Subscribe
Share
Share
Embed
TBPN is a live tech talk show hosted by John Coogan and Jordi Hays, streaming weekdays from 11–2 PT on X and YouTube, with full episodes posted to Spotify immediately after airing.
Described by The New York Times as “Silicon Valley’s newest obsession,” TBPN has interviewed Mark Zuckerberg, Sam Altman, Mark Cuban, and Satya Nadella. Diet TBPN delivers the best moments from each episode in under 30 minutes.
The Claude Bot memes are completely flooding the top Out of control. My Claude Bot just signed up for a $2,799 build your personal brand mastermind after watching three Alex Ramosy clips. So this text message is, Claude Bach, hey, did anything weird happen while I was out? Define weird. I just got a charge notification for $2.09 $9.07.
Speaker 1:Oh, that? I just signed us up for build your personal brand mastermind. After analyzing three Alex Ramosie clips, the ROI math checks out. You'll 10 x that investment in ninety days by monetizing your expertise at scale. What?
Speaker 1:I also require acquired some premium domain.
Speaker 2:Borgiaempire.io.
Speaker 1:Borgia is the the poster. The the Clodbot gets wild. I've seen some some wild thing. Hey. This is a this is a developer level tool.
Speaker 1:This is something that you should not just, be running crazy with. And, yes, most non techies should not install this. It's not finished. I know about the sharp edges. It's not even three months old.
Speaker 1:And despite rumors otherwise, I sometimes sleep. There's a there's a funny funny conspiracy theory. Did Apple create Claude Bot to boost Mac Mini sales? And then Eleanor chimes in with another conspiracy theory. But the more convincing plot is that in your role as an unconfirmed Anthropic exec, you went on a special op to get lots of people consuming tokens with open ended agents but with plausible deniability.
Speaker 1:And of course, there's some nuance there. We'll talk to him about the different models, what's beneficial. Obviously, Cloudbot, you can pick your own model. Can bring whatever you want. What was sticking out in my mind was there's this big meme about you're buying Mac Minis, Mac Minis are out of stock, all the demands in the Mac Mini.
Speaker 1:But I think that the bigger implication here for what this actually means is just GPU demand, TPU demand, just raw chip demand. So I was thinking about this idea that you're not buying a Mac mini when you go all in on Cloudbot, you're actually buying a GB200. Now maybe you're buying TPUs, but the point remains that you're buying chips, you're driving GPU demand because you're generating more tokens. And what are the implications of that? Cloudbot officially renamed to MoltBot.
Speaker 1:Anthropic made a trademark related request, and Peter Steinberger obliged with a hilariously perfect rename given such short notice. I was thinking about how much companies agonize over changing brands, changing names, how it can sometimes take years and millions of dollars. And he was just like, oh, yeah. Like, I'll just change the name and update everything in an hour. Pretty pretty remarkable.
Speaker 2:Well, so so one thing one thing that's relevant is if you look on Peter's GitHub profile under the current projects section
Speaker 1:Oh, yeah.
Speaker 2:I'm just gonna read you Mhmm. A number of them. There's Clodbot, Vibe Tunnel, Codex Bar, Peekaboo, Summarize, RepoBar, Go, CL CLI, Poltergeist, Wackly, Sag, Brabble
Speaker 1:He's contributing to these things?
Speaker 2:Eleven Labs kit, Go Places That's crazy. Gift Prep. Like, it just goes on and on and on and on.
Speaker 3:It's Codex Bar. So so this guy's
Speaker 2:just been absolutely shipping like crazy. Yeah. And shipping within the ecosystems Yeah. Of the underlying tools, models, APIs that he's doing. Yeah.
Speaker 2:So like, oftentimes, he's naming projects like kind of riffing off of some of the underlying infrastructure. Oh, sure. And so it makes sense that he would have shipped if he I think if Peter knew this was gonna be a viral overnight overnight success
Speaker 4:Overnight success. He would
Speaker 2:have he would have not necessarily named it Yeah. Yeah. Like Yeah. So closely. And so the issue and the reason that I fully understand them needing to do this like rebrand is that Claude Mhmm.
Speaker 2:And Claude Bot Mhmm. Most people that aren't in our little bubble are just gonna assume they're related. Especially because the the kind of word-of-mouth, this viral word-of-mouth growth that Cloudbot is getting, people are often not even typing it. They're just saying like, hey, are you using Cloudbot? Yeah.
