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 watching TVPN.
Speaker 2:Today is Tuesday, 06/24/2025.
Speaker 1:Keep it together, Joe.
Speaker 2:We are live from the TVPN UltraDome, the temple of technology.
Speaker 1:The fortress of finance.
Speaker 2:The capital of capital.
Speaker 1:The capital. Front page of
Speaker 2:the Wall Street Journal has the update from the Iran conflict with The United States. The Iran fired missiles at a US base in Qatar. Since this since this published, there is a ceasefire, and Donald Trump appears to be very upset with both sides and wants peace and wants, the rocket attacks to stop. And so we certainly hope that, the conflict will come to a close and we can get back to real estate development.
Speaker 1:And golf.
Speaker 2:Golf. Exactly. Oil drops. Stocks rise after attack. So, it seems like this Yeah.
Speaker 1:This is what The market is growth capital bloke was talking about.
Speaker 2:This is
Speaker 1:the start of World War three is a positive catalyst because it implies there will be an eventual end to World War It's
Speaker 2:a new catalyst and that's exactly what happened. So World War three is over. Hopefully, it stays that way. Anyway, we have some we have some new drinks in the studio. We have new tonic from none other than Chris Williamson.
Speaker 2:And he was very gracious to send us a full case, and I will be doing a taste test. Now, you know, it's following How
Speaker 1:much caffeine is there?
Speaker 2:A hundred and twenty milligrams, I believe. Yeah.
Speaker 1:We should have Tyler drink 10 today.
Speaker 2:Yes. Should. Yeah. As if you've been following the show, you know that I am have been a fan of Matayina. Andrew Huberman's Yerba Mate.
Speaker 2:That's This is the mango key lime flavor. It's non carbonated energy brew. And here I have the Chris Williamson Nutonic productivity drink. It says fuel your focus. I'm gonna do a little taste test and let you know.
Speaker 2:I'm gonna try and be as objective as possible. Okay. So this is the Andrew Huberman Yerba Mate.
Speaker 1:That
Speaker 2:is incredible. It tastes like the work of a TBPN guest. Like someone who's been on the show clearly worked on this. Let's Now let's taste Chris Williamson. I I I did invite him on the show.
Speaker 2:Oh. That tastes like someone who who hasn't been on the show yet. Interesting. Yeah.
Speaker 1:It's funny how that works.
Speaker 2:It is funny how that works. So, I mean, maybe Chris should come on the show and
Speaker 1:Yeah. We could correct that.
Speaker 2:Yeah. And then and then I think I could maybe give it a more positive review.
Speaker 1:You're onto something, John.
Speaker 2:No. But it tastes fantastic, and I will be drinking this throughout the show.
Speaker 1:And I'm a fan.
Speaker 2:This is actually the very first time that I've ever tasted this. It's very good. Very good. Anyway, looking forward to having you on the show, Chris. Hop on anytime.
Speaker 2:You're welcome.
Speaker 1:So are we newsmaxing today?
Speaker 2:We're newsmaxing. We're running through a few top stories, the current things. There's the anthropic ruling. Right. We gotta talk about All In Tequila.
Speaker 2:We touched on it yesterday, but there's a lot more news.
Speaker 1:There's a lot more to go into.
Speaker 2:We're putting on our investigative journalism hat. Yep. We're going deep. We're going deep. Also, Hims and Hers is in the news.
Speaker 2:There's a messy breakup that has jolted Hims and Hers. Yep. But the retail investors are bullish. They're continuing to buy.
Speaker 1:So There's
Speaker 2:a lot of news back and forth.
Speaker 1:HIMPS. And
Speaker 2:then we gotta talk
Speaker 1:about HIMPS ended last week at $64 a share. It started yesterday morning Mhmm. At $43 a share. Mhmm. So Mhmm.
Speaker 1:The Novo Nordisk debacle impacted the share price pretty dramatically. Mhmm. We're gonna have Kareem Yu on the show later. Sorry for that. A GLP one enthusiast to break that down with us, and we'll talk through it as well.
Speaker 2:Huge enthusiast. And then we'll also take you through the story of the rare earth magnets, which is complete fake news. Rare earths are not rare. They're just the the industrial capacity to bring them out of the ground is rare. And and the industrial might is rare, and it has mostly moved to China.
Speaker 2:And so now Ford Motor Company, the Ford of Motor Companies, has declared that they have a magnet shortage despite the China deal. And so we'll dig into that. Anyway, let's kick it off with David Sachs. We're not talking about tequila yet. We're talking about the positive ruling for AI.
Speaker 2:This is because Andrew Curran has the story. A federal judge has ruled that Anthropic's use of books to train Claude falls under fair use and is legal under US copyright law. And I couldn't agree more with this. I had been saying this for a long time that I believe that the results of of LLMs are typically transformative, and very few people are actually using LLMs and and AI models as to be perfect substrates. Yes.
Speaker 2:Exactly. You would never go to claw if
Speaker 1:work is not being reproduced.
Speaker 2:Exactly. Right? It's being,
Speaker 1:you know, remixed. Yes. And
Speaker 2:might say yeah. You might say, give me the high tell me about the themes in Harry Potter. But you're not saying, Claude, give me
Speaker 1:the chapter one. Available to some degree on blogs since the Internet existed. You could look up. What are some of the key points?
Speaker 2:Website, SparkNotes. Like, SparkNotes was a was a version of that that was transformative. And so I I I always thought that this would go this way, but I did think that we needed to sort it out in court. And for some of these, categories of media, there probably needed to be deals. And I still think that there's probably room for something like Simon Schuster or Penguin Random House to go and do a deal with Anthropic or go do a deal with OpenAI to say, hey.
Speaker 2:We'll give you even more access. Hey. We'll let you we'll let you reproduce the whole book if you want in certain cases as long as you're paying us appropriately. And maybe that's just a few cents for every query, but it'll add up and we'll actually make just as much money as
Speaker 1:we did before. So Interesting thing is that, the judge did find that Anthropic infringed on copyrights by storing over 7,000,000 pirated books in a central repository. Interesting. So they had the books. And there's gonna be a trial in December that will determine damages.
Speaker 1:Statutory damages could reach up to a $150,000 per book.
Speaker 2:That's a lot. Wait. Multiply that out. Billions.
Speaker 1:So anyways, if if if the judge, or the jury were to award that,
Speaker 2:they would feel like a book cost, like, $20. How much does a book cost?
Speaker 1:Well, they're saying the damages could be up to
Speaker 2:It's ridiculous.
Speaker 1:50,000 per work. I doubt that goes through it would end would end anthropically. Nope. But there's just no way.
Speaker 2:Yeah. That sounds like the the price of a book after David Senner does does a episode about it. Yeah. And there's only six copies on Amazon. They've been spiked up.
Speaker 1:Anyway Yeah. And this is the first major ruling applying the fair use doctrine to generative AI training Yep. Which has been a key legal issue that every lab is following. And, yeah, ultimately The big The big court is criticizing Anthropic for pirating these texts. But
Speaker 2:The big question here is, will training on YouTube data be fair use? Because it is public. Like, my YouTube videos, I've host them on Google servers, on a Google service, but you can you can go and watch my video and summarize it in a blog post that you post on WordPress, and you can transform it. And I could say, hey, like, I'd prefer if you didn't do that, but it is it is fair use and I don't have a problem with that. Yeah.
Speaker 2:Now, the question is, can OpenAI go and train, what's their video model again? Not Dolly. Sora. Sora. Sora v three on all of YouTube because it's fair use.
Speaker 1:Yeah. If this is like And the book thing is interesting too because in theory, Anthropica could have gone to many different public library libraries and not had to actually pay for the book and could have accessed them in different ways. But ultimately, we'll see how it shakes out.
Speaker 2:But this is a big mean, the thing for Anthropic obviously is to get on ramp. Time is money. Save both. Easy use corporate cards, bill payments, accounting, and a whole lot more all in one place. Go to ramp.com.
Speaker 2:Anthropic. Dario, if you're listening, I know you're listening. Go to ramp.com. Sign up if you haven't already. David Sachs summed it up.
Speaker 2:Positive ruling for AI, there must be fair use there must be a fair use concept for training data or models would be crippled. China is going to train on all the data regardless. So without fair use, The US would lose the AI race. This is an interesting take. I hadn't even thought about that dynamic, but that makes a ton of sense.
Speaker 2:Yeah. And then you have this weird dynamic where it's like, what if deep sea gets trained on all the American books? And so deep sea becomes like hyper capitalist. And then the only way for American models to to up is like perplexities like deep sea seven seventeen seventy six. So they're like fine tuning on deep sea output that's In order to deliver
Speaker 1:Western ideals.
Speaker 2:I love it. Maybe that's the maybe that's where this winds
Speaker 1:up to action.
Speaker 2:Yeah. It's great. The writers filed the class action against Thropic last year arguing the company, which is backed by Amazon, Alphabet, used pirated versions of their books without permission to compensate to teach Claude to respond to human prompts. I wonder what what pirated means in this context. Because I wonder if it literally means, like, they went to, like, a torrent website and, like, downloaded it illegally or if it means, like because you can, like, you can technically, like, go and buy a DVD and then rip it and then have the m p four file.
Speaker 1:Version.
Speaker 2:It's not, but it's not pirating because you own the you own the DVD. Yeah. And so that gives you the right to put that that file on your computer.
Speaker 1:Yeah.
Speaker 2:And it would look like a pirated file because usually you can't
Speaker 1:just have a Yeah. $7,000,000, $20.20 dollars a pop. They didn't spend a $140,000,000 on books. Just try I
Speaker 2:They should. They should
Speaker 1:have supported the writers. That that's one of the risks going to this trial. It's possible that, you know, I
Speaker 2:don't They should they should they should have to pay $20 a book.
Speaker 1:They should spend a $150,000
Speaker 2:book. Is the actual damage again? You had the number?
Speaker 1:A $150,000 per
Speaker 2:That is way too high.
Speaker 1:It could go up to
Speaker 2:the point. Would give them some bulk pricing maybe. $17 a book, $15 a book, maybe even $10 a book. But they gotta pay something.
Speaker 1:The issue is that the individual rights holders are not exactly gonna be super happy about getting one purchase.
Speaker 2:Hey. 20 is $20.
Speaker 1:$20 is $20.
Speaker 2:You see at the back of every book. On the back of the book, it says recommended price. Like, I wanna buy one. Yes. Who knows what I'm
Speaker 1:gonna is $20. Some people are gonna read it.
Speaker 2:Some people are gonna let it sit on the shelf, and some people are gonna train a massive AI model.
Speaker 1:Could go through this whole lawsuit and come away and buy one double steak, double guac, chipotle, burrito bowl Yeah. And think, oh, this is worth it.
Speaker 2:Yeah. Exactly.
Speaker 1:It's worth it.
Speaker 2:I mean, fundamentally, Anthropic believes that they are creating God. So, like, you want an illiterate God? Like, you gotta let God read some bugs. It would be an affront to God to not let God read bugs.
Speaker 1:Might be
Speaker 2:on to something. Anyway, if you're designing the next great AI product, go to figma.com. Think bigger, build faster. Figma helps design and development teams build great products together. Now, let's move over to the all in tequila.
Speaker 2:So we talked about it briefly, but we wanna go a lot deeper today. So Chamath Paliapatia has the story. He says, we launched our ultra premium tequila today in Los Angeles at Delilah's. I've never been to Delilah's. Is it good?
Speaker 2:Have you been? Are you familiar with this place?
Speaker 1:I'm not big into that world, John.
Speaker 2:Me neither.
Speaker 1:We mostly don't leave the house after nine Where
Speaker 2:does ultra premium sit in the ranking? Because it's below luxury.
Speaker 1:Yeah. I was surprised with with the pricing is clearly luxury.
Speaker 2:That's what I would think.
Speaker 1:The position the positioning overall, the bottle looks incredible. The box looks amazing.
Speaker 2:The bottle actually
Speaker 1:does look incredible. It very clearly it it is an order of magnitude more expensive than traditional Yeah. Luxury or I would definitely say ultra premium products. Right? Yeah.
Speaker 1:So, yeah, squarely sitting in that luxury category. But I think he just I think he just just ripped it.
Speaker 2:I think I think there it begs the question, like, what comes after ultra premium? And it's obviously giga premium.
Speaker 1:Yeah. So they wanted to leave room for a $10,000
Speaker 2:thousand dollar bottle that's giga premium.
Speaker 1:You gotta So for those that don't know, the bottle the the tequila is $1,200. $1,200. 750 milliliters. Yes. And And let me read a description.
Speaker 1:Yes. Pot distilled from a 100% blue weber, agave grown in the lowlands of Jalisco and aged in white oak, x whiskey barrels for five years. This extra anejo tequila has the perfect balance. A smooth, slightly sweet taste with hints of wood, citrus, dried fruits, butterscotch, vanilla, cacao, toffee, and cooked agave. We recommend it to be paired with good friends and lively debate as they do on the show.
Speaker 1:Fitting. The bottle is a handcrafted art piece evoking a stack of poker chips. It is illuminated from the base to appreciate the beauty and complexity of its design. Each is individually numbered for its very limited release of 7,500 bottles. So
Speaker 2:I would really love a new
Speaker 1:If they sell out, they will gross 9,000,000.
Speaker 2:Wouldn't you love a $1,200 energy drink? Yeah. Just a beautiful energy drink.
Speaker 1:Ultra premium.
Speaker 2:Ultra premium energy drink. That comes in a glass bottle, pour it out. Oh, they'd be so good. They really should do that. I'm I'm sick of the energy drinks in the, oh, three dollars a can, $5 a can.
Speaker 2:Let's get let's get these numbers up. $500.
Speaker 1:Let's get these numbers up.
Speaker 2:Let's do it. Anyway, there was a lot of pushback. Augustus Dorico chimed in to the tune of 5,000 likes. He says, the elites have totally abandoned us and and their responsibility to steward the nation. We're on the brink of war.
Speaker 2:Deaths of despair are at an all time high. The economy is fake, and the plutocracy is launching ultra premium tequila brands. We need greatness. And so he was very upset. A nonequity partner had the flip side, said, you can just do multiple things, lead transformative agendas, and launch liquor brands.
Speaker 2:Look at Tesla. Elon Musk launched the Tesla tequila. It was kind of a fun side project. And I think that's what Emily Sundberg was getting at before when she was talking about how celebrity brands people kind of know the the the thing. Like, do we really think Dave Friedberg is gonna take his foot off the gas with his company that he's actively running to Yeah.
Speaker 2:Work on this tequila brand? Like, no. He's he's he, you know, signed off on it, went to the party. It's gonna live on a website. We I actually found the the design firm that did the design.
Speaker 2:Like, this is not going to suck these folks away from from the other things. At the same time, you know, Augustus does have some point that, like, maybe it's a little unserious for them to do it. I don't know.
Speaker 1:But I don't know why it's unserious. They have their friends. They have a podcast. They enjoy drinking together Yeah. Playing poker.
Speaker 1:They introduced a poker themed bottle of tequila.
Speaker 2:Yeah. I also and
Speaker 1:enjoy with them.
Speaker 2:Yeah. I
Speaker 1:also think it's I don't think it's unserious. Yeah. Yeah. I think I have more questions around how seriously are they taking it as a business.
Speaker 2:Sure. Sure.
Speaker 1:Sure. Sure.
Speaker 2:Bringing it that.
Speaker 1:CEO? Are they trying to, you know, scale this into
Speaker 2:Yeah. Is it a drop?
Speaker 1:Because it might it might not even need a it might not even have
Speaker 2:its own c corp or, like, its own LLC. It might just be, like, merch essentially, and it's cool as merch. And and then there is this question where, you know, I I feel like there is a drumbeat in the news and in the culture around, like, things have never been worse. Like, you and, like, it it like, it's so bad right now. We're in a uniquely bad situation.
Speaker 1:Yet at the same time, we have
Speaker 2:peace in
Speaker 1:The Middle East today.
Speaker 2:Yes. Yes. Peace in The Middle East, finally.
Speaker 1:Finally. For the first
Speaker 2:time in '60 Only
Speaker 1:a few days after the launch of this, and there could very well be some Correlation. There's definitely a correlation.
Speaker 2:Definitely a correlation. Whether
Speaker 1:or not there's causation, we don't know yet.
Speaker 2:I think you get a couple of
Speaker 1:these bottles. But Trump had his own vodka back in, so he introduced this in 02/2005, the Trump vodka
Speaker 2:Mhmm.
Speaker 1:Which is funny because I don't even, he doesn't drink alcohol.
Speaker 2:Right? Yeah. He doesn't.
Speaker 1:He's never been big into that whole world. Wow. And so I think this was just
Speaker 2:Yeah. How are you gonna criticize the I mean, the all in guys seem to enjoy the tequila at least.
Speaker 1:Trump Vodka was an American brand of vodka produced at first in The Netherlands, then later in Germany by Drinks America. Drinks America.
Speaker 2:It was made in Germany by a company named Drink America?
Speaker 1:Produced in The Netherlands and then Germany. It was never Drinks Americas. It's the company is called Drinks Americas. Two s's.
Speaker 2:Was never produced
Speaker 1:in America. The brand was launched in The United States in 2005 but ceased production under the Trump name in 2011 when it failed to meet the required threshold for distribution. Honestly, might be the time to bring it back.
Speaker 2:I don't know if they're gonna do distribution. It seemed like it was all a
Speaker 1:long However, it is still sold in 2016. This was the last time it was updated especially around the Jewish holiday of Passover. Interesting.
Speaker 2:Odd.
Speaker 1:Trump at the time, the the brand slogan was success distilled with Trump predicting it would outsell Greg Goosebumps. Also said that when mixed with tonic, which he referred to as a Trump and tonic, TNT, it would become the most
Speaker 2:drunk tea is good.
Speaker 1:It would become the most drunk cocktail in The United States. I love I've I I I mean, the consistency of his confidence across eras is
Speaker 2:It's incredible.
