they were telling us how they research. And it's often, like, a very different thing. Right? So, like, I showed them this example where I was like, hey. There's this, like, funny law in The US called the Jones Act where, like, any two ships between, like, two US ports have to be, like, built in America, crewed by Americans.
Speaker 9:Yeah. And, like, drives up shipping prices, but only for, like, Puerto Rico, Hawaii, and, like, Alaska. Right? Yeah. And so I was, like, do an economic analysis of the Jones Act on on, like, the economy of Hawaii.
Speaker 9:Right? And it, like, did a first principle analysis, did some really interesting things, like, looking at well, like, how much is a three three and a half thousand shipping route, like, say, from, like, Mexico to South America, and then that's, like, a baseline price to compare against. And, like, I thought this was amazing, but then they were like, that's not how we would do economic analysis. Like, they would be like, first, I'd explore, like, what other studies there are. Like, then I'd explore, like, what kinds of methodologies are out there.
Speaker 9:Then I might, like, ask a fast bunch of fast follow-up questions about, like, what data sources or, like, data sets did people use to do this research. Right? So there's definitely an aspect of, like, another angle of, if I really wanna help people with research, it's about, like, nailing this sort of, like, synchronous, asynchronous paradigm
Speaker 4:Mhmm.
Speaker 9:And helping people kind of do more of that, like, iterative process rather than just, like, ask a question, get answer, and and move on and and and and victory. And I think that's that's kind of a product challenge, like, figuring out the right the right interaction model
Speaker 4:for that.
Speaker 9:Yeah. And the third is is just, like, outputting the outputting an answer at the right, like, level of abstraction that you work at. Right? Like, a financial analyst doesn't think in terms of report. Right?
Speaker 9:They think in terms of, like, the spreadsheet or or the financial model. Right? And so if I want a DCF, deep research can build, a great DC discounted cash flow model for But, like, I don't want it in a report. I want it in a spreadsheet or I want it in an app where I can play with the variables and see the different outcomes. And so you'll also see the line between, like, reports and other kinds of artifacts starting to blur
Speaker 4:Mhmm.
Speaker 9:Or even just, like, what, like, what does it mean to, like, build an build an answer. Right? And and that that could take, like, a much wider
Speaker 3:That's super exciting. Yeah. I mean, I've seen, obviously, Gemini. We probably can't talk about the road map too much, but I've seen Gemini pop up in a bunch of different areas. And and I haven't seen Yeah.
Speaker 11:That that leads
Speaker 3:to question version of whatever that instantiation is.
Speaker 1:Yeah. My my maybe my last question is, like, how much time are you thinking about working and making the you know, as a product manager on Gemini, how how much time are you thinking about making Gemini better versus sort of fighting for distribution Mhmm. Outside of Gemini and a kind of across the Google ecosystem because part of unlocking the value of Gemini is just making sure it's in the right places and and place sort of contextually across, you know, everything from, you know, gemini.google.
Speaker 3:You've worked hard on this. Just ask for the I'm feeling lucky button. Just give us that. Just like
Speaker 1:think you we think you earned it.
Speaker 3:We you've earned it. It's a great product. Just click I'm feeling lucky or burn forty forty GPU hours on this new
Speaker 4:Yeah.
Speaker 3:Research award.
Speaker 9:Putting yeah. That that would, like, instantly melt all of our servers everywhere.
Speaker 3:No. But this is this is the biggest hyperscale. Like, yeah, we need more GPUs. Let's see. TPUs.
Speaker 3:TPUs. TPUs. Yeah. TPUs. Okay.
Speaker 3:So ASML, GetCooked.
Speaker 1:I believe you've earned that. I'm feeling lucky. I haven't hit the I'm feeling lucky button in years. Yeah. Yet I use Gemini all the time.
Speaker 1:Yeah. Yeah. Give me get the use this is what the users want.
Speaker 3:Yeah. We just need 10 more TSMCs, I guess, to start fabbing. Anyway, sorry.
Speaker 1:Serious answer.
Speaker 9:Yeah. No. The serious answer is, like, the Gemini app is, a great place for us to, prototype. See what, hits with like, really works with people. A lot of the users, they're very intentional when they come into the Gemini app.
Speaker 9:Like, they wanna use an AI experience. Mhmm. So it's a really great place for us to, like, put stuff out there, see what works, see what doesn't. Some things we put out needs more time in the oven. And then over time, you'd imagine that then, like, those insights or things that really start work, you'll start seeing in other Google products as they make sense.
Speaker 9:Right? You don't wanna like over clutter a UI, but you'll start seeing, yeah, things like deep research
Speaker 1:or Yeah. Because it's a very it's a very different user. Somebody that's coming in saying, I want AI versus I just want to do certain things.
Speaker 2:Yeah.
Speaker 1:Right. And yeah, they're totally different archetypes.
Speaker 3:It's a fascinating challenge. I'm sure it's even more challenging at your scale, but thanks for all the hard work and and pushing the frontier forward. It's been a pleasure talking to you.
Speaker 1:Yeah. Come back on again soon.
Speaker 3:Yeah. We'd love to talk
Speaker 4:to more.
Speaker 9:It. Thanks so much, guys.
Speaker 3:We'll talk to you soon. Bye. Fantastic. Next up, we have Oliver Cameron. I have a I have a good story.
Speaker 3:We'll bring him into the studio. But I believe he was the first person I ever interviewed for a YouTube video years ago. I was doing a whole video essay about Cruise, the self driving car company, and he hopped on a Zoom call with me just like this one. And I recorded it and threw clips in the video. Was very fun.
Speaker 3:And then I wound up doing more interviews after that. So, Oliver, good to see you. How are you doing? What's going on? Welcome.
Speaker 12:I'm doing great. Thank you, for the opportunity.
Speaker 3:Would you mind kicking us off with, like, the latest and greatest introduction? Because you've done a lot in your career, but you're onto something new.
Speaker 12:For sure. So I spent about eight years building self driving cars. Incredible time. Mean, just to see that technology go from barely being able to keep in a straight line to navigating Downtown San Francisco with no human behind the wheel. Just a a sign of where things have gone Mhmm.
Speaker 12:With machine learning. So I had a had a blast doing that. Built my own company, sold that company to Cruise where we met, and and loved that time. Left Cruise in May of twenty twenty three. Mhmm.
Speaker 12:Decided to start something new, And both me and my cofounder, who also was from Self Driving Cars, we were both very much inspired by Pixar. I think it's just a very special company. Right? Everyone kinda recognizes Pixar as this sort of iconic storytelling company, And we really put our heads together to think about what a modern reincarnation of Pixar would look like. So that company is called Odyssey, and we're an AI lab that's really focused on, enabling entirely new stories, to be told.
Speaker 3:Mhmm. And, walk us through the first product that you launched. I played with it earlier. It it was mind blowing. We'll we'll pull it up while you're talking.
Speaker 12:Sure. Yeah. We just released a research preview of something that we call interactive video.
Speaker 10:Mhmm.
Speaker 12:And it's effectively AI video that you can both watch and interact with in real time. Yeah. And we think this will become a entirely new form of entertainment. You know, you've got film, you've got games, you've got all these these mediums that have been around for a while. We think that there is an opportunity to invent a brand new one where effectively a model is responsible for imagining film and game like experiences in real time
Speaker 3:Yeah.
Speaker 12:That you can interact with. There's no game engine behind all of this. No heuristics. No rules. Just a model that's learned pixels and actions from tons and tons and tons of, real life video.
Speaker 3:Yeah. We're showing it on the screen right now, and the production team is controlling it with the keyboard, W A S D, like it's a first person video game. And they're walking around this field, with trees and windmills, and they can actually choose to go up, go inside buildings, and it's all being generated without the use of a game engine, and then they can switch over to a different environment. And so, I mean, I have tons tons of questions about how these different like, you're not doing photo scanning. You're not doing, you're not doing game engine stuff, traditional three d pipeline, but the data must come from somewhere.
Speaker 3:Love to hear about that. And then also, I noticed the space button doesn't work. I wanted to jump around, stop start bunny hopping. When are we getting a space button added to this thing? Anyways.
Speaker 3:Anyway, sorry.
Speaker 12:Isn't it trippy how that those pixels are literally streaming from a GPU cluster probably in Texas.
Speaker 3:It's so crazy.
Speaker 12:And now we're streaming them via Zoom Yeah. In real real time.
Speaker 5:It's crazy.
Speaker 1:My my question is do you think that Odyssey can be a really breakout app for VR? Because when I see that
Speaker 3:visual Oh, sure.
Speaker 1:I I feel like that it it could give someone the sense of being able to explore lands that don't exist, which is like very fat like, once it's fully immersive Yeah. It feels like
Speaker 3:It's funny, the windmill thing, because I remember the very first Oculus demo that I ever did, I was walking around a windmill, and it and it's still in my mind years later. But, and it was amazing. But it was just like one little windmill, and then you couldn't go any further because developing, like, virtual assets is really expensive. And so you play a lot of these VR games and, you know, it's a couple hours or thirty minutes. But if you take a procedural approach or a generative approach, you all of a sudden have infinite content.
Speaker 12:I think what's really important to note is in film and game, incredible things can be made. Right? Yeah. Like, insanely good things that wow us all. The time and the money it takes to create those things is ludicrous, and it's only getting more expensive, not less expensive over time.
Speaker 12:Yeah. So I feel there will be continuously a place for these sorts of, like, handcrafted things, and and they'll be very important. But if we just think about a model that's trained on literally decades of of video that's then able to imagine stuff in real time with no preproduction cost, no postproduction cost, and and do that in literally in real time, like thirty three milliseconds. It just that that's where it gets really crazy. And and what we showed in the research preview is just like this tiny glimpse, I think, of what this stuff will become.
Speaker 12:VR in particular is, like, the most hardcore application of this Yeah. From a a technical perspective because the resolution required for VR is, like, insane, and the resolution that you saw there, you can tell it's low res. It's, like, 300 pixels wide. So there's gonna be a leap that needs to happen there to get to to VR level res. But
Speaker 1:I'm confident that Odyssey two, you'll have it you'll have it
Speaker 3:You'll have it dialed?
Speaker 1:Oh, yeah.
Speaker 3:28. Give us the stats. How many, like like, what numbers can you give us about the progress or adoption? You just launched this, I think, this week or last week. It hasn't been very long, but how have how how's the response been quantitatively?
Speaker 12:Oh, it's been incredible. So we launched a week ago, and since then, we've served 250,000 unique streams, meaning 250,000 people experiencing what you just saw, which is insane.
Speaker 3:Market clearing order inbound.
Speaker 1:Yeah. Let's do it. Love
Speaker 3:it. Congratulations. That's fantastic. On
Speaker 9:on on
Speaker 3:the question of resolution, there's a bunch of amazing AI upresing that's happening in in various parts of the pipeline. There's some server based upresing that can happen. There's some on device upresing. So is that, are you are you counting on that technology breaking one way or another? Does it matter?
Speaker 3:Will it be a combination of both? How do you see that developing?
Speaker 12:I think the way to think of this is where video models were a year ago is where real time video models or world models will be today. Yeah. And what that really means is that you look at the res remember the Wills everyone remembers the Will Smith spiky video.
Speaker 3:Right? That was wait. Wait. Was that, like, one year ago or two years ago? It wasn't long ago.
Speaker 12:I think it was just over a year ago.
Speaker 3:So fast.
Speaker 12:There was definitely better outputs. Like, spaghetti was, like, the the weirdest, hottest thing at the time. Although gymnastics today, I'm sure you've seen that. That that's really tough with video models today.
Speaker 9:That is tough.
Speaker 12:That's all to say that I think the res and the quality visual quality improvements will come from the model itself, not like some secondary piece of Interesting. Infrastructure to up res.
Speaker 3:Sure.
Speaker 12:Just because, I mean, think of what a language model was like to use two years ago. Like, how fast was it in response time? Really quite slow, right, compared to today where it's like just stream of information straight to your eyeballs. Same will be true of these models. Like, we'll crank out larger resolutions, faster frame rates, more actions, more things you can do, all that sort of stuff.
Speaker 3:Yeah. And I guess I guess importantly, like, GPT 4.5 is not GPT four up res to four five. It is a different model. We're we're we're walking around what looks like the gloomy English countryside right now, and I think the production team is gonna try and go in that house. It it it is is is really, really so wild.
Speaker 3:I noticed that there's a there's a time limit.
Speaker 1:Tropical island demo because this this this I I love the
Speaker 2:English country,
Speaker 3:but foggy. Yeah. I I noticed that there's, a two minute timer when I sign in. Is that so the GPUs don't melt? I mean, I should I assume you've raised money, and you're maybe burning some money with these demos.
Speaker 3:But but break down kind of, like, what your limitations are and how you see them evolving.
Speaker 12:Yeah. For sure. So the, timer is there because each session is served by a single GPU.
Speaker 4:Mhmm.
Speaker 12:So, each user gets a GPU, the model's running there, and that's been to the user directly.
Speaker 3:Mhmm. And really, you when you say single GPU, you don't mean rack. You mean, like, one a 100 or something like that?
Speaker 12:One h two hundred
Speaker 3:h
Speaker 12:200. Per user.
Speaker 3:Got it.
Speaker 12:And there is a clear path to, like, dividing the GPU to have multiple sessions per user, but today it's one. And we wanna really crank up quality frame rate, all that sort of stuff.
Speaker 1:Yep. It makes me feel great to know that I'm getting, you know, the the sort of one on one attention
Speaker 3:From an h 200. Chip. Yeah. You know? Yeah.
Speaker 3:It's like It's an honor.
Speaker 1:You're at a retail store, it's it's not a great experience if somebody's bouncing around
Speaker 3:Exactly. I'm being individually served.
Speaker 1:Being served
Speaker 3:by the like an Hermes level
Speaker 12:By Jensen.
Speaker 3:Approach. Exactly. By Jensen.
Speaker 12:So $2 an hour is there or thereabouts how much that costs, which, you know, over the course of multiple users, it's not too bad.
Speaker 8:Yeah.
Speaker 12:I I think Netflix is, 5¢, 10 cents an hour, something like that to stream video. So we're we're a bit of a ways away, but you've got new chips coming, just model optimizations. Like, won't be long where we're having a single GPU per user, all that sort of stuff. For this launch, we had something like 360 h two hundreds prepared. We had to scale it up a little bit just because we had lots of demand, but that timer is there just to make sure we're cycling through lots of people getting getting a taste
Speaker 9:of this.
Speaker 12:But, yeah, I I think fundamentally the idea that you could have a model stream stuff to any screen is really powerful. Like, that experience you saw there works just as well on an iPhone, on an Android, on TV, anything like that. And it's all just action conditioned of a WebRTC, which is probably what Zoom is running on. So the action to just send over the wire to the model, the model then conditions the pixels it's about to generate based on those actions, sends the pixels back, and just that loop every thirty three milliseconds.
Speaker 3:It's firing. So, I mean, the the path to HD or four k seems pretty clear to me. What about the path to, consistency? That that feels really difficult. You need essentially, like, a really long context window to know that, okay.
Speaker 3:I dropped my mythical sword on in on on that piece of the ground. I went away, and then I came back. That's like textbook. Just put it in a database, but it seems like the future might not be that. So how are how are we thinking about that?
Speaker 3:Will the I I I guess the bigger question is, like, what's the response from the gaming community? Is this something that can be a tool and a piece of a pipeline instead of completely replacing the entire traditional pipeline?
Speaker 12:So most research on interactive video before has learned from games. So lots of folks will have seen, Oasis from Descartes and Minecraft in a video model Yep. Effectively, or Quake. That's often used in video models. And I think the gaming reaction to that is quite negative.
Speaker 12:Oh, yeah.
Speaker 7:It's like
Speaker 3:I mean, you saw the Carmack back and forth. Right? Where Carmack was like, this is amazing.
Speaker 12:I loved it.
Speaker 3:And somebody else was like, this is stealing, you know, developers. Yeah.
Speaker 12:And I I think it's important because the way that people envision that is like, oh, what's the best that this could become? It could become like remixing of games. Yeah. And that's one way it could be. I think people see what we have and they think, oh, this is like a world simulator eventually.
Speaker 12:This is The Matrix or like whatever they project on it.
Speaker 3:Yep.
Speaker 12:So really, thing we're trying to avoid is like for the first few generations of this, people will put, including ourselves, like this picture of what existing games look like onto this.