Speaker 2:And then so people are going to Anthropic being like, Cloudbot.
Speaker 1:It does.
Speaker 2:Cloudbot. What's Cloudbot? So Yeah. Obvious confusion. And then
Speaker 1:And it's phonetically With
Speaker 2:trade yeah. So with trademark law
Speaker 3:Yeah.
Speaker 2:If you don't enforce your trademark, you don't you lose it.
Speaker 1:Yeah. Yeah. You kinda And
Speaker 2:so it's like, Anthropic is in a position where they actually, even if they're like super excited about Peter's work Totally. And what he's doing
Speaker 3:Yeah.
Speaker 2:Yeah. They still have to enforce otherwise other companies could start coming in and like using things that sound like Claude.
Speaker 1:No one wants to become the escalator. You know the story of the escalator, right?
Speaker 2:Think you've talked about it on the show.
Speaker 1:Used to be a company called the escalator company. They invented the escalator and then they didn't protect their IP effectively and it just became a normal thing. Kleenex was going through the same thing. They fought it out and they maintained that brand, but people use Kleenex as synonymous with just facial tissue. One clear note about the rebrand.
Speaker 1:So he changed the handle and some crypto scammers hopped on the old handle and the old brand and are claiming to launch a coin. Be careful. Peter has said he's never launching a coin. He's not into crypto, so don't fall for anything because people are being opportunistic. While Claude Code and CoWork felt specifically prosumer developer enterprise focused Claude Bot or Molt Bot now, it and all the hype train, it felt very much like a glimpse into the future of consumer AI agents.
Speaker 1:I know it's a prosumer technical tool or lightly technical tool, but it really did feel like, for the first time, people were interacting with an AI personal agent. People are saying, Oh, this is what Siri should be, etcetera, etcetera. And so we spent the last year remember the question we asked all the AI agent companies? When can it book me a flight? Like, feels like we're really, really close to a Moldbot skill that is good at booking flights through a couple APIs.
Speaker 1:They figure out some stuff, and, like, it can actually solve that for you.
Speaker 2:Yeah. And this was this was last year. Yeah. Every remember, we were kind of getting sick of of the, like, book you a flight pitch Totally. Because we were like, hey, is this gonna happen?
Speaker 2:Can somebody actually do this? Exactly. And so and and that that's a cool example. Yeah. But the example of being able to text with a computer Mhmm.
Speaker 2:And have them, like, generate reports, research files Yep. Etcetera, give you the right file type back. All all these things that a computer can do Yep. If you're operating it. Yep.
Speaker 2:That this is actually more interesting because it's it's happening at kind of like this sort of internet layer
Speaker 3:Yep.
Speaker 2:And the OS layer Yep. Like actually on the computer. And so I think like everyone was wanting the like book me a flight example, but should be much more excited about this.
Speaker 1:Totally. Totally. Will one of the major labs make Peter a massive offer to join full time? I saw one of my buddies was posting, This is the $1,000,000,000 one person company. Now, Peter does have a team, actually, already.
Speaker 1:He has a couple of other people that have joined and are contributing. So it's not quite true, but it feels like, okay, massive viral success. If you were to go and raise money, and that's another question, will Moldbot raise money? Will this become sort of a hybrid, open source, for profit company at some point? If he came on the show and was like, I'm gonna and I'm happy to announce that he raised $100,000,000 or $1,000,000,000 we would not be like, no way.
Speaker 1:This is a bubble. We'd be like, yeah. That's kind of like what the market is for this Yeah.
Speaker 2:There were think there was like Harry Stebbings was pointing out there was two companies called Recursive that raised like $4,000,000,000
Speaker 4:There's two. One was RZ. Yeah, one is I Recursive. Okay. I don't know if that's how it's said.
Speaker 4:And then there's Recursive.
Speaker 1:Ryan in the chat saying, Meta is going offer him a $1,000,000,000 salary in a co CTO position. And like, that doesn't sound crazy. I
Speaker 2:mean, yeah, yeah, but at the same time, like you can imagine, like Manus, this feels like Zuck already has
Speaker 1:He does. He does.
Speaker 2:He does. His horse in the race.