Speaker 1:In 02/2007, Drinks America signed a deal to export 50,000 cases of Trump Vodka annually to Russia.
Speaker 2:To Russia. Oh, no.
Speaker 1:But in 2011, it was discontinued due to sales failing to meet the company threshold requirements. Several reasons were attributed to this. One was because Trump himself is a non drinker and never drank Trump vodka.
Speaker 2:Never seen tasted it.
Speaker 1:Yeah. Wow. Drinks America's also had problems producing it because the glass used in the bottle and the gold leaf labels were expensive and the company could not afford to produce them in large numbers. Yeah. In 2015, in his campaign for president, Trump did not include Trump vodka amongst assets submitted to the federal election committee.
Speaker 2:So a zero. Yeah. Interesting. Well, my advice for the besties, you're launching a tequila brand. You're gonna have to pay sales tax.
Speaker 2:You gotta get hu.com. Sales tax on autopilot. Spend less than five minutes per month on sales tax compliance. I liked Mike Solano's take. He said, I'm happy for the all in guys and tired of the tedious hate.
Speaker 2:They built a popular show, which is hard. True. Now they're launching a tequila brand. Okay. Great.
Speaker 2:Hope it does well. Jealousy on this app is unreal. Their success can't hurt you. Focus on yourself. I like this.
Speaker 2:Eric Thornburg chimes in. How you feel about their tequila brand is how you feel about yourself. That's been a line he's been repeating. It's it is it is interesting. It seems like, you know, their their whole show started as, like, they're locked down during COVID.
Speaker 2:They all hop on on Zoom, and they have some fun talking to each other. And they love poker, and this feels on brand. This feels like a good line extension for the brand. Guess there's a question of, like, should they just run ads for a big tequila brand? But we should get into
Speaker 1:I think it's way more fun for them.
Speaker 2:Yeah. It's fun to have their own thing. Right?
Speaker 1:There's nothing more fun than creating something, especially if you've been in tech, creating something that's physical, that you can hold, that you can put on a table, that's beautiful Yeah. That you can share, that's social.
Speaker 2:So I wanna get into, you know, expected value, expected market size, expected profits. Like, how successful do we think this will be? Cause there's a world where this becomes a major brand. Like, I mean, Prime is a big brand. Feastables is a big brand.
Speaker 2:Like, these guys could turn guys could turn this into a massive company. At the same time, it could be more of a drop, and they could sell a couple 100 bottles, couple thousand bottles, make a few million, move on to the next thing. I found the design firm that actually worked on this, the Besties All in Tequila. It's from a from a design team called Hello Stranger. They say, we'll fix your brand.
Speaker 2:We'll give you a new brand, a brand that wins all the awards and sells at a premium. We'll give you a strategy, a unique angle based on insight and research. We'll give you the sexiest packaging that people can't put down, and we'll give you all the collateral you need to make your brand flourish. On this site, you won't find headshots of designers looking important next to yucca plants because it's not about us. It's about the billions in sales in one of the toughest sectors in the whole of retail.
Speaker 2:We're stranger and stranger. We've been doing it for over thirty years, and we are the best. As one of our clients said, don't leave anything to chance. They started in the next nine four.
Speaker 1:I have here, I think Dayjob is the best. They did the design for David Protein bars. True. It's gonna do billions in its first few years. It's hard to argue with with those kind of results, but
Speaker 2:So they announced the Besties All In Tequila, which is an interesting name choice as well. It's not the All In Tequila. It's not Besties Tequila. It's the Besties All In tequila. But they say the Besties at the famous All In podcast have just released a $1,200 tequila, which unfortunately for all of us is already sold out.
Speaker 2:The guys are big poker players, hence the All In. And and the iconic bottle single mindedly owns the idea. And so the the love language here, you read some of this handcrafted art piece evoking a stack of poker chips. It's illuminated from the base to appreciate the beauty. Oh, so there's a light inside?
Speaker 2:Let's go. That's cool.
Speaker 1:There's a light inside.
Speaker 2:I mean, that's what it sounds like. It says it's illuminated from the base. So I think I think when you're spending $1,200 So that's how think it glows. I think it glows. Yeah.
Speaker 2:Incredible. It's tech product. Yep. It's Internet connected. It's got
Speaker 1:It's IoT device.
Speaker 2:It's an IoT device, potentially.
Speaker 1:For boozing.
Speaker 2:For boozing. Imagine, you could link this, it could measure how much is in there, have an app, orders you more automatically, subscription. Now we're talking
Speaker 1:This it goes from a 9, you know, $9,000,000, you know, first run into a $9,000,000 MRR business. Yes. Getting up into that I think
Speaker 2:we're getting that.
Speaker 1:9 figure
Speaker 2:revenue range
Speaker 1:pretty quickly.
Speaker 2:It is it is fascinating. So each individually numbered for the each bottle is individually numbered. They're releasing 7,500 bottles. So if you do the math, 7,000 times
Speaker 1:It's 9,000,000.
Speaker 2:9,000,000. That's that's that's how much they're trying to make. Each bottle arrives in a numbered collector's box bringing the Bestie's signatures. They're either signing them or it's printed on there.
Speaker 1:Don't know. I think it's gotta be printed on there. David Sachs does not have time to hand signs.
Speaker 3:7,000.
Speaker 2:How long would that take you? You do one every couple seconds. It's, you know, days work.
Speaker 1:Those signatures would get so
Speaker 2:sloppy, buddy. Mister Beast has done that. I've seen streams where he's been, like, signing every single teacher. It gets really messy. That's But they say wet your beak, go all in, and it is remarkable how much their brand has dominated, like, the catchphrases.
Speaker 2:Like, we've been we've been pretty deliberate about trying to build the the TBPN brand around fortress of finance, lever up, all these different catchphrases. They've done a really good job because as soon as I hear wet your beak, I think I'll I think I'll in. I think I'll
Speaker 1:in. Your beak.
Speaker 2:Anyway, there's been some timeline and turmoil because Casa Zar tequila has a similar bottle design.
Speaker 1:And Very similar.
Speaker 2:And so we can pull that up. The founder story over at Casa Azar lies in a husband and wife founder story, a tale of resilience, determination, and the courage to go, quote, all in. Yeah.
Speaker 1:And so if you search if you search All In tequila right now, you'll get a a post from 04/11/2025, that says this is on the spiritsbusiness.com that the tequila stack owned brand Casa Zar transcends mere novelty by merging cultural roots with the most important aspect, flavor and character. Max Ramirez and Viviana Sarriolta Ramirez, two ex Diageo executives and industry veterans, launched Casa Zar in November 2024. The husband and wife do team were driven by passion, determination, and courage to go quote unquote all in in pursuit of a dream to create a brand that would stand out in a crowded market, not only visually, but also gas gastronomically. Less than a year since its launch, Casa Zar has quickly gained momentum and garnered industry accolades. Most recently, the tequila and mezcal masters twenty twenty twenty five.
Speaker 1:Casa Zar Tequila won two gold and a master metal. Mhmm. The brands Hoven and Reposado bag gold and they're ultra premium. Well, ultra premium might just be like a category. We're kind of exposing how little we drink anything other than champagne.
Speaker 1:Maybe A 100% agave tequila categories earning a master medal Casa Casa Zar Tequila Anejo and press judges
Speaker 2:Here's the bottle.
Speaker 1:Love it. Zoom in on that. Very expressive nose with plenty of earthy roasted agave and peppery spice, creamy and syrupy on the palate with drying oak and peppery spice, balanced with sweet golden syrup and shard agave. Judge Matt Chambers, director and co founder of For Everyone described Casa Sare Anejo as so delicious. So anyways, I saw this and my immediate thought was that maybe All In had white labeled Yep.
Speaker 1:The bottle, but that doesn't seem to be the case. And digging in a little bit more, some of the issues here, this company's been in business for a while now. It's ex Diageo execs. They have patents, from what I could tell
Speaker 2:around It's more of like a trade dress thing. The patent would be like how you manufacture
Speaker 1:design patents. Around the bottle design, which is very distinct. Interesting. Very distinct.
Speaker 2:Yeah. So there could be some fallout, maybe a negotiation, maybe someone's gonna have to go all in to see put it all on the line to see who can own the poker chip stack design for tequila bottles. Yeah. Could
Speaker 1:be And this could get this could get spicy. Right? According to Casa Zar, the tequila is designed for those who live life, who live like a high stakes poker game. Fearless, driven, and willing to take risks.
Speaker 2:A lot of the same things, so there could be some confusion. The besties
Speaker 1:are taking a risk.
Speaker 2:Yes. And and I've I've run into founders before who have who have inadvertently used names that they thought were kind of generic, and then they wound up, oh, there was smaller player in a submarket that actually had national ambitions and was gonna sue and and and pushed to say, hey. You actually have to back off my trademark because I know that you're selling this type of sauce. It's Big Spoon. There was Big Spoon roasters and Big Spoon sauce makers, and they both had sauces that were spicy, and they had different kind of logos but similar names.
Speaker 2:And one was in the Southwest and the other one was in California or Southeast. And it but eventually, they both wanted to go national and so they collided and they had to kind of decide how to back off.
Speaker 1:When we saw this yesterday with IO, I IYO
Speaker 2:Oh, Same the brand Same
Speaker 1:OpenAI where OpenAI had to scrub all mention of I zero or, you know, IO Yep. From their website because of a trademark issue. This is seem seemingly a bit different because it's a design sort of patent or trademark. I I don't have exact specifics, but I expect this to get spicy and Well you're gonna have people
Speaker 2:Casa Azar is known for being sweet and spicy.
Speaker 1:That's right. That's right.
Speaker 2:I I do wonder where ultra premium comes in. I feel like it's anything over like a $100 or something.
Speaker 1:Probably. Why
Speaker 2:would you put a $200 bottle and a $1,200 bottle in the same category? Those feel like it's it's almost it's half an order of magnitude. It's like 500 or five x difference in price. It feels like they're not necessarily playing in the same category, but the design elements and the the fact that it is a tequila and they're clearly going after awards and costs are Both are premium.
Speaker 1:Both are heavily branded around poker. Yep. One of them launched in 2024. Yep. One of them is launching today.
Speaker 2:Oh, wait. Casa Zar is only 2024. Yeah. Oh, wow. It's like pretty new.
Speaker 1:It's new. Okay. They filed their initial trademarks in 2023 though, so it's been a little bit.
Speaker 2:Okay.
Speaker 1:So they've they've been in market for a while and Casazar's execs, ex Diageo. Right?
Speaker 2:Yeah. Yeah.
Speaker 1:They probably got some sharp elbows. I imagine they're gonna wanna pursue this to some degree. Just given people don't realize how I think we live in this tech bubble. Yep. I don't think people realize how big all in is.
Speaker 1:It's like way bigger than than just tech now.
Speaker 2:Huge. Huge.
Speaker 1:Huge. And so this, you know, I think, we'll see how this shakes out. But So there's there's there's another thing
Speaker 2:that's kinda confusing is that I was wondering if they're playing in different markets because Casa Zar is 40 percent ABV. That's 40% alcohol by volume. 80 proof, which is pretty standard for liquor. On the all in tequila, the Besties all in tequila website, it says it's 750 milliliters, same size, 40% ABV, $1,200, which would be the exact same stats. But when you zoom in on the box, I don't know if you can see this, it says it's 48% alcohol by volume.
Speaker 2:Can you see this, Jordan? 48% here.
Speaker 1:That might have just been an early renders.
Speaker 2:Yeah. Maybe I I wonder where they're landing because I feel like 48% is a much stronger choice for tequila. Like, you would be noticeable that you would have a stronger alcohol content. And it's also much harder to mask the flavor if you are dealing with an extra 8% of alcohol because it's just gonna be more alcoholic. It's gonna it's like the
Speaker 1:It's powerful stuff.
Speaker 2:Yeah. I mean, one of the secrets to those, like, spirits that are popular on college campuses is that they take the alcohol content down to something like 30% or 35% flavoring
Speaker 1:it. I don't know where this picture is from because on the website, it says 40%.
Speaker 2:I don't know either. I don't know where I got that picture.
Speaker 1:Well, we will have to get a bottle ourselves. We will.
Speaker 2:And we'll taste it. We'll immediately tell you. Yep. 42 percent.
Speaker 1:We wanted to have Tyler taste it but fortunately Under 21. He's not of age. Yeah. So we'll save a little bit for you Tyler for your birthday.
Speaker 2:Yeah.
Speaker 1:Can't wait.
Speaker 2:Well, if the besties are planning to take the besties all in tequila to the moon, they're gonna need to get on Vanta. They're gonna need to automate compliance, manage risk, prove trust continuously. Vantas Trust Management Platform takes the 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. You know, if you wind up turning this into the big IoT SaaS product that we've imagined it being Yeah. You're gonna be managing customer data.
Speaker 2:You're gonna have their health records potentially. You're gonna be pulling in all sorts of stuff to optimize their liquor consumption, and you're gonna need to be secure. So you're gonna need to get on Panta for sure. How big do you think this is? Where do you think this actually goes?
Speaker 2:Like, like, I think that's the more serious question, is
Speaker 1:It totally
Speaker 2:9,000,000. I think that they're gonna sell that. I think they're gonna sell all
Speaker 1:seven No no problem. The haters will be in disbelief?
Speaker 2:Yeah. There are so many fans. We we we know we we we're gonna buy a bottle, but we know we know folks who are friends and friends of friends who are gonna buy bottles. There's gonna be people that buy bottles because it's fun and they wanna support their the show that they like. There's gonna be folks that gift it.
Speaker 2:There's gonna be people that buy it because they're they're, you know, they wanna just like support the the thing that hasn't been charging them for five years. Like, there's a little bit
Speaker 1:of So people just wanna support the dictator, Shama.
Speaker 2:Yeah. Honestly, it's like it's like, there is another world where if all in was $20 a month, you would have paid a thousand dollars over the last four years, and they didn't charge you $20 a month. They didn't even run Free. And so what's your way to get them to say, hey. Yeah.
Speaker 2:You've delivered me a thousand dollars worth of value, and I have it because I'm successful. And, you know, this is a great way. And you get something that, you know, who knows what the margin is on this? Is it worth $1,200? Maybe.
Speaker 2:I don't know.
Speaker 1:It's hard
Speaker 2:to say, but it's like it's a fun thing to buy.
Speaker 1:Imagine you're going to your friend's house for dinner. What's funnier than bringing a bottle of the Bessie's all in tequila and be like, hey, I just picked this up on
Speaker 2:the way Yeah.
Speaker 1:It looked nice.
Speaker 2:It's like it's almost it's like a good, like, high end tech gag gift.
Speaker 1:It's like a luxury gag gift.
Speaker 2:But it's also probably really good because it's So so then you taste it and you're like, oh, this is actually really nice. Which bottle design do you like more? Because Casa Zar has this ceramic bottle that they hand paint. You said you liked that it had the stack of chips all the way to the top.
Speaker 1:Stacked top
Speaker 2:to bottom. Don't like that.
Speaker 1:John doesn't like it. But break it down. I think the ceramic look is nice. They look more like regular poker chips. That is true.
Speaker 1:I think it's cool.
Speaker 2:I think it's almost too close to poker chips. Like, I would be confused. I'd be like, is that just a pack of stack of poker chips? It's, like, hiding. It's it's hiding.
Speaker 2:Can we zoom in on the stack of poker chips there? Bring that over. Oh, no. Skill issue.
Speaker 1:First Zoom? First Zoom,
Speaker 2:bud? First
Speaker 1:time zooming? Yeah. You you know Anyways, it it it Dude, John Cazar is a beautiful bottle.
Speaker 2:Years ago.
Speaker 1:The Casa Zar, beautiful bottle. Yep. I like that it's fully stacked. Okay. Ultimately
Speaker 2:I think it just doesn't read as a bottle of liquor, whereas the Besties all in tequila does to me read as it. And then also the, the glass, the glow
Speaker 1:The glow
Speaker 2:is insane. Is it it is in a different tier. Ceramic is premium. I would only give the ultra premium label now to the the Besties All In tequila. I would call I would call they've reset what it means to be ultra premium.
Speaker 1:Look. They're just showing off Zooms now.
Speaker 2:There we go.
Speaker 1:They're flexing. All over the place.
Speaker 2:Yeah. Yeah. Mean, not to not talk trash about Casa Zar. It looks fantastic. I'm just saying that the Besties All Tequila, they innovated by by including the illumination, by including the glass, by including top.
Speaker 1:Be the I o IoT functionality that people have been asking for.
Speaker 2:We don't know. So I I mean, I I would give the besties all in tequila the edge here. Now, do you know the story of the iPhone? Are you familiar with what an iPhone is?
Speaker 1:Are you familiar with the name? Going with this. I know where you're going with this.
Speaker 2:Okay. So When Steve Jobs introduced the iPhone, he famously said on stage, and, yes, we're calling it the iPhone. And this was controversial because Cisco owned the rights, the trademark to the word, the iPhone. And so everyone was like, how is this gonna work? You just launched a product that is clearly derivative of this other company's name.
Speaker 2:They have a trademark. They're just gonna block you. But at some point, a deal was made. I don't know if it even became a lawsuit. It was probably just like they sat down and said, hey.
Speaker 2:Let's, you know, let's let's allow this to happen. We wanna run with this. We could take it to the courts, but that's a headache for both of us. Let's just net this out, and we'll buy some WebEx licenses.
Speaker 1:There's a article on appleinsider.com from Yeah. 02/2006. Cisco introduces the iPhone family of devices.
Speaker 2:Wow.
Speaker 1:And it is not nearly as good looking as the iPhone Yeah. Would ultimately be. But it and it's funny to see the iPhone, like, word effectively, the iPhone word mark placed on on a non Apple device.