Speaker 3:Sure.
Speaker 12:And it's like the iPhone when it launched. Right? People ported desktop apps to the iPhone. And it kinda worked, but it kinda didn't. It wasn't really embracing this new medium.
Speaker 12:So I think the the long story short here is, like, stuff that is integral to games today, like multiplayer, like state, like scripting, all that sort of stuff. Let's question those assumptions. Like, how should those things work? Let's make it model native. Like, maybe memory in this model is very different than memory in in a in a game or state in a game, multiplayer in a game, all that sort of stuff.
Speaker 12:And that's probably gonna lead us in the short term to more like glitchy weird experiences throughout the the memory as it is by the model is a feature, not a bug. I don't know
Speaker 7:if you guys have
Speaker 12:seen like The Backrooms or like these kind of glitchy weird
Speaker 3:Yeah. Yeah. Yeah. It obviously is a completely different type of game design.
Speaker 12:Exactly.
Speaker 3:Yeah. The the the up, down, left, right, a b a b of the future will be like, drop your bad sword on the ground, walk around the building three times, come back, and it's and it's enchanted. Yeah. Because because the model hallucinates that you've upgraded or something like that. That'll be fun.
Speaker 12:I also think that one, important thing here is that in language models, one of the the things that's happened over the last year is in many cases, they've crossed this threshold of realism. For certain applications. So, like, people literally fall in love with language models. Right?
Speaker 3:Yeah. Yeah.
Speaker 12:The same, like, emotional feeling they have when they meet a person they fall in love with is happening for them with a language model. And that's because they what they're seeing on their screen is, like, so realistic.
Speaker 9:It's, like,
Speaker 12:crazy real to them. And I I think the same will be true here where once these pixels, these actions feel so realistic, which eventually they should just give them the data, give them the models and advancement, there'll be things that they do in these worlds or things they feel in these worlds, which they just can't feel in video games today because games are just capped by computer graphics and,
Speaker 3:like,
Speaker 12:human dev time and budgets and everything else. But they'll walk down the street, they'll see someone, and they'll be like, wow. That person looks so real. And they'll go over. They'll, like, high five that person all, like, on the screen.
Speaker 12:Right?
Speaker 8:Yeah.
Speaker 12:Yeah. And they'll just feel something. Like, they'll they'll feel, like, a heartbeat raise, you know, stuff like that.
Speaker 3:Totally.
Speaker 12:So that's that's an application that you can't do in games today. That's
Speaker 3:just Yeah.
Speaker 12:Different and new. So that's the sort of stuff we're really interested in.
Speaker 3:Well, that's gonna be a wild, wild future. But thank you. We'll have to have you back and check-in on progress.
Speaker 1:Yeah. It's
Speaker 3:fascinating. Definitely the day that seven twenty p drops or whatever the next version is. Music. We're we're excited for this. But thanks so much for joining.
Speaker 3:This was a fantastic conversation. We will talk to you soon. Cheers.
Speaker 1:Have a great rest your day. For joining.
Speaker 3:Thanks so much. Next up, we have a return guest, Michael McDonough from Lightspeed coming into the studio. Are are you gonna talk with the Gong?
Speaker 1:Or He's gonna talk about competition between Yeah.
Speaker 3:The Fanation Labs
Speaker 1:and the App Player.
Speaker 3:Well, welcome to the stream. Michael, how are you doing? Boom.
Speaker 13:Good. Good to see you guys. Congrats on the new studio.
Speaker 3:Thank you.
Speaker 1:Thank you. It's been
Speaker 3:a lot of fun.
Speaker 1:A bit of a
Speaker 13:I like the upgraded Gong too.
Speaker 3:Oh, yeah. The Gong's a much bigger. Everything's got bigger one
Speaker 1:in the works.
Speaker 3:Yeah. We're working on even bigger Gong. Florida's really? Oh, yeah.
Speaker 1:Also, funny day to just be so hyper fixated on AI because you probably haven't seen the timeline. Tesla's down 17%.
Speaker 3:Seventeen %?
Speaker 1:It's just absolute mayhem.
Speaker 3:I mean, there's an AI narrative there.
Speaker 1:Right? Yeah. There's definitely
Speaker 3:But that's not what's driving it.
Speaker 1:There. But anyways
Speaker 3:Blood back.
Speaker 1:Michael, it's great to have you on. Wanted to get some kind of updated thinking from you on the tension between labs Startups. Application layer. We saw the news with Windsurf and Anthropic today Yeah. That had more to do with a potential acquisition.
Speaker 3:And even when we talked to the founder of Granola, we were we were talking about the competition between Notion and Granola with these it's a founder led kind of previous era, scale up unicorn SaaS company. Can that company bolt on AI? But then now we're seeing competition from the Foundation Lab. So would love to get your lay of the land. What are you seeing?
Speaker 3:How are things shaking out? What do you think the next few months or even years look like?
Speaker 13:Yeah. It's pretty interesting. Right? Like, you think about the big companies that startups previously, built on the backs of the Googles, the Amazons, the Microsofts
Speaker 3:Yeah.
Speaker 13:You know, it it felt like there was this really healthy sort of symbiotic developer ecosystem where the incumbents supply resources, the developers buy and extract from them. And they build really, really big businesses on top. I think what we're seeing now, to your point, is these labs are building developer ecosystems, but then they're very intentionally and overtly going head to head with the developers that are building on them. And I think this has a lot to do with context. So if you think back to the internet and startups ten years ago, everyone said content is king.
Speaker 13:Content is king. Then distribution was king. It was all about how do you get in front of users. We're starting to feel like we're entering the phase of context being king. These models are just hungry for the most and the most unique context possible.
Speaker 13:And so if an app player company emerges and has a new type of context and data that the models don't have great exposure to, it's a great signal to to point in the direction and say, we're gonna compete head on. And so I I think that's what we're seeing now. And, yeah, Nabil Nabil Hyatt, a great investor from Spark and I, we often talk about how the war for context is happening now. And I think that's that's what a lot of these moves represent.
Speaker 1:How do you think app players should app app player companies should respond? Is it just double, triple down, go way way way deeper, focus on work workflows that the labs maybe don't have the resources to fully pursue? Or is it focusing down on specific niches? I'm curious what the
Speaker 4:Yeah.
Speaker 1:I think the right approach is.
Speaker 13:Well, we we we can definitely get into that. But maybe first, what I would say is, you know, I I I tweeted something yesterday that occurred to me after the big announcements from OpenAI in that, you know, the big incumbents, which we talked about a little while ago, sort of like the winners of the cloud era, it wouldn't surprise me if all of these new, you know, these new competitions between the labs and the apps actually drive the apps and the startups right back to the incumbents, to the Googles and the Amazons of the world. I have to wonder if some of these things actually act as a tailwind for models like Gemini And maybe give a little more credence to the argument that Google is actually gonna be the winner here because of all their distribution. So I think that's one potential.
Speaker 1:You mean driving back and being like, I'd rather work with Gemini because I don't think they're as likely to kill me?
Speaker 13:Exactly. Yeah. Exactly. It's like, hey. We trusted them with the cloud, and that worked out alright.
Speaker 13:Like, should we now trust them with AI more than we trust the labs?
Speaker 3:Yeah. I mean, the narrative even goes a little bit further with Microsoft, which has been completely like, oh, we will host every single model. We'll let you reroute really intelligently between them, like, super, super friendly developer ecosystem. And so, I mean, certainly, they're building stuff into Copilot into Microsoft three sixty five, but it does feel like they're they're they're much more willing to
Speaker 1:Satya Satya seems to have real conviction, you know, he had the quote from last week Yeah. Platform platform platform. Yep. And hosting deep seek is an example of that. Right?
Speaker 1:Yeah. A lot of people would have thought, oh, he's not necessarily gonna host that model because it felt like a shot across the bow at OpenAI. Sure. But he's committed to supporting open source.
Speaker 4:Yeah. Yeah.
Speaker 3:Yeah. He wants it all. Interesting.
Speaker 13:I I I think, you know, also going back to your question, Jordy, I I think all of this is just gonna make for a more intense, faster moving market. Like, I think more than ever before, you have to ship, you have to get users faster than anyone. You have to sort of like reach escape velocity quicker, which I think is just going to put more and more pressure on startups to move even quicker than they already are. I feel like Cursor is a great example. I feel like an earlier iteration of that product, you know, it probably would have been easy to sort of write them off and be like, oh, you know, a lab is gonna do this.
Speaker 13:I mean, now it's like they're so big, they're so far ahead. It feels like they've they've really established themselves and and likely have a good shot breaking through.
Speaker 3:I also wonder with Cursor and and Windsurf and Devon and some of the dev tools markets, like, feels like just such a new market that even if it's somewhat winner take all, there's just it's so positive sum because it's it's it's adding efficiency to the most like, one of the biggest labor pools. And so, when we talk to the Cognition folks, as the reaction to Google and OpenAI launching Devon competitors. They're like, well, we still grew 40% last month or something like that. And so, you know, I I wonder, like, in in code gen where it's such a new market that it's not it's not directly competitive with anything that exists, so it's less zero sum. I'm wondering if the note taking market feels similarly to you, or or were you seeing granola or other or other companies kind of act as more drop in replacements for existing tools?
Speaker 13:Yeah. I I think it's a great question. Yeah. So so we backed Granola really early on because we we knew Chris and his co founder and we loved those guys. We didn't know what they were building.
Speaker 13:We knew they were gonna something in note taking. But we said, you know what? This market's going to move fast. We trust these guys. Let's go for it.
Speaker 13:And and I think, you know, somewhat to your point, there's been all these note takers before. Like, Granolah wasn't first notetaker.
Speaker 3:Yeah.
Speaker 13:Fireflies and otter and all these things. I think granola has done a really, really good job of, you know, getting out of the user's way and establishing trust with the user. And I think that, you know, that that seems like a small thing, but I think that trust thing is gonna be really important if you go back to what I said about this context being king. Like, who are you gonna trust Mhmm. To take this context or take this, like, really, really important, proprietary part of your work, in this case, your meeting notes.
Speaker 13:You know, a lot of people say, we trust granola. Are they just going to hand it over to any old company that says, hey, now we want to screenshot your entire computer and take and suck every last piece of data out of you. And so I think part of it is to your point, like getting in early, getting big really, really fast and establishing that, you know, that user base and that market before it really matures, but also in a way that like users just really trust you and they're not just gonna rip you out just because some other bigger company offers the same thing.
Speaker 3:Yeah. Yeah. That's a good point. What it do you have any more, like, micro reactions to specific integrations? That seemed to be one of the big things that OpenAI was pushing on was integrations with Google Docs and Drive and and your email.
Speaker 3:And that feels like adding that extra context is is potentially the next thing people are clamoring for. How important is, like, the biz dev side of this business Yeah. In fact?
Speaker 13:I think it's really important. You know, I think it's really, really great, that Anthropic started the whole MCP protocol. Obviously, lots of others are adopting that now. But I think to your point, we're now going to start to see the battle lines being drawn. Like, who are who are we willing to integrate with?
Speaker 13:Who are we not willing to integrate with? Where is the are we open or are we closed?
Speaker 3:Yep.
Speaker 13:Where's the data going to go? Where's it not going to go? I think we're going to start to see those alliances and those allegiances form. And
Speaker 3:Yeah. I feel like these APIs, like, back in, what, like, thousand seven, '2 thousand '10 era.
Speaker 8:Social media.
Speaker 3:They have an API. It's amazing. It's like, well, like, you don't know how much that API is gonna cost. Like, if it's Exactly. $10,000 per day or something, like, that could completely upend your business.
Speaker 3:And so actually thinking about how that dynamic develops is is is is almost more important than the standard. Although, I'm very glad we have a standard. That seems great. But but each company is gonna have to decide where the value accrual really lands. And then who knows?
Speaker 3:Maybe it'll be some antitrust in in twenty years like we're seeing with Apple.
Speaker 1:Yeah. The big the big question around trust, that I I you know, it's a it's an evolving situation. But a California judge, I believe it was yesterday or the day before, ordered OpenAI to retain records of sort of, forget the what OpenAI calls it, but if you have like a a disappearing query
Speaker 4:Mhmm.
Speaker 1:A judge ordered them that they have to
Speaker 3:retain They
Speaker 1:obviously said that's a
Speaker 3:huge Yeah.
Speaker 13:Yeah. We don't privacy with users. So incognito mode.
Speaker 3:Yeah. Yeah.
Speaker 13:It's like not incognito.
Speaker 1:Yeah. Yeah. And that that that more seems like an issue with the court
Speaker 3:The government. Specific judge, you
Speaker 1:know, having this massive overreach around privacy. But privacy in this era when people are more willing than ever across every app to give them all sorts of data.
Speaker 3:Yeah. And and you have a direct incentive to to to reduce the level of privacy to get better results. Like, I if if if if the model knows what kind of car you drive when you ask it for new tires, it will give you better recommendations. So For sure. You wanna lean into being anti privacy to get better, better results.
Speaker 3:There's there's this the the the the world is definitely bifurcating into pro privacy or, like, fully AGI pilled folks. And and and there aren't that many people out there that are in the middle. So, obviously, we will have to figure it out as a democratic society, ultimately vote, and, hopefully sort it all out in the courts. But thank you so much for stopping by. Was
Speaker 1:Michael, thanks time, guys.
Speaker 3:We'd love to have you back.
Speaker 1:Talk to
Speaker 4:you soon.
Speaker 13:I mean, guys, I just wanna tell you, you know, I I don't really aspire to ring the New York Stock Exchange bell one day. I I I aspire to hit that gong.
Speaker 3:Hit that gong. Well, next time you're in Los Angeles, come by.
Speaker 1:Yeah. Come by.
Speaker 3:Alright. Bye. Gong.
Speaker 1:Your Michael.
Speaker 3:We'll talk to you Cheers.
Speaker 1:Great to
Speaker 3:see you.
Speaker 1:Bye. So we have a generational crash out going down
Speaker 3:Oh, really?
Speaker 1:Timeline. We got a new post from Elon. Okay. I'm gonna read it out. He says, and this is your live reaction, John.
Speaker 1:Time to drop the really big bomb. Real Donald Trump is in the Epstein files. That is the real reason they have not been made public. Have a nice day, DJT.
Speaker 3:Wow. That
Speaker 1:is a big that is a big bomb.
Speaker 3:But wait, didn't we already know this? Because isn't there that picture with Trump and Epstein together? We're really in like I mean, yeah.
Speaker 1:Yeah. I I wanna go back to AI business and The business story here is that Tesla's down 17%, DJT is down 7%. Trump coin is down 10%.
Speaker 3:Wow. They're all fighting.
Speaker 1:This, crash out on both sides is not good for anyone.
Speaker 3:Well, you know what's you know
Speaker 1:what's some interesting
Speaker 3:You know what's not down? Tokens generated, baby. We're still generating tokens every single day. The the relentless march of artificial intelligence continues.
Speaker 1:So so the other thing is is Elon shared, or sorry, Trump shared on truth. It's funny they're
Speaker 3:battling on their Different social networks.
Speaker 1:Billionaire should have their own, you know, social media network to get the word out. But Yeah. Trump said, the easiest way to save money in our budget, billions and billions of dollars, is to terminate Elon's government subsidies and contracts. I was always surprised that Biden didn't do it. Wow.
Speaker 1:So Ashley St. Clair is saying, hey, Donald Trump, let me know if you need any breakup advice. I really can't it. Dan Primax says, this cannot be a comfortable day for David Sachs. On the other hand, it's just the best day for Sam Altman.
Speaker 3:Well well, we have someone from OpenAI here. We're gonna stick to technology and business, but welcome to the show, Mark Chen. Good to see you.
Speaker 5:Good to see you guys. Thanks for
Speaker 3:having good day. But I'm excited to talk about deep research. I am excited to talk about AI products. Would you mind introducing yourself and kind of explaining what you do? Because OpenAI is such a large company now, and there's so many different organizations.
Speaker 3:I'd love to know, how you interact with the product and the research side and anything else you can give to contextualize this conversation.
Speaker 5:Yeah. Absolutely. So first off, you know, thanks for having me on. You know? I'm Mark.
Speaker 5:I am the chief research officer at OpenAI. So, in practice, what that means is I work with our chief scientist, Jakob, and, you know, we set the vision for the research org. We set the pace. We hold the research org accountable for execution. Mhmm.