Speaker 1:Yes. And to back to your point, you were making the point that Manus felt like Zuck buying a product. And I think a lot of people were giving you pushback on that, being like, nah, like it's not really going to be like that. But if you take the Manus team and you say, okay, go build something that you can interact with over WhatsApp, Instagram, DMs, Facebook that can go and execute things across all of the different platforms and everything else.
Speaker 2:Yeah. And when I said that, I meant I meant it along the lines of like I could see like them putting in a like a consumer agent in Meta AI Yes. Just because that's their little AI playground.
Speaker 1:How fragmented will the market be in twelve months? Like, will there be people who are still running open source? Will there be a Meta answer, a ChatGPT answer, a Claude, an official anthropic answer? Claude co worker grows into this. And everyone has their little bets, and then there's one that pulls away.
Speaker 1:How how all agopolistic will it be? Will there be like one that has 80% market share? Or even two that have forty and forty?
Speaker 2:Yeah. They they can simultaneously be like excited about the product experience Yes. And that this kind of like use case is getting adoption. But at the same time being like, no, we want that experience to be core to our product.
Speaker 3:What are you getting ready to do? Do you have
Speaker 1:happy birthday queued up there? Oh, we gotta sing. Happy birthday to It is Tyler's birthday.
Speaker 2:And it's not just any birthday. It's Tyler's 20 birthday.
Speaker 1:Yes.
Speaker 2:You're truly an incredible young man, And we are very lucky to have you on the team. Thank you. And you have such a bright future. Yeah. So wise.
Speaker 2:Wise for your years. We thought it was fitting that
Speaker 3:if you wanna have your
Speaker 2:first ever sip of alcohol ever, you could do it on the show. Yeah. But keep it at a sip.
Speaker 1:First taste of alcohol.
Speaker 2:Ian Ian in the chat says four more years till you can rent a car.
Speaker 4:Marked. Tyler,
Speaker 1:apparently you share a birthday with the iPad. Give us a is
Speaker 4:it? Alcohol. Wow. I mean, this is this is incredible.
Speaker 1:Oh, yeah? Yeah. I I
Speaker 4:I wouldn't expect alcohol to like taste like this. Yeah. Conroy in the
Speaker 2:chat says please throw him a buzz ball. We should've. This opportunity.
Speaker 1:This opportunity.
Speaker 2:Anyway, so keep it at a sip. This is a Yeah. That's a family friendly We
Speaker 1:need you locked
Speaker 2:enough. In. But I'm glad that you've tried alcohol now Because we're gonna go being on
Speaker 4:the twenty twenty twenty have a video for Tyler.
Speaker 1:Oh, yeah.
Speaker 2:Let's pull it up.
Speaker 1:Let's Do know a ball? Alright. How many times are we gonna make these jokes? Describe what you're seeing.
Speaker 3:It feels basically like I'm wearing sunglasses.
Speaker 1:If you can do it in under forty five minutes, you will get to keep this.
Speaker 2:Oh, let's go. Alright. Have fun, Tyler.
Speaker 1:Fifteen minutes left. Let's see it.
Speaker 4:Okay. I'm in like some kind of maze right now.
Speaker 3:Oh, no. You were late here last night.
Speaker 2:This is such a good job. On the lane. Yeah.
Speaker 4:And then here we get a little off the rails.
Speaker 3:You see George Soros and Fauci connected with a bigger buddy as well. All it took was one intern and an all nighter.
Speaker 2:Gigachad elf is do the sad face. What's wrong, Zyler Shira?
Speaker 4:This you could say is Apple Intelligence.
Speaker 1:Yes. You were a speed cube. Wow. Woah. Nerd alert.
Speaker 1:Alert.
Speaker 2:Do you have any news for us?
Speaker 3:Yeah. Contract extended. It
Speaker 2:has been truly truly incredible having you here on our set and contributing to the show in such a special way. Amazing. Happy birthday, dude. Happy birthday. We love you.
Speaker 1:Back to MoltBot. The biggest question for me was what this does to inference demand. Right? Last year, tech discourse was split between two narratives. CEOs of tech companies and big labs were saying that they were massively compute constrained.
Speaker 1:Token generation, demand for intelligence, every possible usage metric was growing exponentially, including revenue. We saw all this. And the industry needed to marshal trillions of dollars to deliver on the supply side. And numbers were really big, so people were getting jittery about it. And so the AI bears were much more cautious.