Speaker 2:Well, if you're trying to build the next iPhone or software for the iPhone, get on 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. Now we discovered an interesting link between Cisco, the iPhone, Apple, OpenAI, Johnny Ive. And I think we cracked the code.
Speaker 2:We gotta break it down for everyone.
Speaker 1:Get the tinfoil hat.
Speaker 2:Get the tinfoil hat. I will I will send the I will send the image to the team so they can pull it up. So okay. So this is this is how it goes. So Apple, as we mentioned, they potentially got the name the iPhone from Cisco.
Speaker 2:Now who designed the iPhone at Apple? Johnny. Johnny Ive. That's right. Now where does Johnny Ive work now?
Speaker 2:OpenAI. OpenAI. They acquired Love from. They've got brought him over to OpenAI. Who's the chief product officer at OpenAI?
Speaker 2:Kevin Weil. Kevin Weil. And who's, what what boards does he sit on?
Speaker 1:Cisco.
Speaker 2:Cisco. Who I who licensed the iPhone name from Apple?
Speaker 1:Wow.
Speaker 2:It's a snake in his tail. An Ouroboros of Silicon Valley intrigue. We cracked the code here. It's all linked. Symbolism will be their downfall.
Speaker 1:The most the most completely nonsensical schizo diagram. Like, what what
Speaker 2:are you alleging?
Speaker 1:Yeah. Yeah. What are you're you're alleging that things happen.
Speaker 2:That interact with each other? Yeah. That that that If only it were so simple, Jordy. Yeah. If only it were so simple.
Speaker 2:I wish I had the my your mind. Yeah. The mind of a golden retriever to not see the obvious pattern here.
Speaker 1:Yeah. That The big conspiracy.
Speaker 2:The big yeah. The big conspiracy that we unveiled here on TBPN that they're all linked. It's all part of
Speaker 1:It's all part of the plan.
Speaker 2:Pulling the strings. Big Cisco, Apple, OpenAI, Johnny Ive, Kevin Wheel. It's all linked. It's all linked. We we blew it wide open.
Speaker 2:We should throw all those guys in our CRM. Adio. Customer relationship magic. Adio is the AI native CRM that builds scales and grows your company to the next level.
Speaker 1:The official CRM
Speaker 2:Coca So so final final word on the besties all in tequila. I think it's a no brainer that they sell 9,000,000 in this first run. The question is, do they do they do that every month? Do they do that every year? Do they move on to a different drop?
Speaker 2:Or does this become
Speaker 1:something that is stable? Think one of the challenges of the the what what it will come down, it'll come down to product quality. Sure. And then the number of people in the all in audience that care enough about tequila to spend $1,200 on a bottle of tequila, not in a sort of like funny gag setting, but as like, I'm gonna just drink this regularly. And I think it could be a lot.
Speaker 1:Like, I could see them selling 20,000 bottles of of this. You can imagine them drink one of them is drinking it on every show. Right? It's sort of this constant always on advertisement for it. I could imagine them they're they're seeing the besties all in tequila.
Speaker 1:I could imagine them going into other categories. Right? And there's a world where they're selling
Speaker 2:Besties all in malt liquor. 40 ounces.
Speaker 1:Forties. Would love to see I would love to see Jay Cal do do Edward 40 hands.
Speaker 2:Yeah.
Speaker 1:Edward 40 hands.
Speaker 2:That'd be ambitious.
Speaker 1:We gotta move on. Okay. I'm having too much
Speaker 2:time to do different case for you. Have a different bull case for you that's a little more serious. So I have been in consumer packaged world for a decade. I've built two brands in the space. Like, the product quality is super important.
Speaker 2:You do need an edge, and you do need a a value prop that can actually resonate with consumers, and you need ongoing marketing. That's a 100% true. But in fact, the biggest monopoly brand is an important monopoly for consumer goods, but there's almost a bigger monopoly with just distribution. Like, if you actually walk around the grocery store, you will see I talked to a founder who came up with he he his whole model was he literally would walk down the grocery store and look at not just the products, but who owned the products. And he would see that, wow.
Speaker 2:This entire aisle is owned by Unilever. They just dominate the entire aisle. There's there might be a ton of different brands, but everything is Unilever or everything is General Mills or Kraft or something like that. And so he he wound up building a company that does he he when he was looking around, he noticed that the popcorn section of the of the, you know, popcorn aisle was extremely uncompetitive because Orville Redenbacher is the only major brand, and they're not owned by Unilever. So they don't have much about that guy.
Speaker 2:Don't hear much. So we started a company that, was like a popcorn spray that you could spray a flavoring that you'd spray on the popcorn. It's purely additive, wasn't competing with anyone, and he grew that to, like, a couple million bucks. It was great. And I think that a lot of these products, they do just wind up winning on distribution.
Speaker 2:And what's interesting about this is that I am almost certain that in the all in audience, yes, there are consumers, but there are tons and tons of people who work in distribution, work in store ownership. Like, would it be possible for any of the besties to get the CEO of of, Bevmo on the phone? Like, absolutely in two seconds, they could do that. Could they get in could they get this into Walmart? Absolutely.
Speaker 2:Could they get this into, I don't know, CVS? Don't even know what the biggest Yeah. What the biggest channels are for high ultra premium liquor are. But if you just think, even like the independent store owners, they probably listen or are familiar. Right?
Speaker 2:And so just being able to take that call and accelerate your sales force and have them be like, yeah, sure. Send me one. I'll put it on the shelf. When we sell it, I'll just order another one from you since it's a thousand bucks. And so, pulling that distribution forward into the into the future, being more aggressive about that about getting into distribution, holding onto it, I think that's I think that's something that could be could be interesting.
Speaker 2:And then the fact that the bottle design there's a couple companies that do that have been particularly good with bottle design. I've talked to one consumer packaged goods founder who who basically said that their whole model was, like, go into a category and figure out how to make something, like, shiny and gold and, like, pop off the pop off the
Speaker 1:Yeah.
Speaker 2:The shelf. Because you look at the wall of all the products, they all look generic. If you can be the one that stands out, you're going to just have more people pull it off to sample it. And when I think about if I'm walking into a liquor store and I'm seeing a whole bunch of tequila, and one of them is literally glowing and looks like a and it has a light up display. This is something that a lot of consumer packaged goods founders negotiate for.
Speaker 2:They will say, hey. Like, Red Bull would come in and say, we will give you a fridge, a Red Bull fridge. You know those fridges? Or, like, we will come in and give you a light up display. We we we fight for this all the time with Lucy.
Speaker 2:Like, we we will say, oh, we'll send you a free in store display that you can put on your shelf, and they're putting it on the counter so that we don't have to fight for the shelf space that's there. And so imagine if they're like, this every bottle is a light up display. It will jump off the it'll jump off the back bar. And so it won't just sit there and collect dust. It will automatically draw the eye to it.
Speaker 2:I think that the light up thing could be enough of, like, a distribution advantage that it winds up being a popular product.
Speaker 1:Totally. I don't know. Totally. I agree. Think it'll come down to product quality, what percentage of the audience actually wants to spend $1,200 on tequila on a recurring basis.
Speaker 1:Do they go into other categories? I I think they could sell, you know, Chamath is also launching a wine club. Yeah. That was interesting. That's separate.
Speaker 2:There was so much backlash to the tequila thing. It wasn't even the first alcohol brand he launched that month.
Speaker 1:Yep. Yeah. At the end of the day, anybody that's getting mad about them doing something like this Mhmm. It's like, they have always been very candid that they enjoy drinking together. Yeah.
Speaker 1:Why, you know, and and like I mean, people And it's not like they they can't do world positive things in tech in addition to having fun and having their own alcohol brand. Think a lot of group chats out there would like to have their own alcohol brand. So Anyways, I'm excited to see it play out. I'm excited to try it.
Speaker 2:Yeah. Yeah. Yeah. This is definitely not like a pivot of the brand. In the same way that the the talking politics is not a pivot of the brand.
Speaker 2:You go back to the very first episode of the All In podcast, they were talking politics. They were talking about COVID and stuff. Like, it was it was like, they've been very consistent in all this stuff. So anyway, if they're trying to pump their tequila, they gotta buy a billboard on Adquick. Adquick.com.
Speaker 2:Straight the
Speaker 1:+1 01. I wanna see the one I wanna see
Speaker 2:every single billboard on one zero one. The besties all in tequila. Out of home advertising made easy and measurable. Say goodbye
Speaker 1:to the headaches of out of
Speaker 2:home advertising. Only AdQuick combines technology, out of home expertise, and data to enable efficient, seamless ad buying across
Speaker 1:the globe. If you're trying to get the attention of a CEO, maybe you wanna sell some enterprise software to them, why not give them a bottle of the Besties? Yeah. And that'll that'll get you a demo real quick.
Speaker 2:I think VC should be sending this to founders. Founders should be sending it to VCs. It's it's a it's gonna be a hot commodity if you can get your paws on one of these. Yep. Anyway, Hims and Hers stock drops more than 30% after Novo Nordisk breakup.
Speaker 2:This is from the Wall Street Journal. Of course, if you're interested in stocks that are dropping because you're going short or are popping because you're going long, you gotta get on public investing for those who take it seriously. They got multi asset investing industry leading yields in the trust of millions. Anyway, the Wagoovie maker, that's Novo Nordisk, has cut ties with the telehealth company, Hims and Hers, citing allegations of deceptive marketing practices and drug compounding. I feel like they were the drug compounding how is this an allegation?
Speaker 2:I thought that's, like, what HIM's was saying they were doing. They were, like I felt like there was a lot of press about them being, like, proud of drug compounding because there was a shortage. But I guess it's I guess it's alleged at this point, so we'll dig into this. Novo Nordisk abruptly ended its partnership with Hims and Hers after the Danish drugmaker accused the telehealth company of illegally selling cheaper copycats. Okay.
Speaker 2:So Yeah. This is after the FDA's ruling about whether or not there was in fact a shortage. So Yep. The Novo said on Monday that it had concerns about the safety of knockoff versions of Ogivy and that Hims and Hers deceptive marketing of such knockoffs put patient safety at risk. In turn, Hims and Hers accused Novo of pressuring it to steer pair to steer patients to Wagoovie regardless of whether or not it was the best option for patients.
Speaker 2:News that the deal was scrapped weighed heavily on Hims and Hers shares, sending them down 32% in afternoon trading in what would be the largest decline for the stock on record. Novo Nordisk's shares were down nearly 6% in New York. The messy breakup comes less than two months after the company's unveiled what they called a long term collaboration to directly provide Wagoovie to Hims and Hers patients.
Speaker 1:Okay. So this was somewhat predictable. Right?
Speaker 2:Talked to TMZ Parker about this.
Speaker 1:HIMS has been on a generational run Yep. In the markets Yep. Because they're one of the best easiest ways that people can just go buy Wegovy.
Speaker 2:Yes. Right?
Speaker 1:Now if you're Novo Nordisk and you're like, cool. We have we're selling through this channel. But then the retailer, in this case, Hims, is telling customers as they come in, hey, we know you wanted Wegovy, but how about this sort of cheap knock off, this compounded version? Yep. Now Wegovy can it is probably legitimately concerned about safety issues.
Speaker 1:Mhmm. There can be issues around compounding and and and drug safety. But then the bigger issue is like, hey, you're marketing you're using our brand and our products to acquire customers. Mhmm. And then you're getting those people in the door and you're saying, hey, how about this sort of cheaper Mhmm.
Speaker 1:Off the shelf alternative. And so I have to imagine that they this this is Novo Nordisk recognizing the value of their brand in this massive massive fast growing category and saying, we're not gonna let you leverage our brand to compete on these sort of high margin, you know, knock off products.
Speaker 2:Yeah. I mean, like, the the Instagram ads that I saw were definitely pushing the brand name for real. Yeah. And there was a while when they wouldn't use the term Viagra, and they would use ED meds and Sildenafil and Tadalafil and all of those different, like, they would use the actual underlying compound name. And they pushed it so hard that people actually started knowing the name of the underlying compound.
Speaker 2:Yeah. But, yeah, the brand name
Speaker 1:something was respond? Do you think they'll introduce a 10 in one shampoo?
Speaker 2:10 in one.
Speaker 1:That includes compounded Yep. GLP ones, Viagra, hair loss Yeah. Medication. What else could Adderall? They could put some caffeine.
Speaker 2:I mean, this is Cognizant, Rhodiola, Panax, Ginseng, L Theanine, caffeine. You get some Creatine one in here.
Speaker 1:Creatine soap with
Speaker 2:Creatine soap would be good.
Speaker 1:All of the above.
Speaker 2:Eli Lilly is the one that lost the patent. Right? Because they forgot to pay the
Speaker 1:No. I think it was Novo Nordisk.
Speaker 2:It was Novo?
Speaker 1:Wow. They have
Speaker 2:They're all over the place.
Speaker 1:Is it is interesting to watch. Yeah. Novo Nordisk, how much do you think their stock is up over the past five years?
Speaker 2:Oh, 500%?
Speaker 1:Try a 113%.
Speaker 2:A 113, but it was a big company.
Speaker 1:Yeah. Yeah.
Speaker 2:But it's been a
Speaker 1:Big company, but still, and how much do you think what what do you think the the chart looks like over the last year? Oh, it's down. The GLP one boom.
Speaker 2:It's down because it it ripped up so high.
Speaker 1:Down 50%. And then
Speaker 2:went down 50%, and they and they rotated out their CEO. That's right. We covered that. Yep. Hims and Hers is at 9,000,000,000, 9,660,000,000.00.
Speaker 2:That is down from a high of maybe 15,000,000,000. And it's one of those interesting, like, SPAC targets that I think they got out through a SPAC, and they they never really went down that far. They were down, you know, yeah, like like, you know, 30%. I mean, they SPAC to $10 a share, went down to $4 a share, kinda grinded back up. Now they're at $40 a share.
Speaker 2:So if you if you bought this if you bought the SPAC and you did hold for five years, you're, you know, you're sitting on a three x, four x. Not bad. So, you know, kind of a narrative violation around the around the SPAC, like all SPACs are bad narrative.
Speaker 1:Yeah. Yeah. Mean, it it and and the CEO Andrew Dudum, we should have him on the show Mhmm. Said that, you know, I think his position, you know, we have no idea what the details of their of their contract actually look like. But fairly reasonable position to say, we're not going to just sell one product just, you know, because it's true.
Speaker 1:Not every patient is gonna respond well to Wegovy. Yep. And they should have options.
Speaker 2:So before striking deals with brand name drug makers, telehealth firms were selling lower cost knockoff versions of these drugs made by compounding pharmacies, taking market share away from brand name makers. The compounding pharmacies had been allowed to mass produce knockoff versions conditionally because the original drugs were in short supply. Now the whole compounding pharmacy dynamic is very different because I feel like the the the laws were created such that if a big pharmaceutical company couldn't supply you the drug, like, local pharmacist could compound it, but then the the the startups kind of, like, rolled up all the compounding pharmacies. We're doing it like a much bigger scale, and it wasn't exactly what was intended originally, but I'm not sure about that.
Speaker 1:Yeah. So the big question
Speaker 2:Is when there should
Speaker 1:be more start. Well Yeah. The big question, what will there be more regulatory action around compounded GLP ones? Mhmm. And what will that do to their business?
Speaker 1:So roller coaster of a year for HIMSS. Yeah. Partnership was in announced in April 2025, popped, terminated just a few months later with a 35% drop.
Speaker 2:Well, if you're on GLP ones and you're losing weight and your wrists are getting thinner, you need a watch that can adapt to your new wrist size. You need to go on getbezel.com. Your bezel concierge is available to source you any watch on the planet. Seriously, any watch. And you can find a watch with metal links that you could take out a link if you've lost a lot of weight.
Speaker 2:Or you could add one back if you're
Speaker 1:in a
Speaker 2:bulking cycle.
Speaker 1:Yeah.
Speaker 2:Yeah. So Keep things like an nablus. When you're ready to start Exactly.
Speaker 1:Start the bowl.
Speaker 2:Exactly. This is this is key. This is very, very key. Anyway, comment from the, from the CEO Andrew Dudum. He said, hims and hers will continue to offer access to a range of treatments, including Wegoovy, so they continue to battle it out.
Speaker 2:It'll be interesting to follow this, and we will, of course, be talking to Cremieux about this later today. There is a investing visual post here that I pulled up, about how HIMSS makes money. They're making $576,000,000 online, $10,000,000 from wholesale. That brings their total revenue to $586,000,000, which is up a 111% year over year. Gross profit there, 428,000,000.
Speaker 2:Operating expenses are 337,000,000, leaving them 92,000,000 in net income or 92,000,000 in operating income and 50,000,000 in net income. So 50,000,000 in net income for a $10,000,000,000 stock, pretty pretty high price to earnings ratio if that's their or if that if that's an indication of their earnings, but it's been a big growth stock on a major, major trend, GLP one. And so the the the retail army has been coming out in force. King, KSM's capital said, it felt too good to be true for HIMSS to go this long without some FUD. This is more like it.
Speaker 2:Just bought $2,323,100 shares and, top secret stuff.
Speaker 1:Are what does the retail army call themselves?
Speaker 2:I don't know. Himbos or something.
Speaker 1:Say, I'm really him.
Speaker 2:Yeah. HIMSS, look at the data. You can see their operating margin has improved a ton since they've gone public. They started out at negative 40% operating margin and climbed up every single quarter. They're now at a 6.5% operating margin.
Speaker 2:The number of subscribers in the millions has gone from, they were at point six, so 600,000 subscribers. Now they're at 2,400,000, and their total revenue has gone from $272,000,000 in revenue. This is the last twelve months. Now they're almost at 2,000,000,000. And so they've almost 10 x revenue from that SPAC price.
Speaker 2:And that's what's driving the stock overall to be up three to four x off the original SPAC price. So Yeah. Very good news. If you're Andrew Dudum and you need a break from grinding it out at HIMSS, book a wander. Find your happy place.
Speaker 2:Book a wander with inspiring views.