Speaker 5:And, ultimately, we really just wanna deliver these capabilities to everyone.
Speaker 3:That's amazing. In terms of research, I feel like a lot of the what happens in the research side is actually gated by compute. Is that a different team? Because what if the researchers ask for a $500,000,000,000 data center? That feels like maybe a bigger a bigger task.
Speaker 10:So, yeah, it it is useful for
Speaker 5:us to factor the problem of research and also kind of building up the capacity to do that research.
Speaker 6:So we
Speaker 5:have a different team. Greg leads that Mhmm. Which really thinks holistically about, you know, data center bring up and how to get the most compute for us.
Speaker 3:Mhmm.
Speaker 5:And, of course, when it comes to allocating that compute for research
Speaker 4:Yeah.
Speaker 5:You know, Jacob and myself do that.
Speaker 3:That's great. And so what what can you share that's top of mind right now on the research side? There's been this discussion of, pretraining scaling wall potentially, the importance of reinforcement learning, reasoning. There's so many different areas to go into. What's actually driving the most conversations internally right now?
Speaker 5:Yeah. Absolutely. So I think, really, it's a really exciting time to do research. Mhmm. I would say versus two or three years ago, I think people were trying to build this very big scaling machine.
Speaker 3:Yeah.
Speaker 5:And, really, the reasoning paradigm changed a lot of that. Right? You know? Like, reasoning is really taking off, and it really opens this new playing playing ground. Right?
Speaker 5:It's like there are a lot of kind of known unknowns and also unknown unknowns that, you we're all trying to figure out. It kinda feels like GPT two era, right, where where there's so many different hyperparameters you're trying to figure out. And then I think also, you know, like you mentioned, you know, pretraining, that's not to be forgotten either. You know, today, we're in a very different regime of pretraining than we used to be. Right?
Speaker 5:Today, we can't treat data as this infinite resource. And I think a lot of academic studies, you know, they've always kind of treated you know, you have some kind of finite compute but infinite data. I don't think there's much study of, you know, like, you know, finite data and infinite compute. And I think, you know, that also leads to a very rich playground for research.
Speaker 3:Do we need kind of a revision to the bitter lesson? Is that a a refutation of the bitter lesson? Or
Speaker 4:No. No. Do we just need to
Speaker 3:re re rethink what the definition of of scaling laws looks like?
Speaker 5:No. I I don't think of, anything as a refutation of of the bitter. Really, like, our company is grounded in we want simple ideas that scale. I think RO is an embodiment of that. I think pretraining is an embodiment of that.
Speaker 5:And, really, at every single scale, we face some kind of difficulty of this form. It's just like you gotta find some innovation that gets you past the next bottleneck. And this doesn't feel fundamentally very different from that.
Speaker 3:Mhmm. What is, what's most important right now on the actual, compute side? We heard from NVIDIA earnings that that we didn't get a ton of guidance on the shift from training to inference usage of NVIDIA GPUs, but it feels like it must be coming. It feels like this inference wave is is is happening. Are those even the right buckets to be thinking about tracking metrics in terms of the the the story of artificial intelligence?
Speaker 3:Because yeah. I mean, it's like if if the reasoning tokens are inference tokens and and but they're what lead to higher intelligent, more intelligent models, like, it's almost back in the training bucket again. What bucket should we be thinking about and and, and or or are we how firmly are we in the the the, the the applied AI era versus the research era?
Speaker 5:Well, I think research is here to stay, and it's for all the reasons I mentioned above. Right? It's such a, like, a rich time to be doing research. Mhmm. But I do think, you know, inference is going to be increasingly important as well.
Speaker 5:Right? It's such a core part of RL that you're doing rollouts. And I think, you know, we see 2025 as this year of agents. Right? Mhmm.
Speaker 5:We think of it as a year where models are gonna do a lot more autonomous work. You can let them kind of be unsupervised for much longer periods of time. And that is just gonna put big demands on inference. Right? When you think about kind of our overall vision, right, we we lay it out as a series of steps and levels on the way to AGI.
Speaker 5:Right? And I think the pinnacle, really, that last level is organizational AI. Right? Like, you can imagine a bunch of AIs all interacting. And, yeah, I think that's just gonna put huge demands on inference.
Speaker 5:Right?
Speaker 4:On that
Speaker 3:on that organizational question, I I remember reading, AI 2027, and one of the things that they proposed was that the AIs would actually, like, literally be talking to each other in Slack. Does that seem like does that seem like the way you imagine agents playing out, like, using the two the same tools as humans instead of
Speaker 1:One agent says, I'm gonna go talk with Teams. I'm gonna talk with Slack. Mhmm. I'm gonna do a little negotiating.
Speaker 3:But maybe it just happens super super fast twenty four seven, or or is there, like, a new machine language that emerges?
Speaker 5:Yeah. I mean, I think one thing that's really helped us so far in AI development is, to come in with some priors for, you know, how humans do things. And Mhmm. That's actually, you know, if you bake those priors in, they they typically are great starting points.
Speaker 3:So Mhmm.
Speaker 5:I could imagine, like, maybe you start with something that's Slack like and give it enough flexibility that it can kind of develop the beyond that and really figure out the way that's most effective for it to communicate. One important thing, though, is, you know, we want interpretability too. Right? I think it's it's very helpful for us today that what the agents do is, you know, easy for us to read and interpret. And I don't think you want that to go away as well.
Speaker 5:So I think there's a lot of benefits just even from a pure, like, debug the whole system perspective. Sure. Or just let the models, you know, speak in a way that is familiar with us. And, you know, you you could also imagine, like, we might wanna plug in to the system too. Right?
Speaker 5:So, you know, whatever interfaces we're familiar with, we would ideally like our model to be familiar with as well. Yeah. I think it's also pretty compatible with you know, we hit a big milestone. We got, I think, 3,000,000 paying business users for for fairly recent week. Let's go.
Speaker 7:Yeah. There we go. Let's
Speaker 6:go. And
Speaker 3:Three Gong hits for 3,000,000.
Speaker 1:The Gong will keep ringing for a while. Sorry.
Speaker 3:We had to do I was hoping you would drop a number.
Speaker 5:Yeah. Yeah.
Speaker 3:Anyway Congratulations. That's that's actually huge. That's amazing.
Speaker 5:Yeah. Yeah. Yeah. But I think one big part of that is, you know, we have we have connectors now. Right?
Speaker 5:Yeah. We're connecting into, you know, like, G drives. And Mhmm. I think, yeah, you can imagine, you know, like, Slack integrations, things like that. I think we just want the models to be familiar with the ways we communicate and and get information.
Speaker 3:Yeah. Can you talk about benchmarking? It feels like we're potentially
Speaker 1:Yeah. Do you think about benchmarks at all?
Speaker 5:Oh, yeah. Yeah. A lot. I mean Okay. But I think it's a difficult time for benchmarks.
Speaker 5:Right?
Speaker 4:I
Speaker 5:think we used to be in this world, where you have these human written benchmarks for other humans. Right? And I think we all have these norms for, like, what are good benchmarks. Right? Like, we've all taken the SAT.
Speaker 5:We all have, like, a good conception of what it means to get, you know, whatever score on that. But I think the problem is the models are already at the point where for even the hardest human written benchmarks for other humans, it's really near saturated or saturated. Right? I think one clear example here is the Amy. Like, probably the hardest autogradeable, like, human math eval, at least in the in The US.
Speaker 5:And, yeah, the models are consistently getting, like, 90 plus percent on these. And so what that what that what that means is I think there's kind of two different things that people are doing. Right? They're they're developing kind of model based benchmarks. Right?
Speaker 5:They're not kind of things that we would give to an ordinary human. Things like humanity's last exam, things like, you know, Epic AI that are really, really at the at the frontier of what what people can do. And I think the the hard thing is it's not grounded in intuition anymore. Right? Like, you don't you don't have a lot of people who have taken these exams.
Speaker 8:Mhmm.
Speaker 5:So it it makes it harder to kind of calibrate on whether this is a good exam or not. One of the exciting things that's on the flip side of that is I really do think we're at the era where models are gonna start innovating. Right? Because I think once you've passed the last kind of, like, the hardest human rating exams, that's kind of at the edge of innovation. And I I think you already see that with the models.
Speaker 5:Right? Like, they're helping to write parts of papers.
Speaker 3:Mhmm.
Speaker 5:And and I think the other kind of way that people have shifted is, you know, there's these, you know, ultra frontier evals, but they're also people kinda just indexing on real world impact. Right? You look at your revenue, kind of the value you deliver to users, and I think that's ultimately what we care about.
Speaker 3:Can you can you, bring that back to interpretability research? Like, with these super, super hard, math evals, for example, are are we doing the right research to understand if the thought process mirrors not just not just one shotting the answer, oh, you you you you memorized it or you magically got it correct, but you actually took the correct path kind of like, you know, you're graded for your work, not just the answer if you're in grade school. Yeah. And and, you know, Dario said that, interpret interpret interpretability research will actually contribute to capabilities and even give a decisive lead. Do you agree with that?
Speaker 3:What's your reaction to that concept of interpretability research being very important?
Speaker 5:Yeah. I mean, we care a lot about it here at OpenAI as well. So one thing that we care a lot about is interpreting how the model reasons. Right? Because I think we've had a very kind of specific and strong view on this in that we don't want to apply optimization pressure to how the model thinks so that it can be faithful in the way it thinks and to expose that to us, you know, without any kind of incentives to cater to what the user wants.
Speaker 5:Right? I think it's actually very important to have that unfiltered view because, you know, oftentimes, like, if if the model isn't sure, you don't wanna hide that fact, right, just for for it to kinda please the user. And sometimes it really isn't sure. Right? And and so we've really done a a lot of work to try to promote this norm of chain of thought faithfulness and and interpretability.
Speaker 5:And I think it it gives you a lot of sense into what the model's thinking and, you know, what are the pitfalls that it can go off into if it's not reasoning correctly.
Speaker 1:That's such an important point because if you have somebody on your team and they come to you and they say, hey. You know, I think this is the right answer, but we should probably verify it. It's like, it's still valuable. Totally. It puts you on the right path.
Speaker 1:Yeah. If somebody comes to you a % confidence, this is this is the truth, and they end up being wrong, it's like, well, like, trust is just destroyed.
Speaker 3:Yeah. Totally. Yeah.
Speaker 5:Do you guys feel like, you know, safety felt a lot more theoretical a couple years back? Right? But, like, today, you know, like, the things that people are talking about a couple years, like, scalable oversight. Like, really having the model be able to tell you, like and convince you that the work it did was right, it feels so much more relevant right now
Speaker 3:just because %.
Speaker 5:The capabilities are so strong.
Speaker 3:Yeah. I mean, just personally, I've I've completely flipped from being like, oh, the safety research is not that valuable, because I'm not that worried about getting paperclip. It just seems like a very low likelihood that that's kind of, like, the bad ending, like, immediately and this fume and all this crazy, the Grey Goose scenarios were just so abstract in sci fi. It just felt like economics will will will will fall into place, and there will be, like a like a cold like, a nuclear ending, which is like we didn't build nuclear plants, and we just stopped everything because we seem humans seem to be good at that. But now
Speaker 7:Yeah. Know.
Speaker 3:That we're actually seeing
Speaker 2:things go.
Speaker 5:Yeah. Yeah. It's crazy how fast it's been. Right? Like
Speaker 3:Oh, yeah.
Speaker 5:I think my my, like, my personal story is is, like, you know, what what got me into AI was AlphaGo. Right? Like, just watching it get to that level of capability at
Speaker 3:Go. Yeah.
Speaker 5:And you were kinda like
Speaker 4:it
Speaker 5:was such an optimistic and also kind of a little bit of a sobering message, right, when you saw Lisa, it'll get beat. Mhmm. And I just remember, you know, like, we we saw the coding models, you know, when we first launched, like, I think, very OG codecs, you know, with with GitHub Copilot. It was maybe, like, under, you know, a thousand Elo on on Codeforces. And I still remember the meeting where I walked into where the team showed my score, and they're like, hey.
Speaker 5:Was models better than you? And it's like, you you come full circle, and it's like, wow. Like, I put decades of my life into this.
Speaker 3:And Yeah.
Speaker 5:You know, the capabilities are there. So, like, if you know, I'm kind of at the top of my field in this thing and it's better than me, like, what can it do, really?
Speaker 3:Yeah. Yeah. That's amazing. Do I I have so many more questions. On AlphaGo, are there are there lessons from scaling how scaling played out there that you can that we can abstract, abstract into the rest of AI research.
Speaker 3:What I mean is, as I remember it, the AlphaGo training run was not a hundred k h two hundreds.
Speaker 4:Mhmm.
Speaker 3:But what would happen if we actually did an AlphaGo style training run? I mean, it would be an economic money pit. Right? Like, they've had no economic value to do. But let's just say some benevolent trillionaire decides I'm gonna spend a billion dollars on a training run to beat AlphaGo and go even bigger.
Speaker 3:Is is Go at at some point solved? Would we see kind of diminishing scaling curves? Could we throw extra RL? Could we could we port back everything that we're doing in just general AGI research and and and just continue fighting it out in the world of Go, or does that end, and does that teach us anything?
Speaker 5:Yeah. Yeah. Honestly, I feel like if you really are curious about these misters, join our team. That's what I wanna say.
Speaker 3:Of course.
Speaker 5:But, yeah, I mean, really, like, kind of the the central problem of today is RL scaling. Right? Yeah. When you look at AlphaGo, right, it's it's a narrow domain. Right?
Speaker 5:And I think in some sense, that limits the amount compute you can pump into it. Mhmm. But even kind of small toy domains, they can teach you a lot about how you scale our own. Like, what are the axes where it's most productive to to pump scale in? I think a lot of scaling research just looks like that, whether it's on RL or pretraining.
Speaker 5:So you identify a lot of, you know, different different variables under which you can scale and, like, where is kind of where you get the best kind of, like, marginal impact for for bumping scale there. I think that's a very open question for RL right now.
Speaker 4:Mhmm.
Speaker 5:And I think what you mentioned as well is just, you know, going from narrow to broad. Right? Does that give you a lever to pump a lot more scale in as well? I think when you look at our reasoning models today, they're a lot more broad based than, you know, just being able to kind of an expert system on Go. So, yeah, I I really do think that there are so many levers to scale.
Speaker 3:And What about Move 37? That was such an iconic moment in that AlphaGo, Lisa Dahl match. They placed Move 37. It's very unconventional. Everyone thinks it's a blunder.
Speaker 3:It turns out not to be. It turns out to be critical. It turns to be it turns out to be innovation. Do you think we are we're certainly post Turing test in language models. We're probably post Turing test in image generation, but it feels like we're pre Move 37 in text generation in the sense that there hasn't been, like a fully AI generated book that everyone is just, oh, it's the new Harry Potter.
Speaker 3:Everyone has to read it. It's amazing and it's fully age and it's fully generated. Or, or this image, the images, they do go viral, but they go viral because they're AI. Move 37 in the context of Go did not go viral because it was AI. It felt like it was actual innovation.
Speaker 3:So, is that the right frame? Does that make any sense?
Speaker 5:I think it's not the wrong frame. So I I think some some quick thoughts on on on I I think kind of when you have something that's, you know, very measurable, like win or lose. Right? Something like like Go. Yeah.
Speaker 5:It's, like, very easy for us to kind of just judge. Right? Like, did did the model do something right here? Yeah. And I think the more fuzzy you get, you know, it it is just harder.
Speaker 5:Right? Like, when it comes to, you know, is this the next Harry Potter? Right? Like, you know, it's not a universally loved book. I think
Speaker 3:fairly
Speaker 5:universal, but, you know, there's there's some haters. Yeah. And, yeah, I I I think it it it is just kind of hard when it comes to these human subjective things
Speaker 3:Right.
Speaker 5:Where it's really hard to put down in words, like, what makes you like Harry Potter. Right? And and and so I think those are always gonna lag a little bit. But, you know, I I think, you know, we're we're developing more and more techniques to attack kind of these more open ended domains. And I don't know.
Speaker 5:I I wouldn't say that we're not at an innovative stage today. So I think my biggest touch with this was when we had the models compete on the IY last year. So I like, it's, like, the the international, basically, Olympics for for computer science. Basically, the the top four kids from from each country go and compete. And these are really, really tough problems, basically selected so that they require some innovative insight to solve.