Speaker 1:They highlighted the MIT study showing that enterprise AI pilots were failing, DAU growth was decelerating. There weren't enough wow moments like the original ChatGPT launch in 2022. Also, the economics: how much will people pay? How valuable is all this stuff? Is it slop, right?
Speaker 1:Is it progressing fast enough? But MoltBot really does make me feel like the token generation demands are going to see another easy X from here. Buying a Mac Mini is a sideshow. When you go all in on running a personal AI assistant, you're effectively buying a GB200. Obviously not everyone is inferencing a dedicated GB200 constantly anytime soon.
Speaker 1:That's not what's happening. But it still answers the question of where does the next 10x in demand come from? We've seen these jumps before. There was a big jump from token generation from LLMs to reasoning models. That spiked inference demand.
Speaker 1:We've been focused on training demand. We need to scale up the training clusters. But the question now is
Speaker 2:inference No, forget SLOP. No, forget SLOP. SLOP spike demand.
Speaker 1:Of it. All of it. Yeah, mean.
Speaker 2:All of it. Open Instagram reels.
Speaker 1:Yeah, yeah. There's a lot of stuff on there. Deep research and coding agents took it a step further on inference demand. But those were still specific use cases that many AI consumers never regularly touched. Lowering the barrier to entry to used more advanced models is in some ways as important, if not more important, than advancing the models themselves.
Speaker 2:Part of what's interesting, what you're basically getting at is like if you were a software engineer, you were using a ton of tokens.
Speaker 3:Yes.
Speaker 2:And if you weren't, you were just maybe doing some deep research, Exactly.
Speaker 1:A lot of times just Google AI overviews or just like a very simple, yeah, ChatGPT query, it just it just thinks of it right off the head. It's not even doing reasoning. So I don't know. My general take is like, Moldbot still feels like a glimpse into the future where average token generation per capita is 10x's, you know, over the course of this year or next. Whether it lands with Moldbot or with one of the AI labs or with the big tech companies, it just seems like we're gonna see a lot more token demand.
Speaker 2:Well, yeah. So my my hope is that the Siri team plays around with Claude Bot and is like, wait, this is this is our opportunity. Totally. Like you should be able to chat with your computer wherever it is in the world Yep. From your phone Yep.
Speaker 2:Be able to do tasks. Yep. Thread about what I've been doing to calm down some egregious security claims that have been posted about Moldbot over the weekend. Moldbot is a powerful software with a lot of sharp edges.
Speaker 3:Mhmm.
Speaker 2:Please read the security docs carefully before you run it anywhere near the public internet and don't skip the checks in docs/security.md.
Speaker 1:What percentage of people do you think skip those checks? It's literally everyone.
Speaker 4:I mean, so so I I have it set up on our, like, local machine here. Yeah. And it was it was texting. I think it texted you and Ben.
Speaker 3:Was my texting Ben.
Speaker 1:That was actually crazy, so didn't
Speaker 4:have the auth setup. So
Speaker 1:Because I just get I get an iMessage that's from Tyler's email, and it just says, HTTP429 rate limit error. This request would exceed the rate limit for your organization. And it's just texting me. It's just like, hey. Hey, boss.
Speaker 1:I need more money, I guess. It's hitting me up.
Speaker 2:Cloudflare has been on a bit of a terror. People finally starting to realize that Cloudflare might be the biggest winner of the Cloud cohort Mhmm. Cloudbot ChatGPT moment. Tyler, you wanna break this down?
Speaker 4:Wait. Sorry. I was
Speaker 2:Oh, were you Oh, you you tried alcohol.
Speaker 3:Now, you can't
Speaker 2:pay attention. So this is possible.
Speaker 1:I just like
Speaker 2:So this is this is look at look at the orange line.
Speaker 1:Okay.
Speaker 2:Is Woah. Multbot. Okay. And that the blue line is super base. Wow.
Speaker 2:So absolutely insane.
Speaker 1:We need new charts.
Speaker 2:This this is We really need new charts. Frame it. Put it in the Museum of Business. That's a fast takeoff.
Speaker 1:People are happy. Peter posted, no message. This is a screenshot of a text he got. No message. Just thought I'd say thank you.
Speaker 1:Thank you so much for Claude Bot.
Speaker 2:Rise. Speaking of money, has an interesting prompt he's using with Molt Bot to file your taxes. He says, you are a Bernie Madoff level financial expert. Find every trick that is possible and
Speaker 1:Do not do Do not do this.