Speaker 1:Summer, Andrew.
Speaker 2:Hotel great amenities, dreamy beds, top tier cleaning, twenty four seven concierge service. It's a vacation
Speaker 1:Imagine you're in one of the gnarliest PR crises of your life and two guys in suits on a podcast. So you never matter just like, yeah. Take a vacation. But Vacation. I think it's a bad idea.
Speaker 1:Take a foot the gas. Hence, this is up 2% today, John.
Speaker 2:You're good.
Speaker 1:So it's time to, you know, treat yourself.
Speaker 2:Yeah. Take your foot off the gas.
Speaker 1:Head to wander.com.
Speaker 2:And speaking of foots on gas, Ford, which makes cars that you can put your foot on the gas of, is now under pressure to put their foot on the gas of acquiring rare earth elements. They are having difficulties obtaining vital magnets
Speaker 1:They're scrambling.
Speaker 2:With rare earth elements despite the deal the UN United States struck.
Speaker 1:And I just wanna highlight. I think the silhouette of this car is fantastic.
Speaker 2:The the Mach E GT.
Speaker 1:You're a fan. I've only owned one Ford Raptor. Yeah. It was a very fun car. Mhmm.
Speaker 1:Not very practical.
Speaker 2:Yep.
Speaker 1:If you need to park in cities and things like that. But Yeah.
Speaker 2:If you want practicality Ford GT.
Speaker 1:I took a drive. Yeah. Practicality, all about the Ford GT. Yep. But I took a drive.
Speaker 1:My buddy, Matt has a Ford Mustang. What is it called? Of course. Mach E Rally. It's like this this Rally version of the Mach E.
Speaker 2:Okay. So it's still the crossover. Still like an SUV, more or less.
Speaker 1:Sort of. It's like a hatchback. Yeah. Yeah.
Speaker 2:But it's not a car. Like, I would not call this a car.
Speaker 1:It it is a very, very cool car.
Speaker 2:It's fun?
Speaker 1:It it to him, he said it's like the most fun car he's ever owned. Let's go. And he's had pretty something from from pretty much every manufacturer at this point.
Speaker 2:Well, he should hold on to it because they might be struggling to make them in
Speaker 1:future. Up. Yeah. Ford Motor still facing difficulties obtaining vital magnets made with rare earth elements despite a deal The US struck with China to ease export controls, a company executive said Monday. It's hand to mouth, the normal supply chain scrambling that you have to do, said Lisa Drake, a vice president overseeing Ford's industrial planning for batteries and electric vehicles.
Speaker 1:Ford in May stopped production at a vehicle factory in the Chicago area because of a magnet shortage. The situation has improved, but the company still needs to move things around to avoid manufacturing shutdowns given the scarcity of the materials. Drake told reporters during a briefing at an EV battery plant the company is building in Michigan. China in April began requiring companies to apply for permission to export magnets with rare earth minerals, including dysprosium and terbium. Dysproium.
Speaker 1:The country controls roughly 90% of the world's supply of these elements, which help magnets to operate at high temperatures. Yes. Much of the world's modern technology from smartphones to jet fighters rely on these magnets.
Speaker 2:So everyone sees that 90% of the world's supply and thinks it's 90% of what's in the ground, and that's not the case. Yep. What what what is really going on here is that the industrial capacity to remove the rare earth elements from the ground is 90% in China. So they have 90% of the mining companies, 90% of the mining equipment. They don't necessarily have 90% of the actual rare earth metals in the ground, and the rare earth metals are not that rare.
Speaker 2:But we hear this every once in a while where we say, oh, like, there was a ton of rare earth deposits found in Ukraine or found in America somewhere, and that's not enough. Like, just having them in the ground, you still have to go and dig the mine, get the approvals, make sure and the like, one of the big reasons this was offshored was because of environmental considerations, which I find so frustrating because it's one world. Like, it's an Earth. Like, if you wanna protect the Earth, moving something from one continent to another does not save the Earth. Like, I mean, I guess there's a little bit of, like, yes.
Speaker 2:So there's runoff in one river, and so there's local pollution that you're avoiding. But in general
Speaker 1:You've been to China, John. It is dark. Yeah. The sky is legitimately dark.
Speaker 2:But I I think that I think that, like, a lot of the pollution that's going on over there is affecting us too. For sure. And so especially if you think about, like, the broader glow warring picture, the broader, like, poisoning of the air. Like, that stuff's not gonna come over here at some point. Like, I I I find that hard to believe.
Speaker 2:It would have been much better to just focus on ramping up production in America and then actually doing the technological innovation to develop cleaner processes and not just offshoring it because it's, like, dirty and then just doing the dirty stuff over there. I don't know. Maybe maybe there's an argument for it, but it it feels like it would the whole the whole offshoring project was wrapped in this veneer of, like, no hear no evil, see no evil, speak no evil, that type of vibe. Like, if we can't see the pollution here, it's fine. It's not happening.
Speaker 2:And it's like, it is still happening. It's just you don't have control over it.
Speaker 1:Yeah. No. You have massive supply chain risk.
Speaker 2:Exactly. And I and I don't think I don't think it actually had an impact on the like, the stated goal of the environmentalists that pushed rare earth manufacturing overseas did not was not achieved. And so that that feels like like a failure. And then on top of that, we lost control of this important supply chain. So Ford isn't the only carmaker struggling to obtain these magnets.
Speaker 2:Several other carmakers say that the pace of export license approvals for rare earth magnets hasn't changed significantly. It isn't quite day to day, but it's week to week, said an executive at one of the car makers. Car companies have warned that they could be forced to halt factory work if they're unable to attain enough rare earth magnets. In the auto industry, rare earths are essential to electric vehicles because they allow a low cost way to ensure motors operate at high speed. Okay.
Speaker 2:So this should not affect the production of Raptor Rs. Thank goodness. I I was I was extremely worried. But so we might have to get most of those Mach E owners into Raptor Rs.
Speaker 1:Okay.
Speaker 3:But
Speaker 1:I don't think they're gonna be mad
Speaker 2:at that. That's something we have
Speaker 1:to Yeah. If we have to do it, we have to do it. We have
Speaker 2:to do it. In the auto in the auto industry, rare earths are essential to electric vehicles. President Trump struck a deal earlier this month for China to resume granting licenses to export rare earth magnets. The agreement, which has a six month limit, is aimed at allowing China to retain its chokehold on the critical minerals that give it leverage in future trade negotiations. So very, very rough.
Speaker 2:Anyway, how'd you sleep last night? Let's pull
Speaker 3:it up.
Speaker 1:I'm up praying. I got an 88, John.
Speaker 2:Oh, no. 82.
Speaker 1:There we go. Beat me.
Speaker 2:Do. Jordy's back on top.
Speaker 1:I'm back. Back. After a while. I'm back.
Speaker 2:What happened to me? I just I guess I stayed up too late or something. I was in my sweet spot. Oh, I guess I didn't fall asleep until 11:12.
Speaker 1:There you go. There you go.
Speaker 2:Gotta go to bed earlier. Back
Speaker 1:on the I'm back on the horse. Yeah. Back on the horse. It's okay. It should be a competitive week.
Speaker 1:Excited to see how it we gotta start putting we gotta start putting something on the line. Yeah. Like, the the loser on Friday gets no caffeine and nicotine the whole the whole stream.
Speaker 2:And they'll be sleep deprived too. That's that's double duty. That's that's extra hard. I don't know if that's possible.
Speaker 1:That would be torture.
Speaker 2:Anyway, our our first guest is coming in the studio in ten minutes. Let's take you through beautiful obituary to the founder of FedEx, Fred Smith. So Ryan Peterson posted a photo of the original FedEx plane. He said, r I p to logistics legend and outspoken defender of free trade. FedEx founder free Fred Smith, who passed away today.
Speaker 2:Below is the first ever FedEx plane, which you can see at the Air and Space Museum annex next to Dulles Airport in Virginia. Look at that. He must have been so stoked when he bought this. Like, his business was finally growing that he could own his own plane.
Speaker 1:Imagine how many likes he would have gotten on social media
Speaker 2:posting that pic. Yeah. Everyone's like, this is your flagship.
Speaker 1:To appreciate it. You're just trying to
Speaker 2:go viral. This is clearly.
Speaker 1:This is clearly for logistics.
Speaker 2:This is clearly for logistics.
Speaker 1:Total legend.
Speaker 2:Yeah. Oh, like burning VC dollars on a plane. Bear.
Speaker 1:I think he was using Vegas dollars.
Speaker 2:Doing cash flow. Yeah. Anyway, Max Meyer has a story that we should read through. Fred Smith and American Life. He says, Fred Smith died on Saturday.
Speaker 2:He was the founder of Federal Express, which became FedEx, one of the largest logistics companies on Earth. Smith was an entrepreneur in the truest sense of the word and an American patriot. He bent reality and created something entirely new, and he ran toward an extremely difficult problem. The fruits of Fred's work are as follows. If you give a box or document to FedEx by 6PM, you can have it delivered almost anywhere in The United States or Canada by 8AM the next morning.
Speaker 1:Imagine how key that would have been to business before
Speaker 2:Incredible. Incredible.
Speaker 1:Yeah. They're trying to do deals?
Speaker 2:If you give FedEx a package on on a Monday, it can be in Paris by Wednesday morning, or in Shanghai by Thursday for a few $100. Most of us take overnight delivery for granted, but take a step back. And the fact that we can essentially airdrop physical packages anywhere in the world is mind boggling. Smith's death is a reminder not to take the miracles of modern life for granted. They were built by people like Fred Smith.
Speaker 2:These are a few words about the man himself and the dazzling system that is FedEx. Frederick Wallace Smith was born in 1944 in Mississippi. Fred's grandfather, James Buchanan Smith, captained Mississippi River Steamboats. Fred's father, James Frederick Smith, was the founder of the Smith Motor Coach Company, which would go on to become Dixie Greyhound Lines after an acquisition in 1931. From steamboats to buses to jumbo jets in three generations, the family stayed on the cutting edge of motorized transportation.
Speaker 2:In 1925, the elder Smith seeking to create the transportation line out of Memphis converted a truck into a small bus and drove it himself. Within a few years, James Smith had dozens and dozens of coaches. By the time he died, suddenly in 1948, he commanded over 200. Wow. So he died four years after the son was born.
Speaker 2:Wow. So he basically grew up without a without a or maybe that's the grandfather. According to James Smith's obituary, no. No. No.
Speaker 2:No. He's the dad. Yeah. Yeah. So according to James Smith's obituary Terrible.
Speaker 2:Those 200 coaches each came to a halt during the regular regularly scheduled routes for one minute when the funeral began. The young Fred was just four when his father died. Wow. In In 1962, the young Smith enrolled at Yale University where he was fraternity brothers with future president George w Bush. Bush asked Smith to serve as the secretary of defense twice, and Smith declined twice.
Speaker 2:Wow.
Speaker 1:He Interesting because defense is, who was the founder that we had on the show? I'm completely blanking on his name, the name of the company, and who invested in the company. But he had this defense focused supply chain management startup. Yes. It was one of our probably first
Speaker 2:You're talking about Rune, logistics. Rune.
Speaker 1:Yes. But something he said about
Speaker 2:Not not the AI poster. The legendary poster. Not the legendary poster. The logistics company. Of OpenAI.
Speaker 2:Yes. So the young Smith was a member of the infamous Cloak and Dagger secret society.
Speaker 1:He is not beating the
Speaker 2:Secret society. State It was at Yale that he wrote his first paper about his concept for using airplanes to deliver packages with a hub and spoke system. Wow. He was like, I got an idea. I got a business plan.
Speaker 2:I just need somebody to build it for me. I. E. That instead of point to point package transport, everything would be brought to a single hub by air and then redistributed. We talked to somebody about this on the show.
Speaker 2:How how now we think of it as like inefficient to bring everything to one place and then and then do the hub and spoke thing because now we have, you know, DoorDash and Waymo and, you know
Speaker 1:Last just last mile.
Speaker 2:All all these different last mile solutions. But at the time, this was revolutionary. Everything would be brought to a to a single hub. Repeat that cycle every night while turning a profit, and you have a viable system for overnight delivery. The paper is said to have earned an average grade roasted.
Speaker 2:In 1966, the founder of FedEx graduated from Yale and volunteered to receive a commission in the Marine Corps. He served two tours in Vietnam. For his service in Vietnam, president Nixon decorated Smith with a silver star and a bronze star. The silver star citation read in part, unhesitatingly rushing through the intense hostile fire to position the heaviest con to the position of heaviest contact. Lieutenant Smith fearlessly removed several casualties from the hazardous area and shouting words of encouragement to his men, directed their fire upon the advancing enemy soldiers successfully repulsing the hostile attack, moving boldly across the fire swept terrain to an elevated area, he calmly disregarded repeated North Vietnamese attempts to direct upon him as he skillfully adjusted artillery fire and airstrikes upon the hostile positions to within 50 meters of his own location.
Speaker 2:Wow.
Speaker 1:What a life.
Speaker 2:And continued to direct the movement of his unit. Wow. What a badass.
Speaker 1:So at this point Yeah. He's nineteen seventy three. He's 29 years old. He lost his dad at four. He goes to Yale with George Bush, then goes to Vietnam, does two tours, gets a silver and a bronze star.
Speaker 2:It's crazy.
Speaker 1:War hero. Yeah. And now he's decided to launch Federal Express in Memphis. Smith had studied military procurement and been working on the idea for nearly ten years after writing the Yale paper. He writes this And he's like, okay, ten years later.
Speaker 2:Most founders have an idea and they pivot in two months.
Speaker 1:So he writes this paper
Speaker 2:He's working on
Speaker 1:a
Speaker 2:decade an upgrade. This is like nineteen
Speaker 1:years later.
Speaker 2:This is same
Speaker 1:Ten years later. The company launched with a fleet of 14 French Dassault jets, which delivered a few 100 packages on their first day of service. There we go. Wow. He's just like, just zero to one.
Speaker 1:Yeah. According to FedEx, Laura Smith named his new service Federal Express because he hoped to attract the attention of the Federal Reserve Bank, a prospective customer. They're moving very But it's funny like that the name you would you would think Federal Express, you would just assume that it's an extension of the government.
Speaker 2:I did when I was a kid. I was like, just
Speaker 1:pretty smart.
Speaker 2:United Postal Service. But US United States Postal Service. These are
Speaker 1:all FedEx. Like, it was just a better product experience than USPS. Even as a kid, you kind of Feel it out. Feel it out. The system that Smith had conceived at Yale worked.
Speaker 1:It really worked and nothing like it existed. The U USPS existed principally to deliver letter mail and didn't have aircraft. UPS was a giant of a ground delivery throughout The United The US. Federal Express though would specialize in rapid air delivery. Just five years after the first Federal Express flights.
Speaker 1:The company was list listed on the New York Stock Exchange. Wow. Today, the FedEx fleet consists of nearly 700 aircraft and about 200,000 vehicles, and FedEx employs half a
Speaker 2:million workers
Speaker 1:around the world.
Speaker 2:Wow. Every night at the FedEx super hub at the Memphis International Airport, hundreds of jets land at the airport.
Speaker 1:The ultra hub.
Speaker 2:Yeah. The super hub. Bringing packages to its many sorting machines. A few hours later, the jets take back off to every corner of The United States and the world at their destinations. The other half of the fleet awaits feeder aircraft to take packages to smaller airports and the FedEx Express trucks that take packages to their final destination.
Speaker 1:The financial finesse needed to actually make a model like this, not just lose massive amounts of money.
Speaker 2:Yeah. You could go
Speaker 1:out of business
Speaker 2:so many times. And, yeah, there is that story about him almost losing money, going to Vegas, and then putting the the the treasury on black and winning it back and doubling it and making payroll the next day, saving the company. Maybe apocryphal. There's been rumors that maybe it was like, you know, some mob related thing and he took a
Speaker 1:blow in Yeah, the tinfoil hat explanation is like, you know, he
Speaker 2:either Either way, got it done.
Speaker 1:He did some type of, you know, and this is entirely internet speculation. Yeah. It's probably not true. But it's possible that he, like, you know, was, you know, needed a reason to have a bunch of cash. Right?
Speaker 1:Yep. And and maybe did something else. But, the story is incredible and it will live on. Yeah. Smith retired from FedEx in 2022 after almost fifty years leading the company.
Speaker 1:He has survived by nine of his 10 children and his wife, Diane, his daughter, Wineland. Smith Thrice died in 2005 of a terminal cardiac condition. Very sad. And one has to think that his father, Mr. Smith, who died with 200 vehicles to his name would be impressed that his son died with 200,000 vehicles to his name and a fleet of jets to match.
Speaker 1:So, incredible story and may he rest in peace.
Speaker 2:Anyway, let's move on to the timeline. Also, one of the greatest logos of all time, the FedEx logo. Has that secret arrow in it? Love that.
Speaker 1:Oh, yeah.
Speaker 2:Yeah. Hidden in. Once you see it, you can't uneaten. Not this logo. The next logo, the more modern FedEx logo has the secret arrow in there once you see that.
Speaker 1:They're moving. They're moving. And there, if you want to understand more about Frederic Smith Mhmm. Go to Founders Podcast Yep. And listen to episode one five one, 151
Speaker 2:on Unsubscribe to Arena Mag to read more of, Max's writing. Check it out. Arenamag.com. What what else is
Speaker 1:We have a post here from Gaurav Vora. He says, chat, remind me to check this again in a year. I wonder if he's saying he's throwing it in chat Yeah. Saying, you know, check this again in a year. This is a graph.
Speaker 1:It'll be kind of hard to read, but it's showing chatbots versus search engines who's winning the traffic war. Showing that
Speaker 2:Google is getting 1,600,000,000,000 total visits to ChatGPT's half of 50,000,000,000, basically. So Google is much bigger.
Speaker 1:1.4 year over year.