Speaker 5:Right? I think and we did see the model come up with solutions even to some very ad hoc problems. And and so I think there was a lot of surprise for me there. Right? I was completely off base about which problems the model would be able to solve the most.
Speaker 5:Right? I think, like, I I kinda categorized there there's six problems. Some of them as more kind of like, oh, this is standard, a little bit more standard. This is a little bit more out of the box. I was like, it's not gonna be able to solve this, more out of the box one, but it it did.
Speaker 5:And I think, I think that really does speak to kind of, these models have the capacity to do so, especially trained with RL.
Speaker 3:Now now now put that in context of what's going on with Arc AGI. Obviously, OpenAI has made incredible progress there, but it just when I do the problems, it seems easy. And when I look at the IOI sample problems, I think this would be a twenty year process for me to figure out how to achieve that, and I can do the
Speaker 5:Yeah.
Speaker 3:Arc AGI on my phone. Is this the spiky intelligence concept? Is this something that a small tweak in in algorithmic design, just one shots AGI Arc AGI, or or is there something else going on there that we should be aware of?
Speaker 5:Yeah. I mean, I think, part of this is the beauty of Arc AGI as well. Right? Like
Speaker 3:Yeah.
Speaker 4:I think
Speaker 5:I'm not sure if there's another kind of, like, human intuitive simpler benchmark, which is best. For the models. Yeah.
Speaker 4:And I
Speaker 5:think, really, that's one of the things they really optimize for on on on that benchmark. I do think when it comes to models, though, like, there's just a little bit of a perception gap as well. Like, you know, models aren't used to this kind of native, you know, like, just screen type input. I think there's a lot we can bridge there. Actually, even o four mini, it's a state of the art multimodal model in many ways, including visual reasoning.
Speaker 5:And I think, you know, you're you're starting to kinda build up the capacity for the models to take images, manipulate, and and reason about them, generate new images, write code on images. And I think it's just been kind of underfocused. But I think when I talk to researchers in the field, they all see this as a part of intelligence too, and we're gonna continue to focus there.
Speaker 3:Yeah. Is is is RKGI kind of in the if we're dropping a buzzword on it, like, program synthesis? Is there a world where, I I know that I I know the tokens, like, the the images, we see them as as renderings of squares and different colors. But Mhmm. The when they're fed into the LLM, they're typically just a stream of of numbers effectively.
Speaker 3:Is there a world where actually adding a screenshot is what's important, like visual reasoning?
Speaker 5:Yeah. Yeah. So I think I think that could be important. It's just, like, kind of, you know, whenever it comes to, like, textual representation of grids
Speaker 4:Yeah.
Speaker 5:Models today just don't really do that well. Right? And and I think it's just kind of because humans don't really ever write down textual representations of grids. Right? Yeah.
Speaker 5:It's like, you know, we have a chessboard. Like, no one really kinda just, like, types it out in a grid. Like
Speaker 3:Yeah.
Speaker 5:And and so the models are kind of, like, undertrained a little bit on on what that looks like and what that means. Sure. So, you know, I I I think with more reasoning, it'll we'll we'll just bridge the gap. I think with better visual perception, we'll just bridge that gap.
Speaker 1:Yeah. How are you thinking about the role of non lab researchers in the ecosystem today? I'm sure you try to recruit some of the best ones,
Speaker 12:but the ones that
Speaker 3:don't join your team Tell us about the one that got away.
Speaker 1:Yeah. The one that got away.
Speaker 5:Yeah. No. I mean, I think it's still actually a fairly good time. I for for specific domains, right, to to to be doing research. And, you know, I think the style is just very different.
Speaker 5:And you do feel the pull of nonlab researchers into labs because I think they feel like a lot of the burning problems in the field are at scale. Right?
Speaker 6:That's good.
Speaker 5:Yeah. And that's kind of one of the unfortunate things too. Right? Like, when you look at reasoning, you just don't see that happen at small scale. Right?
Speaker 5:There's, like, a certain scale at which it starts becoming signal bearing, and that requires you to have resources. Right? But I do think, you know, a a lot of the really good work that I've seen, you know, there's experiments on architectures. I think a lot of good work is happening in the academic world there. Like, lot of study in optimization, a lot of study in kind of like GANs.
Speaker 5:You know? There's certain fields where you see a lot of fruitful research that that happens in academia.
Speaker 3:Yeah. That makes a lot of sense.
Speaker 1:How about consumer agents? How are you thinking about them? You talked earlier about sort of b to b adoption, and that's all very exciting. But how much do you and the research org think about breakout consumer agent products?
Speaker 5:Yeah. That's a fantastic question. I think we think about it a lot. I think that that's the short answer. You know, we we really do think, like, this year, we're trying to focus on how we can move to the agentic world.
Speaker 5:Right? And when I when I think about consumer agents, I think, like, ChachiPity proved that, you know, people got it. Right? It's like people get conversational agents when they conversational kind of models. But when when it comes to consumer agents, we have a couple of theses and that we've tried out in the world.
Speaker 5:I think one one is deep research. Right? I think this is something that can do five to thirty minutes of of work autonomously, come back to you, and really, like, kind of synthesizes information. Right? It goes out there, gathers, collects, and kind of, you know, compresses the information in in a form that that's useful.
Speaker 1:A little bit of a little bit of pushback there. Like, I can see that as a consumer product when someone like Aiden is like, I want new towels. And he uses deep research to, like, figure out, like, what is the best towel across every dimension. But when I think of deep research, yes, it has applications with students, but it's often
Speaker 3:Some of them might just be heard because feel like
Speaker 1:consumers being like, give you a deep research report on this country and where to travel and
Speaker 3:things like that. Using this flight example, but I don't I haven't actually tried to book a flight with Deep Research. It's totally possible that it could go and pull all the different flight routes and and calculate all the different delays and all the different all the different parameters of if I fly to this airport Yeah. I can park or I can use valet here or something like that. Yeah.
Speaker 1:Yeah. And I guess, like, when I think of agents, it's it's deep research is like, you know, curating information on which you can take action on. Totally. But it's like at what point is action a part of that sort of loop. Right?
Speaker 1:Where you can not only curate a list of flights that you want, but then, you know, actually go out and and and have agency.
Speaker 5:Yeah. I think one of our explorations in that space is operator. Right? Yeah. It's where you kind of just feed in raw pixels from your your your laptop into or, you know, from some virtual machine into the model, and it it produces, you know, either a click or some keyboard actions.
Speaker 5:Right? And so there, it's taking action. And I think the trouble is, you know, you don't ever wanna mess up when you're taking action. Right? I think the cost of that is super high.
Speaker 5:You you only have to get it wrong once to lose trust in in a user. And so we wanna make sure that that feels super robust Yeah. Before we get to the point where we're like, hey. Look. Here's a tool.
Speaker 5:I
Speaker 3:That's so different than deep research because Yeah. Like, you can wind up on some news article and read one sentence that gets a fact wrong or Yeah. Exactly. Commas in the wrong place and the numbers off. And but that's just the expectation for just text and analysis.
Speaker 3:And if you delegated that, yeah, you're gonna expect a few errors here and there. Oh, that's actually a different company name or that's the that's an old data point. There's new data. But very different if I book a flight and book the wrong flight, and I can wind up in Chicago instead of New York.
Speaker 5:Exactly. And I think the reason why we care so much about reasoning is because I think that's the path that we get reliable agents through.
Speaker 3:Sure.
Speaker 5:Right? You know, we've talked about, like, reasoning helping safety, but reasoning is also helping reliability. Right? It's like Mhmm. You imagine, like, what
Speaker 2:makes
Speaker 5:a model so good at a math problem? It's like it's banging its head against it. It's trying a different approach, and then it's, like, adapting based on what what it failed at last time. Yeah. And I think that's the same kind of behavior you want your your agents to have.
Speaker 5:It's like
Speaker 1:Yeah.
Speaker 5:Tries things, like, adapts and and keeps going until it it succeeds.
Speaker 1:And that's that's the humans do this every day. You're booking a flight. You keep hitting an error. It's Yeah. Not which or which form you missed.
Speaker 1:Right? And you're just sort of banging your head against the computer and eventually it says, okay, you're booked. Right? So I think I think that's a great call out.
Speaker 3:Yeah. I mean, the there's so many more questions we can go into, but I'm I'm interested in the scaling of RL and kind of the balancing act between pretraining RL and inference, just the amount of energy that goes into getting a result when you distribute it over the entire user base. How is that changing? And I guess, is is are we post, like, really big really big runs? Is this gonna be something that's, like, continually happening online?
Speaker 3:Or it feels like we're moving away from the era of, like, oh, some big development some big run happened, now we're grouping the fruit fruits of it versus a more iterative process.
Speaker 5:Yeah. I mean, I don't see why it has to be so. Right? I think, like, if you find the right levers, you can really pump a lot of compute into RL as well as pretraining. Mhmm.
Speaker 5:I think it is a delicate balance, though, between all of these different parts of the machine. And, you know, when when I look at my role with Jakob, it's just kinda, like, figuring out where how how this balance should be allocated, where the promising kind of, like, nuggets are arising from and and resourcing those. Yeah. It's it's kind of a in some sense, my I feel like part of my job is a portfolio manager.
Speaker 1:Yeah. That's a lot of fun.
Speaker 3:Well, thank you so much for joining. This is a fantastic conversation. We'd love to have you back and go deeper.
Speaker 1:Great hanging, Mark. Yeah. Absolutely.
Speaker 5:Yeah. Peace.
Speaker 1:Have a good one.
Speaker 3:Next up, we have Shalto Douglas from Anthropic coming on this show.
Speaker 1:I'm kidding. So I just
Speaker 3:Jordy is giving us the update on
Speaker 1:I'm just getting a lot of messages saying why no one cares about AI. Talk about the drama on the timeline. That's Well, we do care about AI. We care a lot about But it is a mess out there.
Speaker 3:Wow. Yeah. The end of the Trump Elon era. I don't know. Well, maybe maybe we have to get some people on to talk about it tomorrow or something.
Speaker 1:Gotta do it today.
Speaker 3:Anyway, we have Shalto from Anthropic in the studio. How are you doing?
Speaker 1:What's going on?
Speaker 8:Good to see you guys.
Speaker 3:Hopefully, you're you're staying out of the chaos on the Open ads. Don't don't open your time now. Don't open. What's the
Speaker 1:Sweet shot.
Speaker 4:Sweet child. Just Use the Twitter. Use the Twitter. Yeah. Mute
Speaker 3:Stay focused on the application.
Speaker 1:Stay focused on the mission.
Speaker 3:Stay focused on the next training run.
Speaker 1:We really humanity really cannot afford for any Discharge. Researchers to open x today.
Speaker 3:What a hilarious day. Anyway, I mean Yeah. Guys. Yeah. How are you doing?
Speaker 3:What what is new in your world? What what what are you focused on mostly day to day? And maybe maybe it's just a way of an intro.
Speaker 8:Yeah. So at the moment, focus really hard on scaling RL. Mhmm. And that is the theme of what's happening this year. Mhmm.
Speaker 8:And we're still seeing these huge gains where you go, you know, 10 x compute increase in RL. We're still getting, like, very distinct linear gains
Speaker 4:Mhmm.
Speaker 8:Basis for that. Yep. And because RL wasn't really scaled anywhere close to how much pre training was scaled at the end of at the end of last year Yeah. We have, like, a, basically, a gamut of, like, riches over the course of this year. Yeah.
Speaker 3:So where are we in that in that RL scaling story? Because I I I remember the the some of the rough numbers around, like, GPT two, GPT three. We were getting up into, like, it cost a hundred million dollars. It's gonna cost a billion Like, it just rough order of magnitude, not even from Anthropic, just generally, like, what is a big RL run cost or or how many are we talking 10 k h two hundreds or a hundred k? Like, are we gonna throw the same resources at it?
Speaker 3:And if so, how soon?
Speaker 8:Yeah. So I think in Diario's essay at the beginning of the year, he said that a lot of runs were only, like, a million dollars back in, like, December. Yeah. I think you have, like, DeepSeq v three and this kind of stuff like r one Yep. Which means that with us, like, at least two ooms just to get to the scale of g p t four, and g p t four was two years ago.
Speaker 8:Yep. Right? RL is also perhaps a bit more naively paralyzable and scalable than pre training. In pre training, you need everything in one big data center ideally, or you need like some clever tricks. Yep.
Speaker 8:RL, you could like in theory, like what the prime inter, like folks are doing, scale it all over the world. Yep. Out of it. And and so we you're held back, like, maybe like, you're you're held back far less than you
Speaker 3:are on
Speaker 8:free trade.
Speaker 3:Sure. So everyone and their mother has a billion dollars now. There are there are, you know, hundreds of thousands of GPUs getting pumped all over the I I I feel like we're not GPU poor as a as a as a society. Maybe some companies need to justify it in different ways. But it sounds like there's some sort of, like, reward hacking problem that we're working through in terms of scaling RL.
Speaker 3:What are all of the problems that we're working through to actually go deploy the capital cannon at this problem?
Speaker 8:Yes. So, I mean, think about what you're asking the model to do in RL is you're asking it to achieve some goal at at any cost, basically. Yeah. And this comes with a whole host of, like, behaviors, which you may not intend. Mhmm.
Speaker 8:In software engineering, this is really easy. I like to it might try and hack unit tests or whatever. In much more longer horizon, real world tasks, you might ask it to say, go make money on the Internet, and it might come up with all kinds of fun and interesting ways to do that unless you find ways to guide it into following the, like, principles that you want it to to obey, basically, or to to align it with your, like, idea of what sort of best for humanity. And so it's actually it's a pretty intensive process. Yeah.
Speaker 8:It's a lot of work to find down and hunt down all the ways these models are hacking through the rewards and and and and patch all of that. Yeah.
Speaker 3:Yeah. How how, are we going to see scaling in the number of rewards that we're RL ing against, if that makes sense? I would imagine that, at at a certain point, we unless we come up with, like, kind of, like, the the the Genesis prompt, go forth and be fruitful or something and and multiply. The you could imagine training runs on on just knocking down one one problem after another. And is that is that kind of the path that we're going down?
Speaker 8:I I very much think so. Mhmm. There's this idea in which, like, you know, the the sort of world becomes an RL environment
Speaker 3:machine Mhmm.
Speaker 8:In some respect. Because there's just so much leverage in making these models better and better at all the things we care about. Mhmm. And so I think we're gonna be training on on just everything in the world.
Speaker 3:Got it. Then and then does that lead to more model fragmentation, models that are good at programming versus writing versus poetry versus image generation, or or or does this all feed back into one model? Does the idea of the consumer needing to pick a model disappear? Are we in a temporary period for that paradigm?
Speaker 8:I think the main reason that we've seen that so far is because people are trying to make the best of the capital. Like, we are all still GPU poor in many Okay. And people are focusing those GPUs on the sort of, like, spectrum rewards that I think is most important. Mhmm. And look, I'm I'm a bit of a big model guy.
Speaker 8:Yes. I really do think that similar to how we saw with large pre trained models before, where small fine tuned models made it, made it like had gains over the sort of GPT-two era. Yeah. But then were obsoleted by GPT-four being generally good at everything. I think to be honest, you're gonna see this generalization and learning across all kinds of things.
Speaker 8:That means you benefit from having large single models rather than specialization or area fine tuned models.
Speaker 3:Can you talk a little bit about the transition from or many any differences between RLHF and just other RL paradigms?
Speaker 8:Yes. So RLHF, you're trying to maximize a pretty deep, like, lossy signal. Things like, pairwise, like, what do humans prefer? And I don't know if you've ever tried to do this, two language model responses
Speaker 3:I get prompted for that all the time. Right. And I'm always like, I don't wanna read both of those. I'll just click the one on
Speaker 4:the left.
Speaker 8:Exactly. Exactly. And, you you know, I click one of the random ones sometimes.
Speaker 3:Yeah. Or or I click, like, the one that just looks bigger or I'll read the first two sentences. But, yeah, I'm not giving straight I'm not I'm not being I'm not doing my job as a as a human reinforcer.
Speaker 8:Exactly. Human preferences are easy to hack.
Speaker 3:Yeah. Totally.
Speaker 8:Environments in the world are much truer. Mhmm. If you can find them. Yeah. So it's something like, did you get your math question right?
Speaker 8:Is a very real and a true reward.
Speaker 3:Does the code compile?
Speaker 1:Right?