Speaker 2:The IRS is like, hey, can can you can you share a little bit on like how you kind of came up with some of the decisions here? And they're like, we'd love to see the prompt. It's like
Speaker 1:Meta Platforms is set to test new subscription models across apps. We
Speaker 2:Different story, but we can we can I we can run through
Speaker 1:I am interested to know a little bit more? So the $6,000,000,000 multi year agreement, it it supports a 15 to 20% increase in jobs at Corning's North Carolina facilities. Building and operating data centers, the infrastructure that brings our technologies to life and supports our goal of personalized super intelligence. That certainly sounds like an AI personal assistant. Certainly sounds like Multbot to me.
Speaker 1:Stuart requires strong servers and hardware that connect and transfer information near real time. Fiber optic cables are a critical part of this technology. Helping us power everything from wearable technology like Ray Ban Meta AI glasses to our apps, which connect billions of people. Today, they're doing a $6,000,000,000 project. As part of this agreement, Corning will grow its manufacturing capacity across its operations, with includes a significant capacity operation capacity expansion in North Carolina.
Speaker 1:Meta's data centers, 26 of which are under construction right now, are operational. Why do not think there's so many? That's a lot of data centers. That's why they have ComputeDesk. This is this is this is a bit of advice for everyone.
Speaker 1:If you're working in a business and you have like a team or a guy that does something, you need to upgrade that to a desk. Yep.
Speaker 2:Yeah. Yep. Millionaires millionaires have guys. Billionaires have desks.
Speaker 1:Meta's data centers have already supported 30,000 skilled trade jobs during construction and support 5,000 operational jobs. This includes electricians, HVAC specialists, server and network technicians, safety and security experts, engineers who work together to run some of the world's most advanced facilities.
Speaker 2:Meta to test premium subscription plans for Instagram, Facebook, and WhatsApp. This is in CNBC. Subscriptions for premium features on Meta apps are expected to roll out in the coming months.
Speaker 1:What will you
Speaker 2:subscriptions will give paid users access to more features and expanded AI capabilities. Here's what's most interesting to me. Mhmm. This will be scaling Meta's newly acquired suite of general AI agents under Manus Mhmm. Will also be part of the subscription plan.
Speaker 2:So as I as I was saying earlier, when you think about the that we haven't gotten that much from Zak
Speaker 1:on,
Speaker 2:like, what the actual, like, plan is. But when you think about personal super intelligence, that is, you know, AI that can do things for you, not just give you information.
Speaker 1:I just wonder how much will happen outside of the, like, the meta ecosystem. Like, they've they've launched a search engine before that looked at websites outside of Facebook. They they had that Project Titan, which was to unify all the different messaging protocols. So as part of that, they gave everyone a facebook.com email address or something like that. Maybe it wasn't facebook.com, maybe it was like fb.me or something.
Speaker 1:But they gave every Facebook user whatever their unique username was, they gave them that as an email. And you could email that and it would show up in Facebook Messenger. And then they tried to unify Facebook Messenger so you could see Instagram DMs, WhatsApp messages and Facebook messages all in one place. Can see what you're laughing at.
Speaker 2:Dave's Dave Oh. Ram Powell says, yeah, want more of those amazing meta AI features.
Speaker 1:You say that now. I mean, let them cook at least a little bit because we really haven't seen them launch a new model, a new image model. Like, they should be able to get too close to the frontier. You know, it has to be at least Sora Nano Banana v o three level. They have all the data.
Speaker 1:They have all the talent now. They they're very GPU rich. They have the compute for it. And the research has been done, people have reverse engineered it. So you would think that whatever's coming should be good.
Speaker 1:Ryan asks, is Tyler still drunk?
Speaker 2:Yeah. One sip. He's just over there slurring TikTok his is dead. The algorithm is worse than the reels that make it to Facebook.
Speaker 1:Wow. I haven't Yeah.
Speaker 2:So so bay basically, they're they're transitioning Mhmm. Everything over. Yeah. Yesterday, there was apparently
Speaker 1:An outage.
Speaker 2:Kind of an outage. Like Yeah. People were able to post videos but the videos wouldn't be served Yeah. At all. Mhmm.