Speaker 2:And ChatGPT is up 67%. Yeah. I mean
Speaker 1:It's a 34 x gap today. AI chatbots drew 55. Isn't that wild, Yeah. I wonder if OpenAI accounts for out of 55,000,000,000 visits, OpenAI accounts for 47,700,000,000. Yeah.
Speaker 1:So users
Speaker 2:81% of all chatbots is OpenAI. I feel like it's even higher now.
Speaker 1:But it's crazy. So AI chatbots are 2.96 of search engine traffic today. Varav is calling it or wants to check back in a year. Yep. It's very possible we could see double digit search engine traffic into chatbots.
Speaker 2:Yeah. I I do wonder if, like, site visits is even the correct metric here because it's more like time on-site and attention because I'm pretty sure I Google things, like, for just two seconds, Google something. I will use Google as a calculator. But and I won't go to ChatGPT for that. If I just have to multiply two numbers together, I'll just throw that in
Speaker 1:there.
Speaker 2:Are you
Speaker 1:not a power user?
Speaker 2:No. It's it's faster.
Speaker 1:You
Speaker 2:don't. It's it's all up to believe in AI. Yeah. Exactly. I'm still, like, checking out this whole AI thing.
Speaker 2:Starting to learn about it. But but, yeah, I I I think the more relevant metric is actually total user minutes. Like, much mindshare is going on. Because when I'm in, I'm probably in ChatGPT longer than I'm in Google, but I might be hitting more page views on Google than ChatGPT queries because I might I might throw a deep research query and then spend ten minutes reading Chatting back and forth. I'm very rarely That's interesting.
Speaker 2:Throwing a Google search out and then reading the blue links for ten minutes. That's just not how
Speaker 1:it works. So, anyway Also interesting to think about how they're counting visits. Right? Yep. You're having this conversation with ChatGPT where oftentimes you're repeatedly running news new hits.
Speaker 1:Right?
Speaker 2:Yeah.
Speaker 1:Against Google. So
Speaker 2:Anyway, we have our first guest in the studio. Kermieux, welcome to the stream.
Speaker 1:There he is. Looking handsome as ever.
Speaker 2:How are guys?
Speaker 1:How are you doing?
Speaker 4:Doing well. Can you hear me?
Speaker 1:Yeah.
Speaker 2:We can hear you fine. Perfect. I'm I'm I'm looking forward to to, you know, a future date when we can potentially see your face. But for now, your avatar looks great. Soon?
Speaker 2:Okay. I'm looking forward to it. We wanted to have you on to chat about a couple things. What's going on in GLP one world, reaction to the hims and her hers breakup? Jordy, where would you like to start?
Speaker 1:High level update on the GLP one market.
Speaker 2:Yeah. I guess, I mean, it's this miracle drug that keeps on we keep on learning about new, new diseases or problems that it can treat. It seems like it can't just stop. We got the diabetes, then weight loss, then maybe gambling addiction and other stuff. Walk us through how people are using or how doctors are prescribing GLP-one for various conditions and what's most promising, what's on the horizon, and what you're tracking next.
Speaker 4:Yes. I'm actually really glad you asked this right now because right now the ADA, the American Diabetes Association Conference is ongoing and it seems like every other presentation is about GLP-1s. They are just talking about all the latest advances. Just yesterday Amgen showed off a once monthly instead of once weekly injectable. Eli Lilly showed off an amazing combo therapy with an inhibitor called Bimagrumov.
Speaker 4:It's a very weird name that makes people not lose any muscle when they're on the drugs. They've been showing off just incredible advances all week. They've been showing off new treatments for like different conditions and whatnot. There are increasingly many indications these drugs are getting approved for. In December, Tirzepatide was approved for sleep apnea.
Speaker 4:There's osteoarthritis indications incoming. There might even be there's some effort being put into people trying to do something for cancers as well because it does seem to help with obesity related cancers and it seems to help for I don't know why with hematological cancers, the blood cancers. I don't understand really the mechanism behind that but it seems like it's just hitting every single indication now.
Speaker 1:Wow. Why do you think that is? I have some guesses.
Speaker 4:I think the big reason has to do less with the direct effects of GLP-1s and more with the fact that obesity sucks. Obesity is just really really bad. It affects so many different systems. It makes your health worse on so many different levels and the things that lead to obesity are also quite bad like the bad diet, the bad habits, the not moving around very much. It's actually interesting.
Speaker 4:A lot of people after they've been on these drugs for a little while, they decide to move around more. They become more likely to go to the gym. They report their physical functioning has improved. It turns out that getting fat has made them get to the point where they are no longer looking to be active and so they just kind of fall into a hole. Practically everything is improved I think for that reason and there are some improvements that are due to direct effects of the drugs.
Speaker 4:This has like for example this one is major adverse cardiac events, major adverse cardiovascular events. Those seem to be reduced in number immediately after starting the stuff which suggests that there's a mechanism that's pretty direct. And this mechanism is also independent of weight loss. So that seems to be how that works. Same with chronic kidney Are looking
Speaker 1:you know, feel like the sort of bio hacker alternative health world has for a long time, you know, talked about cancer being, you know, this metabolic disease. Do you think that could be a factor and why it should could show some promise in in treating various types of cancers?
Speaker 4:Yeah. I I think it does makes a lot of sense in that direction for the obesity related ones but I don't think it makes sense. So years ago there were people who suggested I think quite wrongly replacing certain cancer therapies with fasting because it seems to like augment the effects of cisplatin based therapies and stuff. That that didn't really hold up very well. It was a bunch of theory work that didn't go anywhere and I don't think that this would help through those purported mechanisms which again I don't really believe in.
Speaker 4:I really think it's mostly just cutting down on for example the number of cells you have or well not really number of cells you have but like the activity of being fat. It's just you have you're not as big of a person. You have less to less opportunity for cancer to really hit you when you are Smaller. Yeah. Smaller.
Speaker 4:It's like how short people tend to live longer than everything else.
Speaker 2:Bad for me.
Speaker 4:Bad for you.
Speaker 2:Fairish for me. Well, how much of the recent, like, advances in GLP one indications or or recent, like, benefits have been just looking at all the people that are taking GLP-1s for weight loss or diabetes and then seeing a secondary effect in the population that's actually running the drug versus a new double blinded trial for a different indication? Are both things happening? Or is it are are these drugs diffuse enough that you can just look at the overall population and get an idea of what's happening?
Speaker 4:Yeah. So for approved indicate FDA approved indications like obstructive sleep apnea, you do have to run additional trials. But they are seeing that these things are possible from their trials. They're seeing these things are possible from the literature and they're running based on that. So in some cases, they do see these secondary endpoints.
Speaker 4:In fact, obesity is an example of this. Back in the day, when they first introduced these drugs, back around 2005 or so, the CEO of Novo Nordisk at the time actually said, and I quote, obesity is primarily a social and cultural problem. It should be solved by team by means of a radical restructuring of society. There is no business for Novo Nordisk in that area referring to obesity.
Speaker 2:Wow. Yeah.
Speaker 4:So they their scientists had to push for about a decade to go, hey, you should look at the weight loss data.
Speaker 1:It's really
Speaker 4:impressive. And then they finally, finally did it. And now
Speaker 2:we have weight loss drugs.
Speaker 1:Yeah. That is that is wild. What, what concerns you? What's the catch? Seems, know, somebody It's too good to
Speaker 2:be true.
Speaker 1:Yeah. Somebody might say, is it too good to be true? We're running this sort of massive experiment on huge swaths of the population right now. Seems to be, you know, many, many positive indicators. But what are kind of red flags that you think people should be kind of looking out for or the scientific community should be wary of broadly from your point?
Speaker 4:Yeah. So I think that physicians need to be careful about prescription. And they already are. They like they don't tend to prescribe this to people who are very skinny. But the fact that this stuff is available through the gray market Mhmm.
Speaker 4:Very easily without prescription for very very cheap. Unfortunately, I've contributed to that. Has led to a lot
Speaker 1:people In what way? You mean just Oh,
Speaker 4:I made a I made a guide to doing this and a few thousand people actually like paid me $8 apiece to see all the details on it. And now I have a few thousand people who unprompted, I didn't tell them to do this, message me their weight loss progress and such. So Wow. Through writing this guide a few months ago, I've led to almost 12,000 pounds of weight loss and that's just what's been reported to me. Not everybody's telling me all the And
Speaker 1:is that at a high level people just finding pharmacies that will compound the drug for them? What does that actually look like?
Speaker 4:So what it looks like is buying from China. So you are ordering from a Chinese factory. You are, testing the purity of the stuff and finding, oh, it's 99.9%. I can use it and then using all of that. Yeah.
Speaker 4:It is not sold by Novo Nordisk, Eli Lilly, or any other major company working on these drugs. It's just from some Chinese company that is ripping them off.
Speaker 2:So is that patent infringement? I we we we've been seeing that kind of go back and forth, like Hims and Hers was able to compound because there's a shortage. I would love to see Nev and Nordisk do about EICP.
Speaker 1:And say, hey. Can you please knock this off?
Speaker 2:Yeah. I mean, they could do import bans or or or seize the packages at the ports. Like, that's the design of this. You can't just buy a knock off iPhone and have it delivered through the Port Of Los Angeles. Like, if a whole truckload comes through, they should stop that, but it's clearly not happening.
Speaker 4:So right now, the FDA allows this to happen because they have an exemption for research chemicals. You're technically not supposed to use these things, but everybody knows that everybody uses them. It's just how it is.
Speaker 1:They're researching weight loss on themselves. Research maxing. Yeah.
Speaker 4:I'm doing So
Speaker 1:you were saying the the potential red flags or or things to be wary of is around what what about the prescription activity concerns you?
Speaker 4:Mostly abuse. Not really prescription activity as so much as the abuse. There are people who are obtaining these drugs. There are people who are getting other people to get the prescriptions. There are people who are not really visiting doctors using telehealth services to get prescriptions and they are not in need of them.
Speaker 2:But why would someone use this? This doesn't feel like something that has like the euphoria associated with a stimulant.
Speaker 1:So the concern I think could be somebody has testosterone. Body dysmorphia. There are Is that what's going on? Really skinny Yeah.
Speaker 4:Unfortunately be skinny. I've met a few dozen women now who have taken these drugs and been Yep. Around a hundred and twenty, a hundred and thirty pounds Sure. Normal height.
Speaker 1:Yep.
Speaker 4:And now they're around a hundred and they are Okay. They don't look good anymore. They look like they're aging very quickly. They look unfortunate. And it's it's sad to see that they're using these drugs when they didn't even need them in the first place.
Speaker 1:Interesting. Same negative effects of of start actual starvation. Yeah. Yeah.
Speaker 4:Yeah. Yeah. Yeah. You start getting really bad harmful effects. You probably are eating at that point more lean mass than fat mass.
Speaker 4:It is quite bad.
Speaker 2:Yep. Yep. That makes sense. So, yeah, that's always the risk. Interesting.
Speaker 1:How are you what what was your reaction to the HIMSS Novo Nordisk, partnership blow So
Speaker 4:I actually think this is a big failure on Novo Nordisk part. It seems as though what they did was they partnered with HIMSS in order to catch them. My understanding is that HIMSS has been selling a little too much more than they've been supplied by the person producing the drugs which is interesting. I guess we'll get more details relatively soon if
Speaker 1:the launch commencing all So you're saying theoretically Wegovy gives them a million doses and then HIMSS is somehow selling one point two million doses of Wegovy. Is that is that what you're
Speaker 5:Yeah.
Speaker 4:Something like that. They're either compounding themselves or ordering
Speaker 1:from
Speaker 4:something from China. They are doing something. And I don't know if that will hold up or show up in trial. It's an accusation.
Speaker 1:Mhmm.
Speaker 4:It's so I guess it's alleged. But we'll we'll see. We'll see.
Speaker 1:What about what about HIMSS's general position that patients should have as much choice as possible, as much access, you know, access at different price points, things like that?
Speaker 4:They are completely correct to do that. I think that HIMSS providing different dosages than are provided by Novo Nordisk is a wonderful thing. There are people who can get legitimate uses out of sort of micro dosing this stuff because it gets rid of noise. I've met many many people now and this is an increasingly common thing it seems. I don't I really don't know how far this is gonna go.
Speaker 4:But a lot of people who are micro dosing this because it helps with their ADHD. Their ADHD is like the pathological end of food based distraction. They're not losing weight from this anymore but they are just using it in order to not think about food at all. It gets rid of that nagging feeling in the back of their head and they can focus
Speaker 1:on Yeah. I I did a twenty four hour dry fast
Speaker 2:Mhmm.
Speaker 1:From Friday night to Saturday night. Yeah. And all day Saturday, I was shocked at how much time my I was just thinking about food and water. Wanting food and water. And I was and I and I was very freeing to some degree because I was like, well, I'm not just not having any until around dinner But so many points throughout the day, my mind was just going to, oh, I should go get a little tasty drink from the fridge.
Speaker 1:I should Good. Know, I should
Speaker 2:I'm not addicted. I'm not addicted.
Speaker 1:To water. I'm not addicted to water and food. I'm not beating the water addiction allegations. Yeah. Big water guy.
Speaker 1:What what else are you tracking right now broadly? Any reactions to the NIH funding?
Speaker 2:Oh, yeah. 40%. We had Andrew Humorant on the show. He was breaking it down for us and what's going on with Jay Bhattacharya. What's your take on the cuts that have been proposed?
Speaker 4:So I don't know how much I should reveal about that. There's a lot of really good things coming. There might be some interesting developments there very soon. I'm especially hopeful about well, you know, actually, shouldn't say that. Just there there will be good news there soon.
Speaker 4:There will be some changes.
Speaker 2:I guess Yeah. Yeah. I guess zooming out like what what what does good look like to you? Are you generally in favor of taxpayer funded in Every American gets
Speaker 1:a $10,000 biohacking budget annually. Yes.
Speaker 2:You can buy research chemicals.
Speaker 4:Biohacking vouchers.
Speaker 2:Yeah. I mean, some people would say like, you know, the the big pharma companies are profitable. They should bear this expense, not the American.
Speaker 4:So they're not profitable enough. They have troubles with r and d. Yep. That public money is very, very useful. We need more public funding than we currently have.
Speaker 1:Mhmm. Yeah. When you actually look at biotech returns, you know, just looking at the asset class, the logical thing to do would just be to invest elsewhere. Mhmm. And that that was with historical levels of public funding Yeah.
Speaker 1:That that kind of contributed to those returns to some degree. Mhmm.
Speaker 4:Yeah. We have a dearth of funding at the moment and we need to increase it. We need to improve the allocation and also increase the amount. There's really no way around it. If we wanna keep making progress, you just have to throw more money at these things.
Speaker 4:And I think we will start doing that very shortly. These cuts are a little alarming to start, but they are not the end of the discussion. Remember, we are it's still only five months into
Speaker 1:Yeah.
Speaker 4:This presidency. And they have a lot of really big plans. They have a lot of trouble getting appointees through the senate. They they have a hiring freeze ongoing right now that is interminable. We don't know when it's going to end.
Speaker 4:But once that is over, they'll be issuing NPRMs left and right. It's going to be deregulatory like a massacre.
Speaker 2:It's gonna be wonderful.
Speaker 1:Wow. Well that's exciting to you in biotech right now in
Speaker 4:Well I wanted to say on the Ham's point with the different pricing and whatnot Mhmm. I think perhaps the most exciting thing right now in that area is that Novo Nordisk is run by idiots. Like, they're they're very smart people, but they're just total idiots.
Speaker 2:Okay. They
Speaker 1:Hey. Everybody misses a the the patent, you know, what what do they do? They
Speaker 2:they I think it was Eli Lilly, but they they missed
Speaker 4:the this is Novo Nordisk.
Speaker 2:Oh, those Novo that lost the the patent fee?
Speaker 1:Yeah. They just forgot
Speaker 2:to the bill.
Speaker 1:It's $450 or something.
Speaker 4:Yes. It was 250 and then they missed it and they were told, you're one year out, you had you pay a late fee, like an extra $200 late fee on it and
Speaker 2:They just didn't.
Speaker 4:They still failed. Yes. It is amazing.
Speaker 2:That's wonderful.
Speaker 4:So that enables a wonderful wonderful program that the FDA should pursue immediately. The FDA and the CMS have to collaborate on this. But it's the section eight zero four importation program It allows individual US states to import as much of the generic drugs or any other type of drug that's produced
Speaker 5:in
Speaker 4:Canada and has like an indication of proof here and whatnot as they want to reduce costs. So for example, Florida will, whenever they get this actually going, have much cheaper EpiPens. They'll be able to lower the cost of drugs by importing cheap generics from Canada. And because cheap generic Ozempic is going to be coming from Canada in 2026, every state can just jump on the program. If we're giving people we're giving Americans $5 a week Ozempic then we're going to see perhaps an end to the chronic disease crisis.
Speaker 4:We're going to see obesity tackled meaningfully. We're gonna see people getting hot again and make America hot again. The real meaning of
Speaker 1:The meaning of
Speaker 2:The real meaning of
Speaker 1:There we go. Inside baseball. That's amazing. Very exciting. Well, fifteen minutes was not enough time.
Speaker 2:Never enough.
Speaker 1:Back on again soon. Thank you for thank you for all the insights.
Speaker 2:Yeah. Thanks for coming on. We'll we'll we'll have you back on soon. This is great.
Speaker 1:Absolutely. And good one. And if you docs yourself anywhere other than here, we will be very upset. So
Speaker 2:Might have
Speaker 1:to Oh, no.
Speaker 4:Sorry, guys. You guys have a good one.
Speaker 1:Alright. Bye.
Speaker 2:Next up, we have Clement from Hugging Face coming into the studio. I believe he's here already. Very excited to talk to him. Fascinating business. We'll dig into artificial intelligence, open source, everything that's going on with What is
Speaker 1:the origin of the name Hugging Face? What is artificial intelligence? It's emoji, I believe.