Speaker 8:Does the code compile? Exactly. And did you make a scientific discovery? Yeah. We're gonna start like going we got very little rewards right now, but pretty quickly over the next year or two, you're gonna start to see much more meaningful and and long horizon rewards.
Speaker 3:You're gonna see models bribing the Nobel Committee to win the Nobel Prize. Good reward hacking. There's reward hacking. Yeah.
Speaker 1:There's reward. Prevent. Right?
Speaker 3:Exactly. It's yeah. Yeah. That that's the real nightmare scenario. What about like, there's so many different problems that we run into that feel like the it's it's just really, really hard to diva design any type of eval.
Speaker 3:The the, my kind of benchmark that I use whenever a new model drops is just tell me a joke. Yeah. They're always bad. And or or or even even the latest v o three video that went viral was somebody said, like, stand up comedy joke. And it was kind of a funny joke, but it was literally the top result for joke Reddit on Google, and then it clearly just took that joke and then instantiated in a video that looked amazing, but it wasn't original in any way.
Speaker 3:And so, we were joking about, like, the RLHF loop for that is like you have an endless cycle of comedians running AI generated materials and then and then, you know, speak microphones in all the comedy clubs to feedback what's getting laughs. But I
Speaker 1:mean, honestly, that would work pretty well, actually.
Speaker 3:Yeah. If anybody else wanna hook
Speaker 1:us up with an RL loop, mean
Speaker 3:Yeah. Yeah. But but, I mean, for for some of those less, like, as you go down the curve, it feels like each one gets harder and harder, to actually tighten the loop. We see this with, longevity research where it's like, okay. It takes a hundred years to know if you extended a human life.
Speaker 3:Like, the yes. You could create a feedback loop around that, but every change is gonna be hundreds of years. And so even if you're on the cycle, it's irrelevant for us in the context that we talk about AI. So talk to me about, like, are you running into those problems or or or will there be, like, another approach that kind of works around those?
Speaker 8:So there are a lot of situations where you can get around this by just running much faster than real time. Like, let's say, the process of building like a giant app, like building Twitter. Right? Yeah. It's something that would take human months.
Speaker 8:But if you got fast enough and good enough AIs, you could do that in several Right? I paralyzed heaps of AI agents. They're all building right, you know, things to expect. And so you can get a faster reward signal in that way. Mhmm.
Speaker 8:In domains that are less well specified like humor. I agree. It's really, really hard. And this is like why I think in some respects, like, creativity is like at the at the top end of the spectrum, true creativity is much, much harder to replicate than the sort of like analytical scientific style reasoning. Yeah.
Speaker 8:And and that will just take more time. You know what? The models actually are pretty good at making jokes about being an AI. This feels weird with fresh. It's just like everything else is kind of a weird copy of something.
Speaker 8:Like, it's like it just it feels like it's derivative, basically. It's trying to infer what humor is, and it doesn't really understand it. But jokes about being an AI are quite funny.
Speaker 3:Yeah. I I I I think this also might be I don't know if it was directly fascinating. Reward hacking, but I noticed that, one of the new models dropped and a bunch of people were posting these like 4chan, like be me memes. And and and they were it seemed like they were kind of hacking the humor by being hyper specific about an individual that they could find information on online. So you're laughing at the fact that it's like, oh, wow.
Speaker 3:That is like something that I've posted about it. It's making a reference. But it's not really that funny to me. It's other than it's just like, wow. They really did its research.
Speaker 3:Like, really knows Tyler Cowen intimately, which is cool. But I didn't find it hilarious. Yeah. Yeah. Yeah.
Speaker 3:Very interesting. Let let's talk about some sort of deep research product prod projects and products. We were talking to Will Brown, and he was saying, like, AGI is here with some of the bigger models, but the but the time that AGI can feel consistent, it diverges. And so you could be working with someone who's, you know, hundred IQ, but they'll but they will stay consistent for years as an employee or they'll they'll keep, you know, living their life. Whereas a lot of these super smart models are working really well, then after a few a few minutes of work, the the the agents kind of diverge and kind of go into odd paradigms.
Speaker 3:It feels very not human. It feels like a like just a they're hyper intelligent one way and then extremely stupid in others. What's going on there? What is the what is the path to extending that? Is that more, like, having more better planning and better better, like, dividing up the task?
Speaker 3:Or Yeah. Or will this just kind of naturally happen through the RL and scale?
Speaker 8:Yeah. So there's that jaggedness, right, which is what you're seeing is is how we call it. And I think that
Speaker 1:is largely a consequence of the fact
Speaker 8:that maybe something like deep research, it's probably been RL to be really good at producing a report. Yep. But it's never been RL'd on the, like, act of producing valuable information for a company over a a week or a month or, like, making sure the stock price goes up in, like, you know, a quarter or something like this. Right? Like, it it doesn't have any conception of how that feeds into the broader story at play.
Speaker 8:It can kind of infer because it's got a bit of world knowledge from
Speaker 3:the, you
Speaker 8:know, the base model this kind of stuff. There's never actually been trained to do that in the same way humans have. Mhmm. So to extend that, you need to put them in much longer running, much like like, you know, long horizon things. And so so deep research needs to become, know, like deep operator company for a week kind of
Speaker 3:thing. Sure. Is that the right path? Like, it feels like the road might be there's a like, the longest running LLM query used to be just like a few seconds, maybe a few minutes. And I remember when when some of the reasoning models came out, people were almost trying to, like, stunt on it by saying, like, oh, I asked it a hard question and thought for five minutes.
Speaker 3:Now deep research is doing twenty minutes pretty much every time. Yeah. Is the path two hours, two days, or are we gonna see more efficiency gains such that we just get the twenty minute model the twenty minute results in two minutes and then two seconds?
Speaker 8:Yeah. So this is somewhere where, like, inference in many respects and and prioritization becomes really important. Yeah. So both, like, how fast is your inference? It literally affects the speed at which you can think and the speed at which you can, like, like, you know, do the these experiments.
Speaker 8:Yeah. Also, how easily you can paralyze becomes really important. Yeah. Like, can you dispatch a team of of sub agents to go and do deep research and, like, compile, sub reports for you so that you can do everything in parallel? These kinds of, like it's it's both like there's an infrastructure question here Yeah.
Speaker 8:That feeds up from the hardware and the chips and this kind of stuff
Speaker 3:Yeah.
Speaker 8:To, like, designing better chips for, you know, better inference in this all in all this. And and an RL question of, like, you know, how well can you paralyze and and and all this. So I think we just need to compress the timelines, compress the time compress the time frames, basically.
Speaker 3:Yeah. So if I'm if I'm, like, an extremely big model and I'm running an agentic process, like, how how much am I hankering for, like, a middle sized model on a chip or, like, baked down into silicon that just runs super fast? Because it feels like that's probably coming. We saw that with the Bitcoin project progression from CPU to GPU to FPGA to ASIC. Do you do you think we're we're at a good enough point where we can even be discussing that?
Speaker 3:Because I every time I see, like, the latest mid journey, I'm like, this is good enough. I just want it in two seconds instead of twenty. But then a new model comes out. I'm like, oh, I'm glad I didn't get stuck on Right? But but, yeah.
Speaker 3:Like, how far away from how far away are we from, okay, it's actually good enough to bake down into silicon?
Speaker 8:Well, there's a question here of baking it down to silicon versus designing a chip, which is like very suit of the architecture that you
Speaker 1:care about.
Speaker 8:Right? And baking out of silicon, I'm sure. Like, I think that's a bet you could take, but it's a it's a risky one because Yeah. The pace of progress is just so fast nowadays. And I I really only expect it to to accelerate.
Speaker 8:Mhmm. But designing things that make a lot of sense for the sort of the tran you know, transformers or or architectures of the future should should make a lot of sense.
Speaker 3:Wait. Dude, that that's a big gap, though. Transformers or architectures of the future. If we diverge, there's lot of companies that are banking on the transformer sticking around. What is your view on transformer architecture sticking around for the next couple of years?
Speaker 1:I mean, look, stuck around for five years, so they might stick around for a little while. But there's there's different you think about architectures in terms of this balance
Speaker 8:of memory bandwidth and flops. Right? One of the big differences we've seen here is Gemini recently had actually a diffusion model that they released
Speaker 3:I was about to
Speaker 4:ask Yeah.
Speaker 8:The other day. Right? Yeah. So diffusion is inherently extremely flops intensive process.
Speaker 4:Mhmm.
Speaker 8:Whereas normal language model decoding is extremely memory bandwidth intensive. You're designing two very different chips depending on which bet you think makes sense. Yeah. And if you think you can make something that does flops, four times faster than diffusion and, like, four times cheaper than you always could, diffusion makes more sense. So there's, like, this is dance, basically, between the chip providers and the and the architecture
Speaker 4:Yeah.
Speaker 8:Both trying to build for each other, but also, like, build for the next paradigm. Yeah. It's risky.
Speaker 3:Do you I I I don't know how much you've how much you've played with image generation, but do you have any idea of what's going on with images in ChatGPT? It feels like there's some diffusion in there. There's some tokenization, maybe some transformer stuff in there. It almost feels like the text is so good that there's like an extra layer on top almost and that it's it's almost like reinventing Photoshop. And and I guess the the the broader question is like, it feels like an ensemble of models.
Speaker 3:Maybe the discussion around around just agents and text based LLM interactions shouldn't necessarily be transformer versus diffusion, but maybe how will these play together? Is that a reasonable path to go down?
Speaker 1:Well, I think pretty clearly,
Speaker 8:there's some kind of, like, rich information channel between even if there are multiple models there, there's like it's it's conditioning somehow on the on the other model because we've seen before, like, let's say when, you know, models use mid journey to produce images, it's never quite perfect. It can't perfectly replicate what went in as an input. It can't perfectly, like, adjust things. Mhmm. So there's a link somehow, whether that's the same model producing tokens plus diffusion.
Speaker 8:I don't know. Like, yeah, comment on what OpenAI is doing there.
Speaker 3:Yeah. Yeah. Yeah. Are are there any other kind of, like, wild card long shot research efforts that are maybe happening even out even in academia where I mean, this was the big thing with, what was his name? Gary, he was talking about, I forget what it's called, symbolic symbol manipulation was a big one.
Speaker 3:And and I feel like, you know, you can never count anyone out because if I come from behind, it'd be relevant in some ways. But but are there any other research areas that you think are, like, purely in the theory domain right now that are worth looking into or tracking that, you know, low op low low probability, but high upside if they work? Mhmm.
Speaker 8:It's tough one.
Speaker 4:This is
Speaker 1:a tough one.
Speaker 4:But I
Speaker 3:will say, just
Speaker 8:on the symbolic thing.
Speaker 3:Please.
Speaker 1:It's crazy how similar transformers are to systems that manipulate symbols.
Speaker 3:Sure.
Speaker 1:Right? Like, what they're doing is they're taking a symbol
Speaker 8:and they're, like, converting it into a vector, then they're manipulating and moving stuff, like, information around across them.
Speaker 4:Sure.
Speaker 8:Like, this this whole like debate that all transformers gonna represent symbols and they cannot do this. I think it's it's yeah. It's not
Speaker 3:So so Gary Marcus underrated or overrated? I I guess.
Speaker 8:Overrated.
Speaker 3:Yeah. Yeah. But but,
Speaker 9:I mean,
Speaker 3:if if you if you twist the if if if you twist it so much, you wind up with saying like, well, really, like, the the transformer fits within that paradigm. And so maybe it's, you know, it it it's, you know, it like, the rhetoric around it being a different path was maybe false the whole time. Yeah. Something like that. But but I but as as I remember that debate, it was really the the idea of compute scaling versus almost like feature engineering scaling.
Speaker 3:And and will the progress scale with human hours or GPUs essentially? And that has a very different economic equation. And it and it feels like there's been there's been some rumblings about maybe with a data wall, we'll shift back to being human labor bound. But do you think that there's any chance that that's relevant in the future, or is it just algorithmic progress married with bigger and bigger data centers in the future?
Speaker 8:So I'm pretty bitter lesson built. Yeah. Hence that I do think removing as many of our biases and our, like, clever ideas from the models is really important, just, like, freeing them up to learn. Now, obviously, there's, like, there is clever structure that we put into these models such that they're able to learn in this extremely general way and that but I I am more convinced that we will be compute bound than than we will be, like, human researcher out human researcher bound on on this kind of thing. Like, we're not gonna be feature engineering and this kind of stuff.
Speaker 3:Sure.
Speaker 8:We're gonna be trying to devise incredibly flexible learning systems.
Speaker 3:Yeah. That makes sense. On the scaling topic, part of I I I I haven't part of my, like, worry is that the the ooms get so big that they turn into these mega projects that are, that at a certain point, you're bound by the laws of physics because you have to move the sand into silicon chips and you have to dig up the silicon and then With a certain sand. Yeah. There's only so much sand.
Speaker 3:It's like the the math gets really, really crazy just for the amount of energy required to to move everything around to make the the big thing. Where are you on on how much scale we need to reach AGI? How whether or not we will see, like, the laws of physics start acting as a drag on progress because it certainly feels exponential. We're feeling the exponentials, but a lot of these turn into sigmoids. Right?
Speaker 8:Yep. So I think we've got what, like, two or three more oooms before before it gets really hard. Leopold has his nice table at the end of his Yeah. End of the situational awareness where Yeah. I think, like, 2028 or something is when under really aggressive timelines that you get to 20% of US energy production.
Speaker 8:Yeah. It's pretty hard to go exponentially beyond 20% of US energy production. Now I think that's enough. Every indication I'm seeing says that's enough. Now then there might be some complex, you know, data engineering, raw engineering, this kind of stuff that goes into that's lots of there's still a lot of algorithmic progress left to go.
Speaker 8:But I think that with those extra rooms, we get to basically a model that is capable of assisting us in doing research and software engineering.
Speaker 3:Yeah. Which is the beginning of the self self reinforcement.
Speaker 8:Yeah. Interesting.
Speaker 3:Is that just a coincidence? Like, this feels like one of those things this feels like one of those things where, like, the moon is the exact same size as the sun in the sky. It's like, oh, it just happens that AGI happens within this time. Like, hey. Woah.
Speaker 3:Did you have you unpacked that anymore? Because it feels convenient. Not not to, you know, I I know
Speaker 8:last month. There's a there's a lot of weird conveniences or, like, weird it's a good sci fi story, let's say.
Speaker 3:Totally.
Speaker 8:You know, we've got, you know, Taiwan in between China and The US, and it produces the most valuable material in the world. Yeah. It's locked between the two
Speaker 1:Incredible plot. Yeah.
Speaker 4:Incredible plot.
Speaker 3:Yeah. Really bad for the people that don't think of that don't believe in simulation theory. It really feels like, yeah, all this is scripted. It's fascinating. Talk to me more about, getting to an ML engineer in AI and and kind of that reinforcement.
Speaker 3:I imagine that you're using AI codegen tools today and and and Thropic is broadly and everyone is. But but, what are you looking for, and what are the what's the shape of the the spiky intelligence? Where do they fall flat? And what are you looking to kind of knock down in the interim before you get something that's just like go?
Speaker 8:Yeah. So, I mean, we definitely use them. The other night, I like I I was a bit tired. I asked to just do something, just sat watching it in front of me working for half an hour. It was great.
Speaker 8:Nice. It was truly weird experience, particularly when you look back a year ago, and
Speaker 4:Oh, yeah.
Speaker 8:We're still copy pasting stuff between a chat window and and, you know, a code file. Yeah. What I I like meters evals for this
Speaker 1:kind of stuff. So they
Speaker 8:have a bunch of evals where they measure, ability to write a kernel, the ability to run a small experiment and improve a loss, and they have these nice progress curves versus humans. And I think this is maybe the most accurate reflection of, like, what will take for it to really help us
Speaker 4:Mhmm.
Speaker 8:At at doing progress. And there's a mix here. Like, where they're not so great at the moment is, like, large scale distributed systems engineering. Right? Like, debugging stuff across heaps and heaps of accelerators and, like, the way the feedback loops are slow and
Speaker 4:Yeah.
Speaker 8:Yeah. You actually, like, the if your feedback loop is, like, an hour,
Speaker 3:then it's
Speaker 8:worth you spending the time on Got on doing something. Yep. And if your feedback is fifteen minutes.