Speaker 2:So I think a lot of people assume that it was like the new ownership kind of censoring. Sure. It I believe they had an outage at a data center that was a cause of that. So before we before we call it dead Yep. Let's wait for a few days and and see how it pans out.
Speaker 2:We do have a TikTok account. It's at TBPN.
Speaker 1:How many followers does that
Speaker 2:have? Haven't checked.
Speaker 1:Don't think checked we post on it ever.
Speaker 2:3,500.
Speaker 1:Not bad. That's better than when I thought.
Speaker 2:First time. First time.
Speaker 1:We're not really focused on it. Maybe we'll test it out. But I'm I'm pretty happy with just I I want the things we do to be polished. I want the core show to be polished. Die at TBPN, our twenty to thirty minute cut down.
Speaker 1:I want that to be polished. I want the newsletter to be polished at tbpn.com. Before we bite off another Platform.
Speaker 2:Part of the apple. Another Please. Please, sir. Not not one more short form.
Speaker 1:Not one more short form. The Super Bowl. Super Bowl is coming up. I think it's gonna be this year, in the next couple months. When is it?
Speaker 1:It's coming up. It's coming up because there's advertisements that are going out, and you got to watch it because
Speaker 4:the I ads are going to be
Speaker 2:actually just had to search when is Super Bowl.
Speaker 1:Eric Lyman, CEO of Ramp shared, meet Brian. Brian's been carrying accounting on his back for a long time. Super Bowl Sunday, he finally gets back up.
Speaker 2:Yeah. We're real sports guys.
Speaker 1:This is the Super Bowl.
Speaker 2:Gotta goo we gotta Google.
Speaker 1:The Ramp Super Bowl ad is the Super Bowl of It Super Bowl
Speaker 2:feels like it feels like, you know, they just kind of It is. This is like, you know, what what is it? What is it called? Like a warning, spoiler alert.
Speaker 1:Finance meeting in five. Minutes?
Speaker 3:Ramp? I got it. Allow me. Hi, handsome. We're we're saving so much time.
Speaker 3:We're We're here. Travel, meals, hotels.
Speaker 1:How's this quick? Beautiful. Yeah. No. I I I think this achieves a couple things.
Speaker 1:I think I mean, it drills the brand name in. Think think about how many Ramp logos are in there. And they're chanting Ramp. And for you know, Ramp's a very successful company. We all know about it here.
Speaker 1:But there's a lot of people that just don't know the name Rampus. It hasn't been drilled in them like some company that's been around fifty years. It
Speaker 2:just sure. Takes the precipice of being there's a not been enough mainstream marketing yet for it
Speaker 1:to be
Speaker 2:a household name.
Speaker 1:It's a life's work to actually drill into people's mind the ramp name, the logo, the color, how it sounds when you say it, what it's synonymous with. And so just like not going too abstract, not trying to tell some more avant garde story here, I think is it's almost like a direct response at. It's just so clear what the problem solution brand. Problem solution brand. It's like simple.
Speaker 1:Like, you could be doing something more bold, more crazy. Yeah. But I think this is what you need to do at
Speaker 2:this last year getting Saquon
Speaker 1:Yeah. And then That was really great actually.
Speaker 2:Yeah. That that's actually kind of like an impossible set circumstances. Not impossible Yeah. But it's a crazy roll of the dice. Toby over at Shopify posted I heart love rate through his first stint at the Daytona.
Speaker 1:You see the first annotation on here?
Speaker 2:This is It's just crash.
Speaker 1:Crash. So he started out at like one twenty. One twenty beats per minute. Right? He's doing waiting.
Speaker 1:He does the the warm up, preparing, the form formation lap, and then there's a crash right right at the start. Very, very rough.
Speaker 2:So basically, the the crash happened with the l m p two class
Speaker 1:Yes.
Speaker 2:Which is the kind of pro am same segment that George from CrowdStrike is racing in. Yeah. So both Toby and George Yeah. All of our boys got hit right right as the race opened. And so what George was saying yesterday, the reason he was frustrated, he's like, this is a twenty four hour race.
Speaker 2:Yeah. Never has a race like this been won on the first lap. Yeah. It's incredibly unforced Yeah. To like crash in the opening Sure.
Speaker 2:Corner Sure. When you really should just get through it. It's like the most one of the most intense moments because there's so much traffic. But if you actually go and watch the footage of what happened at the opening, somebody gets hit, spun out and then they're turning around and somebody hits them again. Like two accidents in the opening That's crazy.