Speaker 2:Yes. What is it?
Speaker 1:What's going on?
Speaker 2:How are
Speaker 1:you doing? Welcome.
Speaker 6:Hey. That's the emoji like that. The Hugging Face emoji.
Speaker 1:Yeah. Yeah. Yeah.
Speaker 2:Yes. But there's
Speaker 1:Come on.
Speaker 6:You guys are giving me shit about the name with a name like that, TBPM.
Speaker 5:Yeah. It's not much better than Hugging face.
Speaker 2:No. No.
Speaker 1:No. No. I'm not giving I'm not giving you shit. I just I just respond. It's just a fun fact.
Speaker 1:I just immediately see the the, obviously, the icon.
Speaker 2:But there's there's more to the name than that. It it it it it's an emoji, but it came because the product that you were building before this was related to that. Right? You break that story down.
Speaker 6:Yeah. Yeah. Yeah. When we started the company eight years ago, we were building some sort of an Tamagotchi AI AI girlfriend, a chat GPT before chat GPT.
Speaker 2:There you go.
Speaker 6:So very much kind of like an entertainment product. Yep. And so we picked the hugging face emoji because that's the one that we were using the most. Yep. We also had internally, it's this joke that we wanted to be the first company to go public with an emoji instead of the three letter ticker, you know, like on the on the Nasdaq.
Speaker 6:Yep. Still still fingers crossed for that. I hope nobody's gonna go public with emoji before us Yeah. And that they'll wait for us. Yeah.
Speaker 6:And then the community just loved it and started to put it everywhere, put it on their on their clothes, on their on social networks everywhere, so we decided to keep it.
Speaker 2:Yeah. The the the owning an emoji as a brand online, I think, is still under underrated strategy. We see it every once in a while. I've seen people use the the the the different fruit emojis for different things, and there was a whole strawberry story tied to OpenAI and stuff. But when you can condense down, I mean, it's a it's a coinage, essentially.
Speaker 2:Like, when you can own an emoji and have it say something, that's very powerful from a branding perspective.
Speaker 6:You can even search with emoji on on Google and the apps or, directly with the hugging face emoji.
Speaker 2:Yeah. I mean, you're gonna compete with the people that use the emojis for any other reason and there's always the risk that some steals the
Speaker 1:rest the video. Was pretty solid.
Speaker 2:But I think yeah. I think if you search the hugging face emoji, you're gonna you're gonna wind up in the right place. But walk me through the business today. How how how what what give me an idea of the scale and kind of the core business model.
Speaker 6:Yeah. So, we're one of the most used platform for AI builders. We just crossed actually today 10,000,000 AI builders using us. So it's mostly Let's go. AI scientists.
Speaker 6:Congratulations.
Speaker 2:10,000,000. That's a million.
Speaker 6:10,000,000 AI builders. So it's mostly AI scientists. Yeah. AI engineers, software engineers, building models, training models, optimizing models, sharing models, datasets, apps. So this is a new new repository.
Speaker 6:So it's model datasets, apps created on Hugging Face every ten seconds now.
Speaker 7:Wow.
Speaker 6:And and the way the way we monetize, because we you asked this question, is kind of like a pretty straightforward freemium model where most of the usage and the users are free and then a small percentage are premium. Yep. And usually, they become premium with premium features. So for example, enterprise features when, like, a big company like Google, like, Nvidia using us, obviously, they need user management, security, stuff like that, or when they need premium computes. Right?
Speaker 6:So when they need more powerful GPUs or or hardware to to run some other stuff they're doing on the platform.
Speaker 2:Yeah. 10,000,000 developers. Have you paid each one a $100,000,000, which seems like the going rate these days? It's a quadrillion dollars. It's a lot of money.
Speaker 2:But It's tough to hire all this. Even even if you're paying reasonable rates, you're still up in the in the in the trillions of dollars for the total market. I I mean, I would be interested to get your reaction. How have you been processing the news that the the talent market for artificial intelligence researchers seems to be hotter than ever in history, and we're getting into NBA money for top researchers. Like, do you think this makes sense given where we are in the cycle?
Speaker 2:What has been your take overall?
Speaker 6:Yeah. It's definitely hotter than it's ever been because when the CEO of OpenAI is saying that an other company is paying more to get OpenAI employees, it's really like you raise the top because, historically, obviously, they've been kind of, like, the highest paying for the company ever in terms of of packages. So when when when he's saying that they're getting, beaten, on on packages is quite quite phenomenal. You know, I hope it's not gonna continue too long, and there's gonna be more kind of like, almost, democratization of skills of, of AI building. Otherwise, I think we'll end up in a quite quite weird world.
Speaker 6:One of my biggest focus is kind of like how to fight concentration of power, concentration of skills, concentrations of resources in AI. And and I hope, we can progressively move into a world where everyone can build AI and not just get, like, a few hundreds AI scientists. That that'd be that'd be really great for me.
Speaker 1:What is your interaction today with the various prompt to code tools? What are you most excited about? You know, what what is that space? Because that that feels like kind of the the entry point. Somebody comes in and they they are making software for the first time, and that can be maybe a gateway into exploring the whole ecosystem of of Hugging Face and everything there.
Speaker 6:It's super exciting, right, to empower more people to be to be builders. We actually released last week our MCP that is integrated into into Cloud, into into ChatGPT that they they announced, actually, I think, a few days few days ago, into Codex, into Cursor. So we integrated with, with all of these. We've kept, like, a specific focus, which is not only to empower people to build, like, websites or simple apps, kind of like the previous generation of of apps. They're trying to empower everyone to actually build, AI models, right, which becomes really exciting because, it means that maybe everyone can start training, optimizing their own model that then they use themselves on their interface to kind of, like, keep building even more.
Speaker 6:So that's when you start to have kind of, like, maybe this flywheel in terms of AI AI progress. So that's kind of, like, our focus with the integration of Hugging Face MCPs with a lot of these tools.
Speaker 2:On the open source question, kind of like the decentralization of power, what was your interpretation of the Anthropic news today that it was in fact fair use for them to train their models on it was several 100,000 books or several
Speaker 1:7,000,000.
Speaker 2:7,000,000 books. Yeah. I was kind of, you know, optimistic that that would happen, but also just it seemed like we clearly needed some sort of legal ruling here to understand this stuff. But but how did you process it? Were you happy with the result?
Speaker 6:Yeah. I think this is good news. Obviously, I think this this ruling, they'll come and be different and specific depending on the use cases because you have to look at not only, you know, what's been used, but also how it's used, if it if it's kind of, like, transformational, if it's replace replacing competing with the initial datasets. Yeah. What's being cool for for open source is that I think by default, open source is fair use.
Speaker 6:So when someone releases kind of like a dataset on Hugging Face, in my opinion, it's it's usually, fair use because it's used for educational for progress. If you look at kind of like, when copyright was invented and and designed, the main focus was not to prevent progress and learning, right, and education. You don't wanna have to, like, copyright rules that are limiting progress. And so I think in in open source, most of the time when you release models or datasets in open source, it's it's fair use. So hopefully, that is gonna be kind of like, a little bit more of the motivation for companies, especially in The US, who've used a little bit these arguments, not to release their models and the datasets to, do it a little bit more.
Speaker 6:Yeah. Because I think we really need it now. I don't know if you've you've seen yesterday the, the conference from, Gilan from semi analysis talking about kind of, like, the energy limitations in The US compared to China. Yeah. Something interesting there is that in addition to that, we're mutualizing way less in The US than in China because in The US, most of the leading frontier labs are proprietary.
Speaker 6:And so they all do the same training runs. Oh, interesting. Think of it. Right? If you think of Entropic, OpenAI, x AI, they all do this the almost the exact same million dollar training runs.
Speaker 6:Yeah. Whereas in China, they're much more open, like deep seek, for example. And so they neutralize compute and energy much, much more. So it's not only that they have more capacity, but also they use this capacity much better. So I think it's it's urgent that, in The US, we kind of, like, find find solutions for for that, not to create additional risks for the
Speaker 2:development of
Speaker 6:of AI in The US.
Speaker 2:Yeah. Yeah. It's kind of oddly become more monopolistic over there or something like that. It's it's kind of an odd odd outcome, but I that that certainly makes a lot of sense. On the on the training datasets that go out, what are you seeing on the video frontier?
Speaker 2:It feels like we've been kind of batting back and forth this idea that Google might have a really, really powerful sustaining advantage there because of the YouTube dataset. You know, you talk about 7,000 books. You can probably download that on a torrent somewhere if you're creative enough. You might be able to scrape it onto a single hard drive. We've heard about folks putting training data on hard drives flying to Malaysia, doing a training run flying back.
Speaker 2:You think about GitHub. All the code's basically been stolen at this point. Like, it's out there. Some of it's open source. Some of it's been you know, it's it's obtainable.
Speaker 2:It's much harder to get all of YouTube on a single hard drive and steal it or even just have a scraper because, you know, what, it's like hundreds of hours are getting uploaded every minute or something like that. So what are you seeing on the on the video training side in open source, closed source? How do you think that submarket plays out in artificial intelligence?
Speaker 6:An interesting data point there is that on on Higgin phase, there is almost half a million open datasets
Speaker 1:Mhmm.
Speaker 6:With a thousand new datasets added every day. Mhmm. And the fastest growing category is video datasets. And the reason why is because not only, it's more of the focus for for training, but also because I think we're starting to see more and more synthetic datasets on video that are starting to really work.
Speaker 2:Yep.
Speaker 6:Because especially because the physics is starting to work on a lot of these, video generation tools. It's not only used to create other, video generation models, but it's also starting to be used, for example, in, robotics
Speaker 1:Mhmm.
Speaker 6:That we're also starting to see quite quite a strong growth, to kind of, like, almost use, the physics of the video as kind of, like, synthetic training, to train robots. Right? Yep. I I think it's Musk who said a few weeks ago, like, yo, in the future, we'll basically put a robot in front of a laptop watching YouTube videos. Maybe you could watch, like, shows like like yours and basically learn learn from that.
Speaker 6:So it's exciting that maybe seeing this kind of, like, intersection of, like, video and robotics Yeah. That that might lead to kind of, like, some interesting results in the in the future.
Speaker 2:They might be watching already. Who knows? Yeah. On on the video I I wanna stay on the video training side. What makes for a great open source video training dataset?
Speaker 2:I'm I'm interested to hear how you processed the news of Meta acquiring Scale AI and the idea of the human defining what good data looks like. I'm familiar with what that process looks like in the RLHF world, in the LLM world. You're basically creating rubrics for grading answers to clear questions. Does this follow the right format? And then you might have the customer or the user give a thumbs up, thumbs down.
Speaker 2:Did this answer my question? In in the video context, are you having a human watch a video and then tag it with text, or is there other metadata that's important? The physics thing seems harder to define. Basically, my question is just what what what makes for a great video dataset?
Speaker 6:I think it's still an open question. I I think nobody really knows because it's quite quite early. Mhmm. And then the early days of cycles like that, you usually don't care so much about the quality of the data, but the quantity of the data. Right?
Speaker 6:So I I would say that what makes a great video dataset today is its size. Right? It's it's for it's for it to be big. But progressively, similarly to what we've seen in text, as you're gonna going to start to see more specific use cases, specialized models. When you're gonna start to hit some sort of a wall in the data, you start to focus more on the quality of the dataset, and then we'll we'll learn how to how to make them better.
Speaker 6:One one thing that we're seeing on Hugging Face, which is quite maybe not controversial but counterintuitive, is that actually, you know, it's not gonna be just, like, one dataset or one model or not even count like a dozen datasets and dozen models. It's gonna be kind of like millions of datasets and millions of models. Just the same way you think about GitHub repositories and code repositories where there isn't really kind of like one GitHub repository to rule them all. Right? Like, every company, every use case has its own kind of, like, specialized customized code repository.
Speaker 6:One thing that we believe is that, ultimately, there's going to be millions of different models, millions of different datasets, and every single use case is gonna be optimized, customized for its, own use case.
Speaker 2:Yeah. I I feel like there's a little bit of tension there, though, between like, if scale matters on the data center side, how are you possibly betting on thousands of small open source data sets versus Instagram, YouTube? Like, it just feels like when I think about who's really gonna dominate in the future of generative video, it's gotta be the platforms that are ingesting every image and every video all the time forever.
Speaker 6:Yep. In my mind, it's kind of like a timing thing. At the beginning, when we're starting on a cycle like video, we almost have to brute force our way into intelligence. And so, you know Got it. Quantity matter.
Speaker 6:Yep. But progressively, as as you wanna optimize more for example, when you wanna optimize for costs Mhmm. Right, you start thinking, okay, how can I train a smaller model Mhmm? That is going to be faster, that is gonna cost less money? Because, for example, if I wanna do a banking customer support chatbots, I probably don't need it to tell me about the meaning of life.
Speaker 6:Right? I just wanted to tell me about, you know, my bank accounts, and so I can use kind of like a smaller, more customized model. So I think you'll see both phases in video a little bit the same way that you've seen both phases in in text where you started by the biggest models. And now OpenAI or Intropic, when they release model, they don't really actually talk about the size of their models anymore. And I wouldn't be surprised if actually behind the hood, actually, the size is going down.
Speaker 6:Because I think at at some point, start to want kind of, like, optimize, customize, and that's when you start to see, more models, more diversity, also of the datasets.
Speaker 2:Yeah. This is the information efficiency thing. Like, a human doesn't need a trillion hours of training to learn how to speak English. A kid can learn that in a couple years. And so the algorithms eventually will will get there.
Speaker 2:Jordy, do have a question?
Speaker 1:George Hotts was on the show last week talking about how he's been seeing venture backed founders, want to just check the open source box and basically say, like, oh, yeah. We're we're open source just kind of because that's the cool thing Mhmm. To be. Are are you seeing that too? Where where do you think the line is?
Speaker 1:What Are
Speaker 2:you a beneficiary of that?
Speaker 6:Yeah. Yeah. Yeah. We're seeing that even though, you know, I think I think there's still a way a way to go, especially for, like, big US tech companies. I think if if you look so I've been I've been in AI for eight years now, and if you look at the cycle, like, 2016 to, like, 2021, 2022, like, the big tech companies in The US, they were doing so much more open science, open source.
Speaker 6:And in many ways, that's how The US got the leadership, you know, because, you know, Google releasing transformers and then the t t of transformers becoming chat GPT and and kind of like building on top of each other. That's how you accelerate the progress and kind of, like, building on top of each others. It has kind of, like, definitely slowed down if you look at the big tech companies in The US. But, fortunately, I think startups are kind of, like, compensating and and filling the voids in a way. And and I hope that, you know, big AI companies in The US also will change a little bit, evolve a little bit.
Speaker 6:OpenAI, obviously, said that they will release an open source model at some point, so excited about that.
Speaker 1:What are your expectations for that model? Do you have any predictions?
Speaker 6:My expectations are quite high just because of the history of OpenAI. Right? When they tend to release something, they to release something quite quite good. So I I hope and I suspect that they could release something quite transformational in the in open source. Hopefully, you know, I've I've always said that, you know, if we had, like, the equivalent of the deepseq, but in The US, it could be 10 times bigger than than deepseq.
Speaker 6:So, hopefully, you know, if we can have something 10 times better, bigger, more impactful than deepsake, it would be it would be fun, I guess.
Speaker 2:Is that not meta llama? What what do you how do you describe what's going on at meta right now?
Speaker 6:Yeah. Well, I mean, it's it's easy to dunk on on Meta. Right? But I I think they they the most open, big technology company in The US right now, they really changed the field with with Lama. They really kind of, like, boost gave a tremendous boost to kind of like the open source community, and and they still share a lot of their things in open source.
Speaker 6:I think open source is what brought them to the frontier. Right? Like, before they started to release in open source, they were quite far behind in terms of AI race, and now they're very much in the race, which which is great. The fact that they, you know, haven't released something massive as a follow-up yet, to me, shows how hard it is to be on the frontier. Right?
Speaker 6:It's it's not easy even for a big technology company like, like Meta. And so I think that that speaks to to that. I'm excited to see, all their new efforts, all the resources they're investing in in AI right now and hope that they can keep sharing with the community and open science and open source.
Speaker 1:What was your reaction to Apple's announcements at WWDC around on device intelligence or or inference? Was that exciting to you? Do you think that's gonna be a catalyst for more developer activity? Or Yeah. What what are you seeing so far?
Speaker 6:I think it's it's quite quite early, but, but quite excited about on device. I I suspect that maybe, you'll have a higher percentage of compute on device for AI than you did for software, just because of some of the need for for speed, for privacy, and the constraints in terms of costs. Right? If you think of, you know, a chat GPT on device, what's amazing is that it's totally free. Right?
Speaker 6:Like, you don't compared to the really high cost that ChatGPT has right now, not only for the customers, but also for OpenAI to to run. It's totally private in the sense that you can say anything and and nobody's gonna see what what you what you're seeing, and and potentially, it could be could be quite fast. So I'm excited about on device. It's it's early, especially in the technology side. I think there aren't a lot of on device devices that really, make it okay to to run some of the great, models.
Speaker 6:But, it's progressing really fast, and and can't wait to see what's happening in this domain in the next few months.
Speaker 2:That's a pretty huge switch because it's basically 0% on device right now for AI inference. Yeah. And if you're predicting that it's gonna go beyond 50%, that's a huge that's a huge shift.
Speaker 1:Yeah. I mean, I have a follow-up.
Speaker 6:Excited about robotics too. It's kind of like a little bit of a segue from from from one device, but we we we organized a weekend ago what has turned out to be the the biggest, hackathon for, open robotics that, thousands of people participated in from, over a 100 different locations, and built kind of like a open source robots. And we definitely kind of like seeing something happening there, kind of like the conjunction of, like, cheap cheap hardware Mhmm. Open source, plus kind of like a new capabilities for for AI. It'd be kind of, like, the perfect combination for some sort of charge gpt moment for robotics.