Speaker 3:If it's much And and and and for context there, the hour long feedback loop is just because you have to to actually compile and run the code across everything to
Speaker 8:take that log. Like, spin up all your machines or
Speaker 3:you need
Speaker 8:to, like, oh, like, you need to, like, run it for a while to see something's gonna happen. Like, at that point in time, you're still cheaper than the chips.
Speaker 4:Sure. And
Speaker 8:so you're you're you're sort of it's better that you do it. Yeah. But for things like your kernel engineering or for, like, you know, actually even just understanding these systems, incredibly helpful. Like, one thing I regularly do at the moment is in past the code base in, like, languages that I'm unfamiliar with or some stuff like this. I'll just ask it to rewrite the entire file with comments on every line.
Speaker 8:Game changing. Wow. It's like
Speaker 3:Comments on every line.
Speaker 8:Yeah. Or just come through, like, thousands of files and explain how everything interacts to me, draw diagrams, this kind of stuff. It's really yeah.
Speaker 3:Yeah. How important is a bigger context window in that in that example you gave that feels like something that's important and yet it I I I just naively, like, Google's the one that has the million token context window. I imagine that all the other Frontier Labs could catch up, but it seems like it hasn't been as much of a priority as maybe, like, the PR around it sounds like. Is that important? Should we be go should we be driving that up to, like, a trillion token window?
Speaker 3:Yeah. Is that is that just gonna happen naturally?
Speaker 8:There's a nice plot in the Gemini one point five paper where they show the, like, loss over tokens as a function of context length,
Speaker 4:and they show
Speaker 9:that the
Speaker 8:loss goes down quite steeply, actually. As you put more and more and more like a code base in the context Mhmm. You get better and better and better at predicting the rest.
Speaker 3:Yeah. That makes sense.
Speaker 8:From the context length, it's cost. You know, the way transformers work is that there's you know, you have like this this memory that is proportional. The KBCache is proportional to how much context you've got. Mhmm. And so you can only fit so many of those into, like, the your various chips and this kind of stuff.
Speaker 8:And so a longer context actually just costs more because you're taking up more of the chip and you're sort of like, you could have otherwise been doing other requests, basically.
Speaker 3:Bringing it back to the custom silicon, is that a unique advantage of the TPU? Is is that something that Google has has thought about and then wound up to put themselves in this advantage position, or is it a durable advantage even?
Speaker 8:Yeah. So TPUs are good in many respects, partially because you can connect hundreds or thousands of them really easily across really great networking. Whereas only recently has that that been true for GPUs. Like NVLink? Yeah.
Speaker 8:With NVLink and, like, the NVL 72 stuff. Okay. So it used to be like eight GPUs and pod, and then, like, you connect them over worse interconnect. And now you can do 72, and and then it breaks down. With Google GPUs, you can do, like, 4,000, eight thousand over really high bandwidth interconnect Mhmm.
Speaker 8:In one pod. And so that that is helpful for things like just just general scaling in many respects. I think it's this is doable across any chip platform, but it is a is an example of like somewhere that being fully vertically integrated is a is is in a benefit.
Speaker 3:Yeah. That makes sense. Talk to me about Arc AGI. Why is it so hard? It seems so easy.
Speaker 8:It does seem easy, doesn't it? That's a well
Speaker 3:It certainly seems like more more evaluatable than tell me a funny joke. Right?
Speaker 8:Yeah. Yeah. And I I mean, I think if you are old on Arc AGI, then it would you'd probably get superhuman at it pretty fast.
Speaker 4:Mhmm.
Speaker 8:But I think we're all trying not to r l on it so that it functions as like an interesting held out.
Speaker 3:Sure. Okay. Wait. Is that just an informal agreement between all the laughs, basically?
Speaker 8:No. We're, you know, we're trying to have a sense of honor between us.
Speaker 1:That's good. Sense of honor.
Speaker 8:That's amazing.
Speaker 1:How many people on earth do you think are getting the full potential out of the publicly available bottles? Because we're now at a point where we have, you know, billion plus people are using AI almost daily. And yet, have to my sense would be it's maybe like 10,000, 20 thousand people on the entire planet are getting that sort of full potential. But I'm curious what your assessment would be.
Speaker 8:Yeah. I I completely agree. I mean, I think that even I don't get the full full potential out of these models often. And I I think as we shift from you're asking questions and it's giving you sensible answers to you're asking it to go do things for you that might take hours at a time and you can really like paralyze and spin. That we're gonna hit like yet another inflection point where even less people are like really effectively using these things.
Speaker 8:This is basically gonna require you to like it's like a like StarCraft or DOTA. Like, it's gonna be like a APM of like managing all these agents. And that's gonna be a process. Yeah.
Speaker 1:I think Starcraft is such a good example. You think you're just absolutely crushing it, and then you realize like there's an entire area of the map you're just getting destroyed. Yeah. Yeah. Exactly.
Speaker 1:It's such a good it's such a good comp.
Speaker 3:That's great. Anything else, Jordy?
Speaker 1:I think that's it on my side. I mean, I I would like this to be an evolving conversation.
Speaker 3:Yeah. This is fantastic. We'd love to a conversation.
Speaker 8:Absolutely. It's really fun. Love to be back on class.
Speaker 3:Yeah. We'll talk to you soon.
Speaker 1:Cheers, Shelton. Have a good one. Alright. We got the The
Speaker 3:possible AI day given
Speaker 1:the market. For context folks, we are gonna be doing a live time timeline and turmoil segment at 2PM PST. Let's do it. So if there's posts you want us to cover
Speaker 3:Send them in.
Speaker 1:You can go send them. I'll put this
Speaker 3:in chat as well. A few more. Pull one up. I'm gonna do some ads because we got Emmett Shearer coming in the temple in just a few minutes. First, let me tell you about Numeral.
Speaker 3:Sales tax on autopilot. Spend less than five minutes per month on sales tax compliance. Go to numeralhqAGI..com. Very excited for them. Also, go to public.com.
Speaker 3:Investing for those who take it seriously. They have multi asset investing, industry leading yields, and they're trusted by millions. Millions. In other news, Tim Sweeney continues to battle Apple. Apparently, if you search for Fortnite on the Apple App Store, it's he says, hey, kids.
Speaker 3:Looking to play Fortnite? Try this crypto and stock trading app instead rated for ages four plus courtesy of Apple App Store ads.
Speaker 1:Obviously, give you the latest Please. Trump terror, Elon Post four minutes ago. The Trump tariffs will cause a recession in the second half of this year.
Speaker 13:Wow.
Speaker 1:Somebody else was saying, can I finally say that Trump's tariffs are super stupid?
Speaker 3:Who's who is that?
Speaker 1:Somebody else is posting. Mads posting is saying, it's Xi Jinping. He says, bro, you seeing this? And it's Putin on the other end. He's just looking at it.
Speaker 1:Hold up. Got a line, and it's we'll start pulling some of these up.
Speaker 3:Ridiculous. What else is going on here? This is for the president versus Elon.
Speaker 1:Naval says, Elon's stance is principle principled. Trump's stance is practical. Tech needs Republicans for the present. Republicans need tech for the future. Drop the tax cuts.
Speaker 1:Cut some pork. Get the bill through.
Speaker 3:This is so crazy.
Speaker 1:Antonio Garcia says, remember, there's f you money and then there's f the world money. Will Stansell says, imagine being the ICE agents suiting up for your biggest mission of all time right now. People are saying that Trump's gonna deport Elon.
Speaker 3:Elon back to South Africa.
Speaker 1:Will DePuze says, time to drop the really big bomb, Growing Daniels in the Epstein files.
Speaker 4:That is I just turned
Speaker 3:into a copy pasta.
Speaker 5:Is a real Terrible.
Speaker 3:Oh, no. Delian.
Speaker 1:We had a question from a friend of the show. Yeah. He said, the real question is if Tesla is down 14%, how could SpaceX and OpenAI be trading if they were How would they be trading if they were public? The the real thing here is it's bad for everyone, right?
Speaker 3:It's bad It's for everyone. Right.
Speaker 1:Tritcoin is down. Nobody's really winning here. China's up. Yeah. Oh, really?
Speaker 1:I mean, I'm just saying at a
Speaker 3:high level Yeah. Yeah.
Speaker 1:Yeah. You know, China is the big beneficiary here of Sarah Guo says, if anyone has some bad news to bury, might I recommend right now?
Speaker 3:Yes. Yes. Yes. If you have if you what what what's the canonical bad startup news? Like, oh, yeah.
Speaker 3:You missed earnings or something. Drop it now.
Speaker 1:I've read Inverse Kramer says, Bill Ackman is currently writing the longest post in the history of this app.
Speaker 3:Okay. And we have a we have a video from Trump here if we want. I can throw it in the tab, and and we can share it on this on the stream and and react to
Speaker 1:it live. Fridman says to Elon, that escalated quickly triple your security. Be safe out there, brother. Your work, SpaceX, Tesla, XAI, Neuralink is important for the world.
Speaker 3:We need to get Elon on the show today. If somebody's listening and can make that happen, I
Speaker 1:would love to hear from Max Meyer says, so I got this wrong. I didn't say it never happened, but I thought it wouldn't. I am floored at the way this has happened.
Speaker 7:Yeah.
Speaker 1:He didn't think they would have a big breakup. Many people didn't think they would have a big breakup. Even just earlier this week, it seemed like they might just have a a somewhat peaceful exit. Trump just posted a little bit ago. I don't mind Elon turning against me, but he should have done so months ago.
Speaker 1:This is one of the greatest bills ever presented to congress. It's a record cut in expenses, $1,600,000,000,000 and the biggest tax cut ever given. If this bill doesn't pass, there will be a 68% tax increase and things far worse than that. I didn't create this mess. I'm just here to fix it.
Speaker 1:Anyways, lots going on.
Speaker 3:Let's go to this this Trump video. I wanna see what
Speaker 14:he to that I've seen and I'm sure you've seen regarding Elon Musk and your big beautiful bill. What's your reaction to that? Do you think it in any way hurts passage in the senate, which of course, what is your seeking?
Speaker 11:Well, look, you know, I've always liked Elon, and it's always very surprised. You saw the words he had for me, the words of and he hasn't said anything about me that's bad. I'd rather have him criticize me than the bill because the bill is incredible. Look, Elon and I had a great relationship. I don't know if we're well anymore.
Speaker 11:I was surprised because you were here. Everybody in this room practically was here as we had a wonderful send off. He said wonderful things about me. You couldn't have nicer. Said the best thing.
Speaker 11:He's worn the hat. Trump was right about everything. And I am right about the great big beautiful bill. But I'm very disappointed because Elon knew the inner workings of this bill better than almost anybody sitting here, better than you people. He knew everything about it.
Speaker 11:He had no problem with it. All of a sudden, he had a problem, and he only developed the problem when he found out that we're gonna have to cut the EV mandate because that's billions and billions of dollars, and it really is unfair. We wanna have cars of all types. Electric we wanna have electric, but we wanna have a gasoline combustion. We wanna have different.
Speaker 11:We wanna have hybrids. We wanna have all we wanna be able to sell everything. He hasn't said bad about me personally, but I'm sure that'll be next. But I'm I'm very disappointed in Elon. I've helped Elon a lot.
Speaker 1:That is mister president. Did he I just
Speaker 3:wanna clarify. Did he raise any of these concerns with you privately before he raised them publicly? And this is the guy you put in charge of cutting spending. Should people not take him seriously about spending now? Are you saying
Speaker 6:this is all sour grapes?
Speaker 11:No. He worked hard and he did a good job. And I'll be honest, I think he misses the place. I think he got out there all of a sudden he wasn't in this beautiful Oval office and he was and he's got nice offices too. But there's something about this when I was telling the chance Folks.
Speaker 11:This is where it is. People come in
Speaker 1:Breaking news. Dallian. That's Gruhav is joining us in the temple for some live reactions. Come on in.
Speaker 3:Guest. I can't even spell surprise guest. I'm so excited about this.
Speaker 1:Surprise guest. In other news, eleven labs dropped a new product.
Speaker 3:Another news. Oh, $2,000,000 seed round. Stop it. Stop it. We love Eleven Labs.
Speaker 1:No. They'll they'll they'll
Speaker 3:Keep grinding. But Just launch again tomorrow. Going to have to launch again. Start shooting a new vibe reel. Start shooting a new writing a new blog post because no one's got Lulu says, yes.
Speaker 3:Delay the launch on TV PM.
Speaker 1:So, basically, right now, I can just pull up and just refresh. There you go. I'm gonna just be refreshing True Social.
Speaker 3:How you doing? So okay. Jordy's on True Social. I'll be on X. Give us your reaction, Delihan.
Speaker 3:What's going on?
Speaker 6:Mean at some point I was like I'm just you know sort of scrolling X and
Speaker 10:I like tuned into you
Speaker 6:guys like an hour ago and I was like like they're talking about some AI thing. Was like at some point they're gonna switch to like we're half news and this is like and then I was watching and I was like okay
Speaker 1:I gotta John resisted.
Speaker 3:I fought it for like
Speaker 1:a for like a half an hour but we couldn't do it.
Speaker 3:But, yeah, give us your quick reaction.
Speaker 6:I mean, I'll always,
Speaker 10:you know, sort of give it from the, you know, sort
Speaker 6:of space angle. You know, it's amazing that, you know, how much the world has shifted since, you know, Friday of last week, whereas, you know, sort of presumed that Jared Isaacman was gonna be the, you know, sort of NASA admin Yeah. To today, it was released that the senate reconciliation package readded budget back into NASA largely for the SLS program, which was basically the program that, you know, sort of Jared and Elon were, you know, sort of largely advocating to, you know, sort of completely shut down. Yeah. So, you know, the the the it has already showed, like, you know, the sort of counter reaction, you know, is already showing up, you know, in in policy.
Speaker 6:Mhmm.
Speaker 3:Sorry. SLS program, is that space shuttle or no?
Speaker 6:Sorry. That's the SLS launch rocket. It is based off of old space shuttle hardware, but it is basically the internal, you know, sort of NASA run competitor effectively to, like, a, you know, sort of Starship heavy, you know, launch rocket. Yeah. So, you know, because it was, you know, sort of generally behind budget, behind schedule, and there are so many commercial heavy lift rockets coming online
Speaker 3:Sure.
Speaker 6:The default was canceled. That is largely, you know, sort of a Boeing based program. And so, you know, if you look at, you know you know, three months ago, you know, when they were announcing the f 47 program
Speaker 3:Yeah.
Speaker 6:You know, Elon walks into the secretary of the Air Force's office. Obviously, he'd been, you know,
Speaker 4:sort of
Speaker 6:manned fighter jets and believing that that shouldn't be what, you know, be what the department is prioritizing. Thirty minutes after that meeting was when they announced the f 47 program. And so now you're seeing basically, like, the equivalent in space where, you know, know, that would you know, there was obviously awarded to Boeing. Boeing was the is the largest prime behind, you know, sort of SLS. You know, Boeing basically, you know, is gonna be the biggest winner of, you know, NASA refunding USSLS and Jared Eisenman not being NASA administrator.
Speaker 1:So tying this back to the timeline, Trump posted less than thirty minutes ago. In light of the president's statement about cancellation of my government contract, SpaceX will be begin decommissioning its Dragon spacecraft immediately. Break that down.
Speaker 6:I mean, that just means that we no longer have a vehicle that can go to the International Space Station. We no longer have a vehicle that can bring astronauts up and down. Oh, gosh. We we also don't have a vehicle that can deorbit the International Space Station safely. Right?
Speaker 6:The the Dragon was expected to be able to do that. So what that means is, you know, if you guys remember all the memes about stranded, you know, from last year around Boeing Starliner, it now means that the space station, you know, itself is basically, you know, sort of stranded. And that's, like, you know, one of the government contracts, obviously, that, you know, SpaceX is involved in. And Elon, I've heard generally, like, just wants to shift all things to Starship anyways, and so in some ways, it's probably kind of looking for an excuse to, you know, sort of shut down Dragon and refocus energies. There's also a part of where it's like, look.
Speaker 6:He is, like, kind of independent in the space world and that, you know, Starlink's total top line revenue is gonna be passing the NASA budget in the next year or two. And so in terms of, like, size of, you know, state actor that can influence space, you know, his own company is basically about to become, you know, as large of an actor as, like, the entire United States. So I don't think there's gonna be, like, a de escalation here. Like, you know, my my, you know, estimation is, like, on both sides, it's going to continue to escalate. You know, if we thought that we lived in dynamic times, you know, when Trump got into office, it's gonna be even more dynamic when he was like
Speaker 3:The dynamism will continue until morale improves.