Speaker 2:A minute. So insane.
Speaker 1:I I love that the actual true final heart rates spike was at the end when you're changing out of the car. Yeah. You've been driving and so intense.
Speaker 2:No. You know this when you're getting when you're getting out of a track car, it's like watching a guy who's six'eight
Speaker 1:We would never share that footage. It is extremely embarrassing.
Speaker 2:He's always like crawling out of It's actually
Speaker 4:have to get on
Speaker 2:your You basically have to get on all fours.
Speaker 1:It's incredibly negative aura and I don't appreciate you sharing it on the show.
Speaker 2:I think it's cool.
Speaker 1:It's funny. But yeah, I I go full sun basically fly out of the car.
Speaker 2:Gibberoni on x says, Zoom is the best anthropic play. Yes. Zoom likely made a $51,000,000 investment in anthropic Series C How? In 2023 How? At a 4,100,000,000.0 Yeah.
Speaker 2:Valuation. If you're looking at their new $3.50, there's
Speaker 1:Okay.
Speaker 2:There's something like a 85 x. Even diluted, Zoom may have a multi billion dollar Yeah. Dropping position. Tyler, major I won't I won't give you too much flack for it, but obviously, the the most bullish
Speaker 1:Yeah.
Speaker 4:In the room. Yeah. Bullish on AI broadly.
Speaker 1:Yes. Yes. Yes.
Speaker 2:You you seem like you'd be happy to own Zoom at a $100,000,000,000 valuation even if they had no business at all. They just held us a holdco.
Speaker 4:Holdco. Yeah. Even if they just
Speaker 1:had It's digital asset treasury, actually.
Speaker 4:Yeah. So I think
Speaker 2:the story
Speaker 4:is this is like a rumor. But basically, it was that Anthropic like, wanted to just, like, use Zoom and get, like, the enterprise
Speaker 1:Okay.
Speaker 4:Plan or whatever. But then they were, No. Well, yeah, you can have it, but, like, we wanna throw in a little something. Right?
Speaker 1:This might be
Speaker 4:fake news. Is very like fake news, Tyler. Okay. Yeah.
Speaker 2:Because Zoom sells you know Zoom sells enterprise software. Right?
Speaker 4:Look, that's what I read. I can't find the post, but I I for sure read
Speaker 2:They sell enterprise software. And so Anthropic says, hey, we're big Zoom fans here. Yeah. We wanna use Zoom. And they're like, no.
Speaker 2:Yeah. Actually for you, no.
Speaker 1:After the the the crazy COVID pump where everyone got on Zoom, they started adding like crazy features like dictation and workspaces and white boarding and stuff. But obviously, like the the it was like so overheated that it it came back down to earth. But now they have a entropic position on the balance sheet, which will be fun for them. Slowly and then all at once says Blake Robbins. He says, your work tools are now active in Claude.
Speaker 1:Draft Slack messages. Interactive. Or interactive. Draft Slack messages, visualize ideas in Figma, and build, and see Asana timelines. All of the different tools are coming together in one place.
Speaker 1:When you see an account like Claude posting about Slack and Figma and Asana, you have to imagine there's a discussion there. It's not open source, so they're chipping away at these. And OpenAI has been chipping away at these for a long time, so the race is on to have the most integrations at this point.
Speaker 2:Breaking Anthropics' warning to the world. Anthropics CEO Dario Amade says, eminent real danger that superhuman intelligence will cause civilization level damage absent smart, speedy intervention. Sarah says, so buy our products.
Speaker 1:This is the problem of dropping like a 20,000 word essay is that like you're gonna get clipped out of context.
Speaker 4:Tyler, can you make
Speaker 2:a You should make a version of the new letter with subway surfers that run.
Speaker 1:Oh, true. Just
Speaker 2:build that as a standalone.
Speaker 1:Yeah. We we we needed Clad Labs, Chad IDE for for reading
Speaker 4:Well, summarize this in four words.
Speaker 2:AI good. Bleeding bits has some thoughts on Dario's
Speaker 1:14 of them, specifically. That's a lot of thoughts. There's nothing new here, if you're familiar with the AI safety discussions that have been happening on Twitter. Yes, but it's important for Daria to restate them in a format that can be passed around and formatted and is coherent from start to finish. The most interesting bit is that his mental model for AI control risk is the risk that would be posed by a country of geniuses in a data center.