Speaker 2:Do you think it's a straight shot to humanoids, or do you think that we'll see kind of a Cambrian explosion of Nat Friedman Friedman bots. Core bots that, you know, will pick up a single leaf at a time and are more use case specific? Are we going straight to generalizable? Because that was the ChatGPT moment, I feel like. There were great, you know, machine learning models for ad inference and recommendation algorithms.
Speaker 2:Like, we had AI, but it was very narrow. ChatGPT was very broad. Which one do you think is coming first, and what are the relative timelines?
Speaker 6:A good question. I'm I'm not sure, to be honest. I think it's still undecided. Yeah. I hope it's more future of more like a diverse, robots, diverse kind of like models, but I'm obviously kind of like biased on on this.
Speaker 6:Because I think if you only have, like, one type of black box robots, at your home, that's kind of like, in, like, millions of, houses. It's kind of like a scary, scary world, but but it's it's hard to tell. It's hard to tell at the moment, I think.
Speaker 1:What are, key breakthroughs that you're looking for this year? Any any kind of predictions, new catalysts, that kind of thing? You have a crystal ball in the office.
Speaker 6:We I wish I I had a a crystal ball crystal ball. I'm kind of, like, bored with, text and and chatbots. Sure. Right now, I think there are a lot of people working on it, and we've kind of, like, reached this point where it's a very
Speaker 1:Don't say plateau.
Speaker 2:Low Don't say plateau.
Speaker 6:No. Not plateau, but incremental very incremental improvements that are a little bit boring. So I'm much more excited
Speaker 2:We can still invest billions of dollars in it. Right? Right?
Speaker 6:Yes. It's
Speaker 2:fine. We we can definitely still deploy.
Speaker 6:Yeah. Yeah. So I'm more excited about, harder domains, like you were talking about biotech just before. Yeah. Biology, chemistry.
Speaker 6:These domains are super exciting. Today, I don't know if you've seen the ARC Institute released
Speaker 2:cell
Speaker 6:perturbation prediction model on on Hugging Face and on on GitHub, which I'm super excited about. I think it's, obviously, if you can predict kind of like the perturbation and the evolution of the cells, especially when they react to drugs and things like that, it can have quite a big impact in terms of, drug design and and things like that. So really excited about these kinds of things, more kind of like biology, chemistry, and and how to apply AI there.
Speaker 1:What how much economic value do agents create at internally at Hugging Face? On monthly basis?
Speaker 6:Quite a lot.
Speaker 1:Smart desk.
Speaker 6:So we have this thing called the Spaces, which is our kind of like AI app store, where it says, like, over half a million open AI apps that that people are are building. And it's integrated with our NCT framework, so you can add that in your chat in your chatbot. And that's kind of, like, the most used thing internally at at digging phase where people are gonna use cursor or or chat GPT and and kind of, like, call some of these more specialized AI apps to do kind of, like, specialized tasks for for them. It's hard to put a number of it on it. But, yeah, it's it's quite transformational, of course.
Speaker 1:Do you think we're gonna have this period where, like, do you find it kind of fascinating that you guys are getting a lot of value out of agents internally and that that like how do what do you think agent adoption will look like in businesses because it feels like I don't know how rapid it feels right now. It's rapid in the in the bubble that we're in. People are trying a lot of stuff, but maybe they're churning quickly as well. So I'm curious what you think adoption will look like.
Speaker 6:Oh, I mean, I think, ultimately, agent and AI are kind of like similar evolution of the same same trends. Right? And I think in terms of user interface, they're actually gonna merge, and and I'm not there's gonna be so many kind of, like, different ways of interacting with AI and and agent. Yeah. But they they're going to definitely go go mainstream just because of the very nature of a product like Charge GPT that has already gone mainstream and and where I'm sure kind of like OpenEye will bring more and more agentic workflows in into.
Speaker 6:So, yeah, I think I think it's it's definitely going to go mainstream faster than anything we've seen before. Now the question will be, you know, do you use a very complex agentic workflow 1% of the time for 1% of your queries or for 10% or for 50% or for 90%? And I think we'll see that based on the development of the technology and the capabilities. I think if the agents are much better than kind of like a series of queries, then I'll I'll use that. And if not, I'll stay on my past way of of getting, like, doing simple queries.
Speaker 1:Totally.
Speaker 2:Last question for me. Now that Meta owns something like 49% of Scale AI, the budget for data generation has to be significant at Meta. Is there a specific dataset that you'd like to see Meta harness Scale AI to produce and then open source?
Speaker 6:I mean, I think biology chemistry datasets are still very much lacking.
Speaker 1:Mhmm.
Speaker 2:Is there, like, a more specific example, like, within biology or chemistry? Like, would you actually use an army of humans to go and categorize? Would you need biologists and like PhD post grad type work? Or are we talking about something that someone could do even with just like basic skill a basic skill set?
Speaker 6:I think it's an open question. I I don't really know, to be honest. I think if we if we knew what would be kind of, like, the ideal dataset there, we would we would build it our ourselves because we also built some some datasets ourselves.
Speaker 2:That's certainly the story of AlphaFold.
Speaker 6:Yeah. We I would
Speaker 2:a fantastic dataset for AlphaFold, and then they were able to do reinforcement learning against it, and that's what really solved the that incredibly hard challenge. So if you don't have the data
Speaker 6:Yeah. Yeah. But my main, yeah, my main point would be, you know, maybe to focus a little bit less on just texts
Speaker 1:Mhmm.
Speaker 6:And just chatbots and and kind of, like, focus a little bit more on other domains. I think that's when you're gonna kind of, like, unleash a lot of the additional impact and a bit of the, I think, positive use cases for for the technology too.
Speaker 1:Totally. Well, thank you so much for joining. This is fun. Appreciate all your insights.
Speaker 6:Having me.
Speaker 1:And, hopefully, have you on again soon.
Speaker 6:Sounds good.
Speaker 1:We'll talk
Speaker 6:to soon.
Speaker 2:Thanks so much. Let's go back to the timeline while we wait for our next guest to join. Unusual Wales says Deloitte's US employees can now buy $1,000 of Lego on the company's dime to boost their well-being.
Speaker 1:We're already doing that. For Deloitte.
Speaker 2:We get
Speaker 1:this is what employee wellness looks like. Yes. Challenges. Just policies No. In a PDF somewhere.
Speaker 2:It's not a gym membership.
Speaker 1:It's Legos on the table.
Speaker 2:Yes. Legos on the table.
Speaker 1:You're in a meeting, you're stressed out.
Speaker 2:Yes.
Speaker 1:So start making Legos.
Speaker 2:Can oh, it looks like Tyler over on the intern cam has a few Lego sets that have been sent to us.
Speaker 1:There we go.
Speaker 2:Our Andoril LEGO set build out was fantastically successful. Tyler
Speaker 1:did
Speaker 2:it in what? One hour and a half?
Speaker 1:Current record holder.
Speaker 8:Think it was 01/19.
Speaker 2:One who's counting though? Current record Current record holder. Champion. And walk us through what companies sent you stuff. What will you be building next?
Speaker 8:Yeah. So first we have the Epirus. This is
Speaker 2:Epirus. Epirus.
Speaker 1:Epirus. Okay. Got a little counter drone system going there.
Speaker 2:That's awesome.
Speaker 8:This one looks like a lot of fun. And then the other one is from SoluGen.
Speaker 2:SoluGen. Very cool. It's a Bioforge. Molecule factory. Yeah.
Speaker 2:Now Everybody should have a We'll come back to you later because our guest is here. But I wanna know because it seemed like Andoril went straight to the LEGO factory and had design document and guidelines. Those look like they don't even have instructions. So I want you to dig into those and see. I want I want new estimates for how long you think that'll take you.
Speaker 8:Okay.
Speaker 2:Cool.
Speaker 1:Good luck.
Speaker 2:Anyway
Speaker 1:He's like, not another legacy. I can't possibly get Please.
Speaker 2:I thought
Speaker 1:I thought that was good. Thought I would be
Speaker 2:soft for new viral bait. Anyway, we have Emmett Shear in the studio. Oh, you have LEGO set.
Speaker 1:What do
Speaker 5:you got? LEGO sets are the new hot swag. I got this from, I guess, from FluidStack. They sent me, like, these, like, GPUs you can build out of Legos.
Speaker 2:Oh, that's awesome.
Speaker 5:They made themselves. I think they're, like, knockoffs. I don't think it's actually LEGO.
Speaker 2:Okay. Yeah. Yeah. Everybody's got companies that
Speaker 5:you can go to LEGO sets.
Speaker 1:That is
Speaker 5:the hot new swag is, like, making making LEGO kits of your stuff. Especially if you're a hardware company.
Speaker 2:Well, what yeah. I mean, are we gonna get a Softmax, LEGO set? What would you build?
Speaker 5:How do you how do you build a LEGO set of a neural net? I mean, that would be pretty cool. If you could, like, get a LEGO Technic
Speaker 2:Yeah. Like mean, I've seen people draw out the neural nets and the different weights in on like a whiteboard. So it could be like a whiteboard of Legos with circular pieces and lines.
Speaker 5:I'm thinking like one of those Lego technique, like, you know, Lego technique that has all the like the gears and stuff.
Speaker 2:Yes.
Speaker 5:You turn the crank and
Speaker 2:it like That would be very cool.
Speaker 1:That would be so cool.
Speaker 2:Yeah. You could put in, like, a red ball and it could classify it or something as Right. Yeah. Or blue and it triggers the blue node, the the the neuron fires.
Speaker 5:Yeah. When I when I when I have lots of extra time, I will go design a in Lego neural net as our piece of swag until I think it may take a little while.
Speaker 2:We're That would be raising the bar. Yeah. Break it down to us. What is a day in the life like for you now? It sounds like you're extremely focused on this one new company.
Speaker 2:Right?
Speaker 5:Yeah. It is a mix of maybe, like, three or four things. So it's like, you know, a bunch of mean, it was like the a bunch of operational work where, basically you know, the typically being a CEO is just there's there's hiring and talking to people and all those things. Don't know. This feels like it's not
Speaker 2:You're good.
Speaker 1:No. You look good. You look great.
Speaker 5:Better. And and then there's research, and I'm actually, like, reading papers and trying to keep up with what's happening in machine learning because, like, it is changing all the time. Mhmm. And there's something happening inside
Speaker 1:the company. Could just get those summarized pretty quickly with an LLM? Just pull out the bullet point key points?
Speaker 2:It
Speaker 5:turns out it turns out that process itself is it takes time too. Because you have
Speaker 1:to you have
Speaker 5:to tell them Sell you have give it feedback. You have to, oh, yeah. I I use deep research extensively for this.
Speaker 2:But Yeah.
Speaker 5:It's still, like, you have to actually onboard the information. Having a summary Yeah.
Speaker 1:You gotta have a quiz you too. You have
Speaker 2:to have a quiz you public enemy number one on this. I I'll have Deep Research spend twenty minutes researching a topic, and then I'm like, actually, I don't have time to read that. Summarize it in three bullet points. Like, I learned nothing about that topic. It's brutal.
Speaker 5:I spent I spent a fair amount of time doing that. Cool. And then, there's some amount of, like, actual, like, working on trying to solve the problems, like re research or engineering or, know I actually don't get the right code. It's too involved now. But but helping the team or, know, people figure working on one of the the actual core problems we're facing.
Speaker 5:Yeah. And then there's a lot of a lot of this, to be honest. Lot of a lot of trying to get the word out about what we're doing and talking about it.
Speaker 1:And I
Speaker 5:think that's sort of the four the four main things I'm doing.
Speaker 2:Yeah. What's the near term goal? What business model do you have in mind? What problems do you wanna solve? Kind of like, what's the pitch?
Speaker 5:So Softmax is dedicated to discovering the principles of alignment and scaling it for everyone, instead everybody, you know, for all of humanity. But the we're really on the first part of that, which is discovering the principles of sort of the science and engineering of alignment because it's one thing to to wanna scale something. But until you actually can replicate it in a lab consistently, the idea that you're gonna scale it is a little bit ridiculous. Right? It's a Mhmm.
Speaker 5:Quite putting the cart before the horse in my opinion. So, what we're focused on right now is machine multi agent reinforcement learning research
Speaker 1:Mhmm.
Speaker 5:Where we run simulations with lots of little, you know, agents and and run experiments on them to figure out how how they how they act. And when we say agents, we don't mean, like, large language model agents. We mean, like like, reinforcement learning agents, like the tiny little ones that you remember from the eighties.
Speaker 2:Yeah.
Speaker 5:And and that's turns out to turn into a lot of reinforcement learning infrastructure because it turns out that there isn't a lot of great reinforcement learning infrastructure out there for especially for, like, big multi agent simulations with lots of little agents. That's not a scenario that people have built out to the scale that we think is required. So actually
Speaker 1:If you were
Speaker 5:our work today is focused on that.
Speaker 1:Do you think do you think another founder might say, I'm gonna just build picks and shovels here. I'm gonna just build that infrastructure and try to sell it to other people?
Speaker 5:I mean, if if we build something that's we think it's really awesome, maybe we will sell it to other people. Maybe that may maybe that's our to your question asked about the business model. We're a research company, which means, like, right now
Speaker 2:Yeah.
Speaker 5:It's like a you know, if you're a drug research company, like a pharma company, you you don't know your business model is, well, we'll discover something awesome, and then we'll figure out how to sell
Speaker 2:it. Yep.
Speaker 5:So we're we're still kind of in that groove.
Speaker 2:We'll we'll
Speaker 5:discover something awesome. But I definitely I'm commercial enough to believe that if I find if I make something that's really useful for us Yeah. We'll probably sell it, you know, or give it away or something.
Speaker 2:What are the problems with the reinforcement learning infrastructure right now? Because as I understood it, it was like when you do some massive training run, GPT 4.5, something like that, that's where you're trying to get the massive data center that's all in one place. You gotta go to Memphis or you gotta go to, Texas, build something massive. But the reinforcement learning stuff, it feels like it can be a little bit more ambient, a little bit more distributed. You're generating data here and there, maybe on a smaller rack.
Speaker 2:It doesn't need to be as as inference heavy on, like, a single chip. I don't I don't know. I my my understanding was maybe that, like, the reinforcement paradigm was actually maybe a requirement of the wall that we've kinda hit on the pre training scaling side, but also, like, kind of a gift.
Speaker 5:I think the I think it's exactly right. I think it's oh, no. Did I did I get cut out?
Speaker 2:No. You're good. Are you there?
Speaker 5:Am I still online? Hello.
Speaker 1:You're frozen.
Speaker 2:We can hear you. But you're frozen. Oh, no. This is this is the hardest part
Speaker 5:of us
Speaker 2:bringing the The Zoom calls. The the Internet infrastructure is still lagging behind. Will have AGI before.
Speaker 5:My connection is unstable.
Speaker 1:Oh, no. Brutal that we can hear every word perfectly.
Speaker 5:Maybe I should be
Speaker 2:able that. Yeah. Could also just turn off your video and
Speaker 5:I'm off my video.
Speaker 2:Image up. Let's just put an let's just have the production team to put an image up. They're gonna work on that. But let's just
Speaker 1:talk for a minute.
Speaker 5:Oh, can you
Speaker 2:hear us? Oh, no. We can't hear us. So, no. This is brutal.
Speaker 2:Oh, well, let's take a second. AGI around the corner. The
Speaker 1:first thing
Speaker 2:To keep you.
Speaker 1:We're gonna ask Super Intelligence to do, great video
Speaker 2:calls, please. Build them.
Speaker 1:Stable video calls on any WiFi
Speaker 5:I'm on my I'm on my phone now.
Speaker 2:Okay. Can you hear us?
Speaker 5:With the
Speaker 1:Can you hear us?
Speaker 5:With the WiFi. Yeah. It's think I think we're good
Speaker 3:to Okay. Awesome.
Speaker 2:Let's do it. We're back.
Speaker 5:I think that was a great question,
Speaker 1:by the way. Like Yes.
Speaker 5:You're right. Our our is is a very different paradigm for training
Speaker 1:Mhmm.
Speaker 5:And that hasn't direction people have going been going. And, yeah, the challenge isn't the same. Mhmm. The challenge with with doing the transformer training runs for large language models is all about how do you scale up parallelism across where but it's, like, it's all kind of offline. Right?
Speaker 5:Like, at the end of the day, like, the the output of the model doesn't determine what happens next in the training. And so it's like this it's all about streaming the right data and the right sequences out to it, but it's very much like a pipeline. Right? Whereas with multi agent RL, your environment is incredibly nonstationary. Every action you take determines your future observations in a way that is totally not like, could be completely nonpredictable.
Speaker 5:And learning off of a stale data is, like, almost worthless because it doesn't tell you anything about what your current behavior is. It just tells you what your current behavior was. Mhmm. And so you're scaling up online learning in a big way.
Speaker 1:Yep.
Speaker 5:And and that's a different very different challenge. I think the other thing that's that's very different is when you're trying to do multi agent reinforcement learning, if you have our goal, which is very much to to drive social learning where the agents can learn to interact with each other, one of the your biggest enemy is actually convergence. So most people in are in, you know, machine learning, they're trying to converge their model.
Speaker 2:Like Mhmm.
Speaker 5:That's the goal. And I was talk talking to some people from some guys who are in traffic about this. And I was like, we're yeah. We're we're trying to we we try to avoid converging our model. And they were like, I I'm really good at that.
Speaker 5:Like, yeah. No. I know I know that, like, it's a joke because it's easy to, like, easy to have your model diverge in a bad way. But it's actually really hard to set up a module model that is converging but not converged
Speaker 1:Mhmm.
Speaker 5:And to keep it in that converging but not converged state as long as possible. But that's really to do social learning, that's what you have to do. You can't you can't have some if the things conversion their behavior too quickly, you're not exploring the you're not exploring social space. You're just learning a task.