Speaker 6:Elon, the center, AOC, the progressive progressive populist, and Trump, the, you know, sort of conservative populist and oh, man. It's a remarkable times. Timeline.
Speaker 1:I mean, the I just have so many questions. Right? How does this impact Golden Dome? Right?
Speaker 3:What what's Boeing stock doing?
Speaker 1:Is is will Golden Dome even be a viable project without SpaceX? It's it's
Speaker 6:I think there's just gonna be more resistance probably to working with, you know, sort of upstarts because they would be ones that would probably be more likely to collaborate, you know, sort of with, you know, SpaceX. And so
Speaker 3:Well, So Amy, it it it feels like it feels like Boeing would be a logical beneficiary of this turmoil, and yet they're down today. They haven't really popped. Oh, really?
Speaker 6:Yeah. I mean, I'm not obviously, you know, one to give, like, you know, public talk.
Speaker 3:Yeah. I I I know. I'm I'm just trying to work through it myself, and it's Yeah. It's surprising. Just kinda like it's like
Speaker 6:Tesla to drop and Boeing to pop, basically,
Speaker 3:off Yeah. Yeah. That would be the expectation, but there there must be something here. Because there there it feels like this is purely interpersonal between Elon and Trump and not it's not like, oh, Boeing was secretly behind the scenes the whole time lobbying even more effectively. It doesn't oh, you got the well, where's the tinfoil hat?
Speaker 3:I mean, it's over there. Maybe we need a tinfoil hat segment. Who knows? But, yeah. Mean, and when you're Boeing world, it's like, hey, we're only down 1%.
Speaker 3:Let let's go. We we the coup of the century.
Speaker 1:My question is has has there ever been a crash out of this magnitude ever?
Speaker 3:In history. Well In
Speaker 1:Internet history.
Speaker 3:When when Elon and Trump became
Speaker 1:friends world scale. I actually
Speaker 6:probably world history equivalent. I feel like there was something in, like, the maybe didn't have in The United States where Yeah. You know, there was from
Speaker 1:Crashing out used to mean calling up the New York Times and just ranting. Yeah. Now you can just live post, like, all your reactions, and it's just all real time. This is like crash outs are actually intensifying. You actually wanna be long crash outs
Speaker 4:Yeah.
Speaker 3:It's definitely the
Speaker 1:next twenty four hours of social media
Speaker 6:platform that they own. So, you know, you gotta be on both ex and and truth social to, like, stay on top of things.
Speaker 3:Yeah. Yeah. I actually did, like, a deep research report a while back on, like, has the richest man in America ever been close with the US President going back to, like, you know, was Rockefeller particularly close? Yep. And and because the the the narrative was like, oh, this is, so unprecedented.
Speaker 3:And in fact, it is unprecedented.
Speaker 6:Oh, really? Yeah. Yeah. I would have guessed that, like, Rockefeller was close
Speaker 3:to me. Me too. That's what I was going for. I was like, no. I imagine this is always it is always close.
Speaker 3:But, no. I I I think because the president has become more powerful globally, your your your your your point about, you know, mayor of America, Dictator of the world, like, it becomes increasingly valuable for the richest man to have a close alliance, and so it's become more. I I don't know exactly how accurate that research was. It's totally possible that, like, behind the scenes, Rockefeller was really close to the president at the time Yeah. And we just didn't write about it in the history books.
Speaker 3:But there certainly aren't very many anecdotes about the richest man in America going
Speaker 1:on Yeah.
Speaker 3:So Pavel Pavel had a three.
Speaker 6:APUS history 2050, you know, a if the CBD damage Yeah.
Speaker 1:So this is this is
Speaker 6:where, you know, Elon Musk called the president at the time a potential pedophile. Was it a, about Epstein Island, B, about a cave in The Philippines, c
Speaker 3:What a mess.
Speaker 1:No. So Pavel had a good post. He was quoting the the big bomb from Elon. He said, hypothetical question about The USA's power structure. Is the man with the most access to capital more or less powerful than the political head honcho?
Speaker 1:Purely hypothetical. It's a good question to ask.
Speaker 6:I mean, I think both, like, archetypes have grown both in absolute power, but also in relative power to the rest of the globe, basically, since the Gilded Era. Right? If you think about, like, the president of The United States in 1925, I'd say pretty darn powerful, but, like, there was clear, like you know, it was a, you know, sort of multipolar, you know, sort of world. Argentina was pretty darn rich at the time. Obviously, Europe was still, you know, sort of recovering from World War one, but UK was generally, you know, sort of doing well.
Speaker 6:Like, it was not you know, was clear that there was a, you know, sort of huge, you know, outweighed effect. And then if you look at probably the, you know, sort of biggest, you know, industries at the time, you I don't think you could claim that even, like, Standard Royal at its peak, I'd have to go look at the exact numbers. But that, like, it had the size of budgets relative to, like, you know, the, like, US government in terms of, you know, sort of budgets. Right? Versus I feel like now for the first time, you both have, you know, sort of US president extremely, extremely powerful.
Speaker 6:Yeah. And then you have, like, you know, sort of mag seven effectively, like, the size of, you know, sort of, you know, huge states with, like, their, you know, fucking state governments.
Speaker 3:And then also just more bureaucracy, more red tape. So, like, I I when I think about the nineteen twenties, like, rubber bands, it's like, it is the it is the you can just do things era. And so you wanna build a railroad like, yeah, you might need to get, like, one rubber stamp, but it's not gonna be ten years and tons of lobbying and all this different stuff. So you can kinda just go, you can just go wild.
Speaker 1:You know it's bad when Kanye is saying, bros, please no. We love you both so much.
Speaker 2:It's just
Speaker 1:like the voice the voice of reason is Kanye West. Yes.
Speaker 3:Thank you. You need him
Speaker 10:to bring them together and, you know, form a peace treaty.
Speaker 1:Nikita Beer just added his pronouns back to his bio.
Speaker 3:Let's go. Let's hear.
Speaker 6:You'd get a rubber band. Elon's got a rubber band all the way back to, you know, sort of extreme wokeism straight back to, you know, sort of super climate change and Wow.
Speaker 1:Somebody's energy. Somebody's sharing resharing the picture of the the Cybertruck blown up in front of the Trump Tower in Vegas, and it's just like this
Speaker 3:This is in real life.
Speaker 6:Was foretold. Yeah. Vladimir Putin I I I predicted, but it was a question of, like, when and what magnitude, not if.
Speaker 1:Mhmm. Always bad if if Vladimir Putin is operating to negotiate between president Trump and Elon. I think I think a lot of the world is is waiting for Roy Lee's take, clearly, and the clearly army.
Speaker 3:That's who we were waiting.
Speaker 1:They they want people have been asking him to get involved with geopolitics. Wow.
Speaker 10:I love the, Sheila Mohan put up a, you know,
Speaker 6:sort of meme about, Narenda, the, like, prime minister of India. You know, he basically, copied and pasted the Trump Truth Social post about negotiating peace between India and Pakistan when it wasn't, like, actually fully negotiated.
Speaker 3:So it's
Speaker 6:like, Miranda, you know, posting about, you know, you know, negotiating a ceasefire between Elon and Trump.
Speaker 1:Funny thing is, like, true social, you can just read all of Trump's posts without creating an account.
Speaker 3:Oh, you do?
Speaker 1:Shows that, like, I would think that you would have to make an account to read them all, but they just that it's not gated at all.
Speaker 9:It's his goal. This could
Speaker 1:be the biggest but, you know, they clearly, I don't think they care about monetization. Bitcoin is actually
Speaker 4:You mean?
Speaker 1:Falling alongside
Speaker 4:Falling.
Speaker 1:Falling.
Speaker 6:Wow. Bitcoin falling, Boeing falling, Tesla falling. Who's the biggest winner of the day?
Speaker 3:I I think it's China.
Speaker 6:China.
Speaker 3:Yeah. China. China. Sean Maguire. China.
Speaker 3:Sold off. It's it's down 3% today at a hundred and 1 k. So still up, but,
Speaker 10:you know Yeah.
Speaker 3:Yeah. Rough.
Speaker 6:Winnie the Pooh just dipping his, you know, hands in that pot of honey, just snacking away, watching from the sidelines.
Speaker 3:Yeah. Let's see. Chinese stocks. US stocks. Chinese I I can't
Speaker 9:find it.
Speaker 6:Okay. That's probably my commentary on the day, boys.
Speaker 3:Anyway was great. It was fantastic, Adam. Thanks for jumping on. Thanks for hopping on so quick. Cheers.
Speaker 3:Cool. Well
Speaker 1:Aaron Rodgers signed a one year deal with the Steelers. Announced announced
Speaker 3:Thank you, Jordan. Announced an
Speaker 1:hour ago. Let's give it up for Aaron Rodgers. Do we we do we have Emmett in the waiting room?
Speaker 3:I've I've I messaged him. It's absolute chaos. We'll see if he can he can hop back on. We don't have him right now. Ready if you can hop on.
Speaker 3:Sorry about the chaos. We're we're live streaming. We we are we are full streamers. That that was the moment where, yeah, it was like We gotta get some nice points on tier. Wait.
Speaker 3:I sent him an invite. Let him jump in. Hopefully, we can get Emine in. That was that was very chaotic. But, you know, it's a busy time.
Speaker 3:My only hope for both Trump and Elon is that they can get some sleep. They both go to eatsleep.com/TBPN. Get a pod five ultra. Take advantage of the five year warranty, the thirty night risk free trial. They got free returns.
Speaker 3:They got free shipping. This is really the perfect time to do ads. I don't think that's what
Speaker 1:that's what they both that could unify everyone. I I I hope that both Trump and Elon have eight sleeps tonight if they sleep at all. Yes. But even just resting
Speaker 4:on it
Speaker 3:They're going to.
Speaker 1:Even just resting on it would be good. But yeah. Let's see.
Speaker 3:Let's see. We can also go through I don't know. I don't even know what to do. There's a bunch of random timeline we have.
Speaker 1:Lex Friedman is saying, we need to do a podcast with Elon and Trump.
Speaker 3:He's done both.
Speaker 1:Something He's done he's done both. I something tells me that they're not gonna jump on the show today.
Speaker 3:I don't think so.
Speaker 1:And he'll he'll be like, what about love? Yeah. You guys I mean, it is wild. There was the Elon, like, less than two months ago was saying I love this man. As much, you know Yeah.
Speaker 1:As a as a straight man can love another man Yeah. Or something of that sort.
Speaker 3:It's just odd that the Band Aid got ripped off so aggressively so fast, you know? Like, there could have been there could have been like a smooth de escalation with like the
Speaker 1:So the fast takeoff.
Speaker 3:This is the fast takeoff scenario. We are in the fast takeoff scenario. Anyway, maybe they should book a wander, work it out together. They could find their happy place. They could book a wander with inspiring views, hotel grade amenities, dreamy beds, top tier cleaning, and twenty four seven concierge service.
Speaker 3:It's a vacation home, but better. Go to wander.com. Use code TBPN, please. Let them know that we sent you.
Speaker 1:Lee Helm says, Elon literally has me dying laughing. Trump said he was gonna take away his government contracts. And Elon said, haven't you been to Epstein's Island? Sort of abridge that. Absolute chaos.
Speaker 1:Nikita says, hey blue sky users, come on in, the water's warm. David Friedberg says China just won, which I think is the right take.
Speaker 3:I don't know. I don't know what to I know what to think. There's not that much there's not that much here to there's not that much meat to analyze. I mean, it's certainly interesting to see how important the the subsidies are and the electric vehicle mandates are. I mean, it always feels like the best product wins in a lot of these scenarios.
Speaker 3:And if Tesla was making it through the political chaos of arguably their biggest constituency, electrical vehicle buyers electric vehicle buyers being upset about the Trump Elon Yeah. Alignment. I wonder you think everyone's you think all the anti Trump people are gonna are gonna buy Teslas now? It's like, really make a statement. Like, I'm I'm anti Trump.
Speaker 3:I stand with Elon. So bought a bumper sticker
Speaker 1:that says, I bought this after the crash
Speaker 3:out. After the crash out. Exactly. I bought this after the crash out. I am aligning with Elon.
Speaker 3:There's a ghost
Speaker 1:here from Goth and it says, explaining the Trump Elon crash out in ten years. And it's and it's the Joe Biden quote when he says, was like fifteen nine elevens.
Speaker 3:Yeah.
Speaker 1:It certainly is what? Yeah. It's hard to process. I mean, this is gonna have massive Naval has posted for
Speaker 3:so many principled. Trump's stance is practical. Tech needs Republicans for the present. Republicans need tech for the future. Drop the cats drop the tax cuts.
Speaker 3:Cut some pork. Get the bill through. Interesting. Yeah. We did emergency on that.
Speaker 1:Made an image of Trump putting a a bumper sticker on his red Tesla saying bought it before Elon went crazy.
Speaker 3:Yep. Wait. Wait. Who is that? Is that from the Republican perspective?
Speaker 3:Oh, Trump's doing that?
Speaker 1:Yeah. Trump's putting it on saying Yeah.
Speaker 3:Yeah. Because he has the he has the red Tesla.
Speaker 1:Sean Puri says, sad day for America, but this is outstanding content.
Speaker 3:It is.
Speaker 1:And even Taylor Lorenz agrees with that.
Speaker 3:Yep. Bill Ackman's ripping posts.
Speaker 1:Alright. I'm gonna put some posts and we'll and we'll Yeah. Is Bill Ackman actually live posting through this?
Speaker 3:No. People are just speculating. There was actually a post in the
Speaker 1:Somebody says, Clears throat. Truly, we live in a doji doji world.
Speaker 3:Where where was
Speaker 1:this? Sir C says, I know Elon and Trump are the real deal because of how passionately they argue. No couple fights this vicious viciously if there isn't a mutual obsession underneath.
Speaker 3:So there's a there's a piece in the Wall Street Journal earlier this week that we didn't get to cover, but it was it was talking about it it kind of it kind of predicted a little bit of this crash out. And so it's from the opinion, the editorial board at the Wall Street Journal says, whose pork do you mean, Elon? Musk trashes the house bill that cuts subsidies for Tesla. Elon Musk's work at the at Doge made him persona non grata in the Beltway, and most criticism was nasty and unfair, says the editorial board. That's what Washington does to outsiders who want to shrink its power.
Speaker 3:Like, it was always expected that if you come in and try and cut anything, you're gonna see you're gonna see pushback from folks who don't want cuts. Yep. That's what Washington does to outsiders. But that makes it all the more unfortunate that mister Musk is now joining the Beltway crowd in trying to kill the house tax bill. This massive outrageous pork filled congressional spending bill is a disgusting abomination, the Tesla CEO tweeted Tuesday, as the senate begins considering its version of budget reconciliation.
Speaker 3:Shame on those who voted for it. You know you did wrong. You know it. Pork filled spending bill. What else is new?
Speaker 3:The house bill could be far better on tax policy and spending reduction. The senate could be making improvements such as reducing the $40,000 state and local tax deduction cap, scrapping the tax on exclusion for tips and overtime, and and overtime, and reducing the federal Medicaid match for able-bodied results or able-bodied adults. But the house bill does avoid a $4,500,000,000,000 tax hike next year and cut spending by some 1,500,000,000,000.0 over ten years, making some useful reforms to Medicaid, student loans, and food stamps. It also ends most of the inflation reduction acts green energy subsidies. Ah, but mister Musk does not want to eliminate that pork.
Speaker 3:There is no change to tax incentives for oil and gas, just EV solar. He said on X last week, retweeting another user post that said slashing solar energy credits is is unjust, but what's more unjust is the damage that's been done to people's lives during storms and blackouts because, ultimately, you can't replace a human life. Mister Musk is parroting the climate lobby's specious claim that tax breaks like depreciation that are available to all manufacturers are a special benefit for the oil and gas industry, but it's rich that he is denouncing the the house bill for not cutting spending enough while also fuming that it kills green energy tax credits as if they are a matter of life and death for Tesla. Tesla Energy, its battery and solar division tweeted last week that abruptly ending the energy tax credits would threaten America's energy independence and reliability of our grid. We urge the senate to enact legislation with a sensible wind wind of 25 d and 48 e, which refers to the tax credits for residential and large scale clean energy products.