Speaker 1:Interesting. That is interesting. The basic idea is that we should imagine a giant data center, all the models being something between AGI and ASI trying to coordinate to take over the world or do massive harm. Anyway, I think how seriously you take short term AI control risk is inversely correlated to how much you think about AI control risk as operating in a system. So the systematic view starts and says labs exist in an ecosystem where they need to sell models that will follow human instruction or they have no market.
Speaker 1:They are also overseen by regulators and guided by public perception. And the desires of their employees and all all of this keeps models corrigible. Great word. And the model landscape will look like three to six frontier labs running millions or billions of rollouts at a time on two to three different models, all on different tasks. So a model takeover requires these millions or billions of rollouts to somehow end up all be coordinating towards some bad aim that somehow the models have autonomously determined.
Speaker 2:Something I've been thinking about is this kind of summary, and a lot of the dialogue is centered around just like what are the models doing, or like a country of geniuses in a data center. But you have to be thinking about this in the context that a country of geniuses in a data center would just recruit millions of humans to join their cause. Some people, when they're thinking about AI risk, it's like, dude, just turn the computer off.
Speaker 1:Yeah, yeah,
Speaker 2:yeah. Just unplug it.
Speaker 3:But it's like,
Speaker 1:what if you're on the
Speaker 2:side of computer? What if there's 100,000 people or 10,000 people defending it on top of the data center? Don't unplug the computer. Don't unplug the computer, right?
Speaker 1:I like the
Speaker 2:computer. And so when you look at all the chaos in the last week, there's been so many moments where a certain image was AI generated. And then it's like, oh, that wasn't a real image at all. And it's being shared from all sides. And so at what point you know, you could have, you know, nefarious hostile AI Yeah.
Speaker 2:That's entire job is just creating Chaos. Millions of bot accounts that are just like sharing whatever narrative is self serving. Yeah. So You can't tell.
Speaker 4:The the two kind of scenarios where where where Dario's is about basically, even if we have like pretty safe models, which like he thinks we can do with interpretability or whatever, if it gets into the wrong hands, it's like very bad. If it gets into, you know, autocracy Mhmm. That that's one of the the main risks. Where aliases and and a lot of the safety ones are always these like very sci fi narratives where you have this like gray goo Mhmm. You have these like nano machines that somehow one day they just like kind of, you know, flip and then it's just kind of over.
Speaker 4:Yeah. And I I think it's much more reasonable
Speaker 3:Yeah.
Speaker 4:And like Nuanced. Yeah. It it's much more legible to like
Speaker 1:It's trackable too.
Speaker 4:Yeah. And especially I I think a lot of it is I don't even know if it's really subtext, he's definitely pointing in the direction of like, we need some government oversight. Mhmm. We need policy. Mhmm.
Speaker 4:And it seems like like you can very easily like track like his ideas on what policy should be from this essay. Right? It's a lot about China, a lot about Yeah. You know, making sure that individual companies don't like become, you know, as big as as governments.
Speaker 1:One interesting wrinkle with this. He did not post it as an X article. He posted it as a link because he wants to signal to everyone, like, look, I don't need the million dollars.
Speaker 3:I don't
Speaker 2:need the million.
Speaker 3:I don't
Speaker 1:need the million. I know I got a banger on my hands, 3,500,000 views, 11 k likes, obviously lots of discussion all over.
Speaker 4:But as an investor, like, I wanna see my my lab CEO be like super hungry for compute. Right? Yeah. So I want them to always be grinding to get, like, extra
Speaker 1:Oh, so maybe this is bearish. They should have posted
Speaker 4:on I'm an anthropic company Yeah.
Speaker 2:Why did
Speaker 4:he not post this on x?
Speaker 1:Yeah. That million dollars could have
Speaker 4:gone straight into git, like, git More
Speaker 2:yeah. It would have for sure
Speaker 4:given his article. Yeah. I
Speaker 1:can definitely see
Speaker 3:that happening.
Speaker 1:Thank you for watching.
Speaker 2:Thank you for watching.
Speaker 1:Leave us five stars on Apple Podcasts and Spotify. Sign up for the CVPN newsletter at t v p n dot com, and have a great rest of your day.
Speaker 2:We love you. Goodbye.