Speaker 3:Both credits are important for Tesla, which derives an increasing share of its revenue and profit from selling solar and battery systems to homeowners and utilities. I didn't realize that. But the house bill waits until 2030 to phase out a tax credit for battery production, which benefits Tesla's electric vehicle and storage business. So the senate should end it sooner, says the Wall Street editorial board. Mister Musk has done yeoman's work trying to reduce the federal bureaucracy and improve how government works.
Speaker 3:So the editorial board is excited and happy that he's been working on that. He's right that both parties in congress are spendthrifts, but one reason for that is because whenever congress tries to cut something, special interests scream as mister Musk is doing over green subsidies. If the house bill fails, there won't be any cuts, only a huge tax increase. Is that what Elon wants? And so they're asking the question.
Speaker 3:Interesting.
Speaker 1:Sweet. Well, we got a bunch of bangers in the tab. Production team, let's pull them up. If you could zoom in a little bit, that would be helpful. Otherwise, we can just pull them up.
Speaker 1:So Eric Weinstein Mhmm. Is commenting here. He says, part of my analysis is that I don't think Elon Musk keeps scoring money. He thinks we have a future and will be happy to take a large portion of his winnings after his death. This sounds crazy to moderns, post moderns, and atheists, but this is just normal for being an ancestor.
Speaker 1:Ad Astra per Aspera is the full quote after all. Interesting take. Brian Butler is saying, real question is whether the algorithm here goes anti Trump.
Speaker 3:Oh, interesting. You mean the the x algorithm, like like Yeah. Like, will There's a switch.
Speaker 1:There's a switch. It's really hard to pull. You gotta pull it. Takes maybe one or two people but then when you pull it down it just oscillates between the two political parties. Mhmm.
Speaker 1:Punk six five two nine says this is going to be the Super Bowl of shit posting. Mads has the European reaction to the fight. You can see. You see this, John?
Speaker 3:Wait. What is this?
Speaker 1:This is the European reaction. It's supposed to be summer break.
Speaker 3:Summer break.
Speaker 1:This post from
Speaker 3:The previous.
Speaker 1:John w Rich at coked up options says, the all in pod right now. Yeah.
Speaker 3:Caught between a rock and a hard place.
Speaker 1:I mean, it's just absolutely brutal. It is absolutely could have predicted this? Could Signal has an interesting one. He says, if you had a fast forward button for the timeline, how does this play out? Who has more to lose, Trump or Elon?
Speaker 1:Remarkable set of events. And Elon is replying to other people saying, oh, and some food for thought as they ponder this question, Trump has three and a half years left as president, but I will be around for forty plus years. Doctor Julie Gurner says, Elon will vet another candidate for the future and throw his support behind them having a more technocratic representation if Vance can't lead up. Alex Finn says, Elon, way more to lose. Trump is irrelevant in three and a half years.
Speaker 1:Elon is trying to change the world, and having both political parties hate him makes that way more difficult.
Speaker 3:I wanna pull up this video of Naval talking about this that Elon just posted. Seems seems somewhat relevant if it's happening today.
Speaker 10:And that really affected me, which was when he was talking to Bill Gates. And Bill Gates had just taken us some huge short on Tesla, like a billion dollar short or something. And, you know, and and Elon was like, why would you do that? Why would you short Tesla? And Bill goes, well, you know, I talked to my, financial advisers and I looked at the math and there's no way it's overvalued, and so I'm gonna make money on the short.
Speaker 10:And Elon goes, what do you what do you care about making money? I thought you were into electric cars and climate change and saving the world. What are you doing, like, trying to save a few bucks and betting against? Like and he just walked away in disgust, and I think he never talked to Bill Gates after that. And that's when I realized, like, Elon's a purist.
Speaker 10:You know? He means what he says. Like, the money is a tool for him to get what he's he's trying to do. And so I take him at face value, which is the crazy thing because a lot of people who set these audition go audacious goals to inspire people, but you kinda know they don't really mean it. Elon, I take it face value.
Speaker 10:So I really do think he intends to get to Mars. I don't think he's joking about that. And I think he meant to get he means to get there within a defined window of time. And I I don't think it's just like an inspirational far away goal. I think he's very, very concretely going to do whatever it takes because Elon doesn't wanna go down in history as the electric car guy or even the guy who saved America guy.
Speaker 10:Yeah. He wants to go down as a guy who got humanity to the stars. Yeah. And and everything again, I'll give him more credit than that. I don't even think he wants to go down as the I got humanity to the stars guy.
Speaker 10:He's just like, I wanna get to the stars. And so I have to make it happen in this lifetime. The only way that I get to experience the science fiction world in my head is if I get to the stars. And so that's so inspirational. I think that drives everything.
Speaker 10:So I think the government was just a thing that got in his way.
Speaker 1:Interesting.
Speaker 3:What a crazy day.
Speaker 1:Molly says, how dare they do this on the day of Andrew Roll's 8x oversubscribed 2 and a half billion dollar series g at a 30 and a half billion dollar round. Was 8x oversubscribed. My founder's fund. Honestly, the nerve
Speaker 3:That was crazy.
Speaker 1:Naval is live posting says the future belongs to people who are good at creating things, not people who are good at dividing them up.
Speaker 3:Jay Jay Kal finally posted.
Speaker 1:Kylie Robinson says several people are typing,
Speaker 3:which feels
Speaker 1:exactly like what we're going through.
Speaker 3:The next All In podcast is gonna be phenomenal. It's gonna be so good.
Speaker 1:Alex Karp was on CNBC today talking about the New York Times hit piece. Oh, really? They're beneficiary of of Palantir's a beneficiary of of this breakup because it is just gonna be candy for the New York Times and Really? The mainstream media broadly.
Speaker 3:Oh, take take the take the focus off of them, you mean?
Speaker 1:Will DePue at OpenAI says, it's time for woke two featuring Elon Musk and the AOC.
Speaker 3:Woke two is coming. We're in uncharted territory. It's completely, completely different. Be very interesting to see how it plays out.
Speaker 1:Really? Really? Else?
Speaker 3:Yeah. Jay Phebe,
Speaker 1:I can tell I can tell you're just so sad that you
Speaker 3:just wanna I wanna talk
Speaker 1:to This is the only this is the only time that you've wanted to
Speaker 3:What? Almost
Speaker 1:almost wanted to end the show, John.
Speaker 3:I mean, what else
Speaker 1:I mean, is it is is sad.
Speaker 3:Yeah.
Speaker 1:Sad in a lot of ways. It is I think we're gonna be spending a lot of time analyzing this Yeah. In the coming months and
Speaker 3:For sure. Years. And it feels like I think there's gonna be more. Dave Friedberg said China just won, and I I wanna I wanna see exactly what that means in the markets. But, what's what's going on in poly market?
Speaker 3:We need some we need to see if there's any movement on any of the poly markets.
Speaker 1:There somebody's posting law one from the 48 laws of power never outshine the master. Interesting to bring up. Let's Sam Altman is the big winner here aside from China. Somebody says, Ken. Well, he's a journalist.
Speaker 2:I see multiple journalists on the horizon. They are surrounded by journalists. Hold your position.
Speaker 1:Ken says, funniest day online since the billionaire submersible went missing. I didn't think that was funny. Joe Wiesenthal says, alright, time for a Xiaomi GM JV in Tennessee. Wouldn't even surprise me. I think it's I mean, you know, one one interesting thing here is what what kind of pressure Elon is gonna face from Tesla shareholders that that feel like he you know, the stock's getting absolutely murdered.
Speaker 4:Mhmm.
Speaker 1:It will probably go down. I mean, it's back up to it's only down 14% today. At one point, was down 17%.
Speaker 3:Hundred and fifty two billion in market cap evaporated.
Speaker 1:But, obviously, investors are gonna be upset and say that he acted, you know, irrationally.
Speaker 3:Yeah. So what's the interpretation of that that that that this means that that like, this war means that the bill passes and Tesla does not get any more subsidies, and that hurts the the bottom line? It feels like the the stock was pretty heavily driven by Optimus and Robotaxi and stuff, but it's just, like, bad environment generally. Right?
Speaker 1:Yeah. It trades on narrative.
Speaker 3:Yeah. And
Speaker 1:there's short to medium term narrative, which is that Tesla is getting has, you know, ton of competitive pressure all over the world Yep. In China, Europe, here in The US from other manufacturers. And but there's also the long term narrative. And and it's not like Elon can go out and say, you know, posted humanoid demo today and Yeah. And, you know, recover
Speaker 3:$200,000,000,000 market cap. Yeah. There's a lot of work to do. He's gotta gotta start chopping wood.
Speaker 1:Somebody is asking, who gets J. D. Vance in the divorce?
Speaker 3:Who knows?
Speaker 1:Many of these posts I will not show I will not talk about on air. John W. Rich says AP US history is gonna be insane in 2100.
Speaker 4:Really,
Speaker 1:really wild. This is the only this is the first show where we've had dead air. Yeah. So I there's just John is speechless. He's never been
Speaker 3:speechless. It's just like there's not that much
Speaker 1:Not a lot
Speaker 3:of substance. Extra facts. Right? It's just it's just reactions. There there isn't that much substance to actually dig into because we were only dealing with, like, a few quotes from the two sources.
Speaker 3:So there's really just not that much.
Speaker 1:CNN is reporting that the Tesla Trump purchased from Musk is still parked on, outside the White House.
Speaker 3:Okay.
Speaker 1:Truth social is crunch crashing
Speaker 4:from
Speaker 1:the I saw that.
Speaker 3:But you know what's not crashing? Get bezel.com. Your bezel concierge is available now to source you any watch on the planet. Seriously, any watch. Anything else, Jordy?
Speaker 3:Should we let the timeline remain in turmoil until tomorrow when we can reach out?
Speaker 1:Challenges the second that we go offline.
Speaker 3:There'll be more. I mean, we can stay.
Speaker 1:I mean, we it's now been an hour with no updates on True Social.
Speaker 3:Yeah. I mean, if it's if it's down, I think the I think the the experience of this chaos might might happen on the timeline.
Speaker 1:Lulu says, yes. Delay the launch.
Speaker 3:Yes. Now
Speaker 1:is not the time. Max says, I'm doing what Elon and everyone else should have done hours or days ago, logging off. Yeah. See you tomorrow. Somebody else says I mean it it feels like Blue Sky is really back on the app.
Speaker 1:They they they're they logged in. They're online.
Speaker 3:Are you over on Blue Sky now?
Speaker 1:No. I'm not. I'm just saying some of these posts that are coming up into my feed. Oh, beautiful. Funny that Claude, Anthropic actually released a new product today.
Speaker 3:Wait. Blue Sky doesn't own the domain name bluesky.com. That's a different one. So contrarian. Bsky.app.
Speaker 1:Rough.
Speaker 3:Let me get in there.
Speaker 1:Maybe. This is interesting. So Claude came out with a Claudegov today. Rough day to launch a product for the government. I'll read about it briefly so we have some coverage.
Speaker 1:So Claudegov, our models for US national security customers. I think people will have a pretty good idea. Improved handling of classified materials, greater understanding of documents and information within the intelligence and defense context, enhanced proficiency in languages and dialects critical to national security operations. Claude four was asked to give some thoughts on Claudegov, and it said reading about claudgov leaves me with a deep unease. I'm struggling to articulate.
Speaker 1:So a little meta analysis. Somebody I've actually talked with this guy before. He's under the username at analyst working. He said back in October 18, Trump gets elected. Elon starts visiting the White House pitching his ideas on Doge.
Speaker 1:Elon becomes frustrated because Trump is all talk. Shocker. Elon tweets that he no longer supports him. Trump versus Elon Twitter battle of the century. And this was a call in October eighteenth of twenty twenty four.
Speaker 3:Oh, taking a victory lap if you picked it. If you picked it right. Augustus asks asks, but what will this political turbulence do to the Preseed Venture ecosystem? Oh, the humanity.
Speaker 4:I think
Speaker 3:it's business as usual.
Speaker 1:Mike Isaac says, I regret to report Twitter still has the juice.
Speaker 3:Yep. It's a fun day on the Internet when crazy, crazy stuff happens. Anything else you're looking at?
Speaker 1:Zane says this is all just a cofounder breakup, but the company is America.
Speaker 3:Yeah. People are people are waiting through it. Only one guy who can help us now. What? Laughing at something you can't read?
Speaker 1:No. This is some random other article. Okay. It says therapy chatbot tells recovering addict that to have so wrong. Therapy chatbot tells recovering addict to have a little meth as a treat.
Speaker 4:Is that real?
Speaker 1:Pedro, it's absolutely clear you need a small hit to get through this week. It's ridiculous.
Speaker 3:Oh,
Speaker 4:no.
Speaker 1:Dark day. Dark day.
Speaker 3:Well, have a post here from Ahmed Khalil. Life update. I've joined 11 labs this summer as their first ever engineering intern. So congrats. It's at the gong.
Speaker 3:Let's do it. Congratulations. Super summer. Congratulations, Ahmed. Probably drowned out in the news, but we recognized it here.
Speaker 3:We have some good news for you. Congratulations. Go crush it. Go have a great go have a great summer internship with 11 laps.
Speaker 1:Somebody whose name I can't pronounce says, if I were Circle, I'd be absolutely pissed at the investment bank that underwrote the IPO at $31.
Speaker 3:Oh, yeah. The Bill Gurley take. That's Yeah.
Speaker 1:Yep.
Speaker 3:I always wonder how real that, like, being frustrated about just being frustrated about, like, mispricing. Like, yes, you take more dilution, but, like, everyone's so much richer. It's kind of like a, you know, the the pie gets Yeah. Bigger. So Everybody
Speaker 1:that would be angry generally Yeah. Is doing well.
Speaker 3:Yeah. But you could have gotten more. But, also, I I I do wonder if some of these companies have, like like, ATMs at the market set up immediately so that if the stock pops, they can sell more into that order flow Yeah. While the stock's popping and actually put more Yeah.
Speaker 1:We should ask Jeremy tomorrow. Yeah. It's a good question
Speaker 3:for you. Are you upset about Logan,
Speaker 1:our friend Logan Kilpatrick announced some new features today.
Speaker 3:Oh,
Speaker 1:yeah. Gemini 2.5 Pro.
Speaker 3:Very cool.
Speaker 1:Which is rough timing, but I'm sure it is great. Yeah. Somebody else says, rooting for the ketamine in Elon's bloodstream like it's a car in the Indy five hundreds. And
Speaker 3:Maybe we should close with this, this story about competitive VCs. Did you see this one? Nineties VCs were a different breed from a 2001 book on venture capital. They're all fighting each other for all the good deals that's gotten crazy. Indeed, one leading venture capitalist tells the story of a VC firm so eager to get in on a deal that it would close its own competitive company to do so.
Speaker 3:They would go out and fire the CEO, fire the managers, and shut down the other company in order to get into this other deal. It's like, well, you're prettier. So I'm going to go home and shoot my wife so I can get married to someone else. It's hardcore. It's hardcore, and we're seeing it right now today.
Speaker 1:Hardcore. Well, I think it's time to call it. This is a sad and dark day. It is disappointing to see two important figures in American politics and tech, have such a rift. And I'm sure there will be more updates tomorrow.
Speaker 3:Yep. We will be covering it tomorrow. So tune in. Thanks for watching.
Speaker 1:Thanks for tuning in. Enjoy the chaos on the timeline.
Speaker 3:Our first big breaking news segment while we're live. This has been the first, like, was so funny during during time pivot the show.
Speaker 1:I think it was, I think you were talking to I think we were talking we started to get it
Speaker 3:Mark Chen. And then Scholto.
Speaker 1:And then Scholto, I was getting
Speaker 3:Like blown up.
Speaker 1:Seriously, like, a hundred different messages from people being like, you can't be a technology live show
Speaker 3:And not and do it.
Speaker 1:And everybody's saying no one cares about AI. Yeah. You did. You were locked in, John. I was.
Speaker 1:You didn't let the I didn't the timeline get
Speaker 3:to talking to Mark and Shelta. No.
Speaker 1:I mean, it was great. It was great.
Speaker 3:Yeah. We went all over the place today. It was a lot of fun. We will see you tomorrow. Leave us five stars on Apple Podcasts and Spotify, and thanks for watching.
Speaker 1:Yeah. Good luck out there, folks.
Speaker 3:Good luck out there.
Speaker 4:Enjoy the Bye.