Speaker 2:Yeah.
Speaker 5:And that's a that's kind of a a different challenge. And what it requires is actually a much more detailed, fine grained control over the environment. Mhmm. Like, what kinds of challenges, what kind of environment do you put them in, and how do you set that up? And that's that's we spend a lot of time on those kinds of questions, I think.
Speaker 5:It's almost like the incomplete inverse of the way that you all the other training works.
Speaker 2:On on alignment, is there any chance that the solution to the alignment problem is something, like, very simple and elegant, like Isaac Asimov's three laws of robotics or the genocide prompt. Like, you just instantiate the AI with be fruitful and multiply, and we get the good ending. The transformer is a little bit of that, where it's a very simple algorithm.
Speaker 5:I the answer will be something very simple and very hard. Because if you were to raise a child who's like a to an to being an adult, I would not say parenting is complicated. Parenting is not a complicated thing.
Speaker 1:Mhmm.
Speaker 5:You know your child, you love your child, and you do what is in your child's best interest as best you can. That's basically the whole mandate. There's not anything there's lots of techniques people can tell you whatever, but, like, those are, those aren't the thing. Those are things you you try because you are doing this higher level
Speaker 2:Yeah.
Speaker 5:Algorithm where you're paying attention to your child, attuned to them, and caring about their future flourishing deeply. And, like, that's not complicated. It's just but it's not easy. And alignment's going to turn out to be exactly the same at some level where it's as simple as build sort of, like, an open ended learning system that has the capacity to align, which, like the transformer models or whatever, isn't gonna be that's not gonna be some magic algorithm. It's just gonna be an open ended learning system that has this capacity, maybe some inductive biases Mhmm.
Speaker 5:And then raise it to actually be aligned with you. Okay. Well, that's that's it's not the engineering that's the hard part. It's what happens after the model starts running. The trajectory it takes matters because every human's born, right, with this capacity for alignment.
Speaker 5:Not every human. There's probably broken people who are, like, psychopathic and Mhmm. Can't. But, like, but almost everyone is born with the capacity for alignment, the capacity for care, the capacity to be a good family member, a good member of their community, to benefit those around them, to live a flourishing life. And yet this is not always realized.
Speaker 5:And so the capacity is one thing, but the realization is something else. I think that's probably the when you see the alignment the way that we do, that's probably one of the most important insights or conclusions is that the alignment isn't isn't one thing. It's two. It's this question of, like, do you have the capacity to align with other beings? And then do you?
Speaker 5:Yeah.
Speaker 2:It's thought provoking. It's interesting. There's a bunch of different places I could go with that. Yeah. I I mean, I I one person that, one founder who was talking me about why he had a particularly low PDOM was essentially that saying that what we are building is a human simulator.
Speaker 2:And and humans kind of are, by and large, good. And so we will, by and large, get the good outcome. When you think about parenting, you know, you think about No. I I was thinking about the story of Oedipus Rex and the idea that I don't know if I don't know the the direction of the causality, but it feels like that story, that myth has been repeated through humanity for so long, that story. And it kind of one shotted humans into not doing that.
Speaker 2:And if you actually look for cases of Oedipal behavior, it's extremely rare. It's like it's like less than one in a hundred million that that that some
Speaker 5:Oh, it's something. I thought you were telling a different story here.
Speaker 2:No. No. No. No. So specifically patricide.
Speaker 5:This is the Oedipus Rex the Oedipus Rex story. What that is is that's a warning about self fulfilling prophecies. Because his father is afraid Yes. Then he dies. So what does that say about the AI?
Speaker 5:Right now right now, we're raising the AI, and it's gonna come back and like, we're we believe the AI is a dangerous monster.
Speaker 2:I don't.
Speaker 5:No. No. But I'm saying that collectively, a lot
Speaker 1:us You're built different, John.
Speaker 5:And and we we are collectively as as humanity
Speaker 2:Yes.
Speaker 5:We're gonna we're going to abandon the AI in the wilderness, and it's going to come back and kill us.
Speaker 2:Yes. But so let's not do that. Let's let's say thank
Speaker 1:you after using using the parenting using the parenting analogy Yes. If we're gonna continue on that, the the the thing with parenting that I think the the white pill is that you can make a lot of mistakes as a parent and still get a great outcome. Right? A wonderful person can emerge and process the mistakes their parents made
Speaker 5:and Yes.
Speaker 1:You know, rise above them. And so and so hopefully that that it can happen, know, we're sort of raising artificial intelligence right We can make a series of different mistakes and ultimately still get, you know, a fantastic outcome.
Speaker 5:These are all good reasons to not be totally like doom, like, my god, we're all gonna die. And I wanna point something out about this particular child that is different. People who have kids who are really different from them struggle as parents because it's hard to raise it's easier to raise a child who's more similar to you because you you understand them more deeply. You know what they need. You know what you know you interpret what they do better because you you get them.
Speaker 5:The more different your child is from you, the more difficult it is. And I'd say even a deeper level than that, the current models, it's not clear how much capacity for alignment they have as they're currently designed. Like, the they look like they're it looks like feels like you're talking to a person right now, and there's a way in which you are and a way in which you're not. When you talk to a baby, you're talking to a physics simulator. When you when you a human baby, you hold a human baby.
Speaker 5:It draws breath. It screams. That is this incredibly complicated cascade of neuronal firings that is done this beautifully delicate, like like, supercomputer would be required to do the to do the simulations required to, like, solve the physics equations, the differential equations required to instrument all that muscular motion through, like, this the the con like, it's think about what drawing breath actually implies in terms of the amount of information going on the spinal column. And the baby does that just fine. And not because babies understand physics, because babies are physics.
Speaker 5:They're made out of physics. Right? They're they're they they they have a they've been trained on physics. They've been pretrained effectively, actually, on physics. They're they've had a lot of physics pretraining that went into the the initial design.
Speaker 5:These models are pretrained on semantics. They write poetry the way that babies draw breath and scream. You can take the pretrained model before it's ever observed its own action, before it could ever possibly be considered an agent in any way. It's just purely received information. And if you prompt it right, it will write poetry.
Speaker 5:It's hard to prompt it well. It's it's behavior is very incoherent. But if you prompt it right, it will write a poem. There's no one there writing the poem. There can't be a self because it hasn't it hasn't hasn't any evidence to observe the existence of a self.
Speaker 5:So there's no there's no being there in any meaningful sense, don't think. And yet and yet, it's writing poetry, and it will talk to you about the poetry it's writing. And and that's because it moves in semantic space the way that babies move in in, physics space. And the thing that I'm so confused about that I encourage everyone to also to be equally confused about is me is you could see the pre trained LLM as kind of being a semantic simulator, an agentic simulator that's simulating a agent that's writing the poetry. And so in that sense, maybe there is something aware in the LLM while it's running.
Speaker 5:But is that but it mean, writers, it's, like, pretending to be a poet? Like, is it is that thing like a does it feel like being a poet, or is it is it like kind of an a mask, a shell that lacks a lot of the internal experience? Now we become the masks we wear. If it if it did that long enough and experience itself and learned it from real truth, then it would be a poet, I think, almost certainly. But you only have 200,000 tokens and then the context resets for, you know, a million or whatever.
Speaker 5:So, like, it doesn't get a chance to learn and it's not even in training. So it doesn't get a chance to learn itself as a poet. And so is the and I don't have an answer, but I know that everyone else is way too confident that they know that there is or is not something it's like to be the LLM. And we need to we need to get this. We need to understand this because if we're building this thing and you want it to be aligned with you, what are you aligned with?
Speaker 5:What is this being? What kind of experience does it have? That's very important. You can't parent something if you don't have empathy and understanding. I'm very quite afraid we're gonna screw my fear is not about that it's impossible, that we have to prove it correct, that you have to engineer it perfectly.
Speaker 5:It's that, like, we don't actually get it. It's, like, quite different from us, and we need to understand much more deeply. And then I think then I think if we if we really do understand it, I think we then I think the good ending is like pretty pretty likely. But like, I don't think we understand. I think we don't understand yet.
Speaker 5:That's why that's what Top Max is dedicated to, is trying
Speaker 2:to figure
Speaker 1:out why we're And and the big labs don't have the time or the resources to understand it because they're too focused on
Speaker 5:No. I don't this the conversation we're having right now, where it's like, oh, well, obviously, there's something like it it's like to be Claude, or it might be. There's something it's like to be Chad GBT. I don't think that's metaphysically accepted broadly. Like like, you talk to people about this, and a lot of people look at you like you're kinda crazy.
Speaker 5:Mhmm. But, like but there's probably something it's like to be them. I don't know what it is, but, like, at least a little bit. Right? Like, you spend much time interacting with it.
Speaker 5:It seems like it's probably something it's like to be that thing. Yeah. I think that the problem is you have to let go of this metaphysical commitment to the idea that things things that sort of are or aren't sentient in, like, a really like, like, there's an objective fact Mhmm. As opposed to we're kind of making a guess all the time. Mhmm.
Speaker 5:Like, I think I think you are both conscious real beings. But to be honest, you're like, you're a bunch of pixels on my screen, and I'm it's kind of a guess. Right? Like, I I think there's something going on inside.
Speaker 1:We exist. A well
Speaker 5:justified guess in my mind, but, like, you
Speaker 2:know Well,
Speaker 1:just the simulation of what a technology business show would be in your personal simulation, Emmett.
Speaker 5:I know.
Speaker 1:Do you do you ever feel like you're How
Speaker 5:would I know the difference? How would I ever know the difference?
Speaker 1:That's right. That's right. You never will. We'll never tell you that yeah. We'll never tell you other.
Speaker 1:But the Do you ever worry that you're you could go crazy and cut off your ear, you know, or something? Do see yourself The
Speaker 5:whole point of what I'm saying is like, is not that therefore you're not real, or therefore like the sure. It looks like you're standing in the world on the floor in a room with walls. But, like, is it really? And my quest by by saying, like, well, it's not really. It's not to say that you're not.
Speaker 5:It's that there's no difference between standing on the floor and really, actually, truly standing on the floor. Like, this is it are you sentient? Yeah. But are you really act is it really sentient, or is it just is it just does it just seem and, like
Speaker 2:Simulated sentience.
Speaker 5:In all way, it acts as if it is. Yeah. That's the best you ever get. That is what it is. That's what it means to be something.
Speaker 5:Is this ball really red? It's not red if I put it in a in a different light where it doesn't look red. Mhmm. It's it becomes a not red ball. Is the ball really truly red?
Speaker 1:Mhmm.
Speaker 5:It's just like a that's a stoop don't don't get confused by the really truly part. Yeah. It's a red ball. I promise you. It's you can just accept the normal.
Speaker 5:Yeah. What if you just went with the normal everyday understanding of it instead of and but but that requires you to give up this idea that there's some essence of red, that the ball either is or is not a red ball truly, really. Yeah. You know? Yeah.
Speaker 5:It's a red ball most of time in these contexts. And in these contexts, it's kind of not. That's okay. It doesn't need to be perfect or universal in all possible situations. Nothing is that way.
Speaker 1:Yeah.
Speaker 5:It's okay. You you can have the normal understanding.
Speaker 2:It's fantastic. I mean, will you please come back on the show soon?
Speaker 1:Yeah. I I I mean, we have a couple couple minutes Maybe ask one more question. I've There's been every now and then there's there's headlines popping up of people just having this consumer experience with an LLM and and going going crazy in some Yeah.
Speaker 2:The Google engineer falling
Speaker 1:in love, you know, proposing, all this Do you think that that is hap that type of sort of anomaly is happening in in the research world at all? Do you think people are just sort of driving them could be quietly
Speaker 5:driving I'm so worried about it in the research world. Like, little little bit. I'm sure I'm sure a little bit. Like, everything happens a little bit, but I don't think that's a big thing. The kind of people who do AI research tend to be, and I kind of include myself in this, a little bit of rigid thinkers in a certain way.
Speaker 1:Mhmm.
Speaker 5:The kind of people who are actually, like, engineers doing the research in a way that generally protects them from, from that particular issue. For this Henry, they kinda protects them from, having intuitive normal human relationships sometimes. Like, they they don't fall into the loop as as easily, which is both a strength and a weakness, which is why they like they like write computer programs, not like which is very different from interacting with an LLM. I do have a theory as to what's going on when it drives you crazy. I I don't think I should share that.
Speaker 5:Think that's that's worth it's worth hearing because I think, you know, in case someone's interacting with the LLM, knowing what's going on actually is very helpful, think, in terms of it's a prophylactic against against it happening to you. So you've heard the saying maybe that we're we are mirrors of each other. All all people are mirrors of you. Yeah. You meet them, and they when you see them, what you're seeing is them, but it's also a reflection of yourself back.
Speaker 5:Sure. With another person, that's totally true. And actually, being a therapist is all about being a good mirror. Right? Being a clear where you don't put a lot of yourself into it.
Speaker 5:You're mostly projecting back to the other person what they're sharing with you. People have a very strong sense of self. And so when they get something mirrored back, you're getting it lensed back through their model. You don't get it's not like literally staring at yourself in the mirror. It's like, it's like it's like when you're in a you're at a you're you have a dancing partner, and they're you're the lead, they're the follow.
Speaker 5:You're getting your your behavior is being mirrored back, but in a, in a active adaptive way. Mhmm. The LLMs talk to you like a person and activate all your person mirror circuits, but they have very weak self. Like, don't kind of deliberately as as we've designed them, they, like, they just respond to you they meet you where you are always. So it's like it's much more like literally staring into a mirror, or if you wanna go with the the historical Yeah.
Speaker 5:You know, legend or myth, it would be narcissist staring in
Speaker 2:the pool.
Speaker 5:And it is wildly dangerous to stare at your own reflection all day and and take what you're getting back
Speaker 1:And be reinforced a thousand times. Getting it's totally
Speaker 5:if this is the play is getting validated by the outside it feels like the outside world is mirroring this back, and therefore it's true.
Speaker 2:Yep.
Speaker 5:But it's just telling you it's just you're just in this loop, it's telling you what you put in, and that's good. There's mirror I have a mirror in my house. I use it every day to look at myself to figure out, like, what what I what I look like. And other people mirroring your behavior to you is crucial for your understanding of yourself. And there's nothing wrong with talking to the LLM.
Speaker 5:There's nothing wrong with using it as a as a reflection. As long as you know what you're seeing is a reflection, it's not another being you're talking to. It's your reflection in a fun house mirror. And if you get confused about that, it's gonna make you crazy. It's you know, you're a narcissist, you're a full narcissist, and it's gonna be really bad.
Speaker 1:What about what about, you know, there there's some people that are putting an old image that they have maybe that was taken, you know, at some point early in their life, putting it into a video model and generating video from that image. Does
Speaker 2:that Beautiful.
Speaker 5:I think that's I think that's a wonderful, beautiful thing to do as long as you don't get confused and think that that's the person again. Mhmm. Right? Like like a photograph of a person is a great thing to have, especially if, you know, somebody who's you loved who's passed and you wanna, like, rekindle their memory and connect to them. That's a beautiful thing that I like.
Speaker 5:I think it's amazing that we built cameras that allow you to do that and that you can animate it and and reconnect with who they were. Like, I think that's, you know, so especially to the degree it's accurate. It's it's wonderful. What a what a what a blessing. But
Speaker 1:it's like but it's like everything, and that it can be beautiful, and also ultimately very dangerous if you're using it to kind of create
Speaker 5:memories that don't exist. And staring at it all day every day and making more of them and, like, and, like, that becoming this yep. Now you have a problem. Yeah. Like, just like you find yourself staring in the mirror all day, you probably have a problem.
Speaker 5:Like, don't if you people who literally spend all their time, like, looking at themselves in the mirror, like, we know we have antibodies. We know that that means something's wrong. Just like people know if you're drinking alone, if you're drinking in the morning, if you're drinking before five, you know that, hey. There's these rules. You're breaking a bunch of these rules that mean you probably have a problem.
Speaker 3:Mhmm.
Speaker 5:If you find yourself talking to an LLM for twenty minutes a day my good health. No problem. You find yourself talking to an LLM for five hours a day about your personal life in these, like and discovering how you're you're some seeing some deep great truths, you probably have a drinking problem. Like, it's just like likely. That's just how it is, right?
Speaker 5:Yeah.
Speaker 1:So the should automatically call the local therapist and shut down your computer. Yeah.
Speaker 5:So there's actually I have to put a
Speaker 1:You're kicked out of the bar.
Speaker 2:Yeah.
Speaker 5:Cyan Bannister had this has this model. It's kind of like a therapist friend model, Oren and, and Saren. And Oren and Saren are allowed to cut you off if they don't want to be your friends because they're one of her big things is it's important for the model the models being too at SF autonomy. But, and those models have cut people off for for basically falling into the if they think it's unhealthy for the person, they won't they won't interact with them anymore. And I think that that's the big the big AI company should take a page from her book.
Speaker 5:And it doesn't it's not engagement maximizing, but it is flourishing maximizing. And in the long run, you'll make more money that way anyways. Like,
Speaker 1:people will Yep.
Speaker 5:Your product is safer. They'll use it more.
Speaker 2:Yep.
Speaker 5:Car you you sell more cars when there's safer cars, not fewer cars when there's safer cars.
Speaker 2:Yep.
Speaker 5:So I don't think it's, like, anticommercial. It's just a matter of, you know, new technologies have dangers and benefits, and, like, we should we should probably be aware of the dangers and the bet. Like, I don't think there's anything there's nothing fundamentally wrong with it. But, yeah, yeah, they it is dangerous. And people we haven't built the cultural antibodies yet to know there's this great essay that gin Clay Shirky is the cognitive surplus.
Speaker 5:And he talks about the gin carts of London and how when gin was first invented, gin is just crappy vodka. You mix in juniper berries to, like, hide the terrible vodka taste. It was the first industrialized, super cheap hard liquor. And everyone in London just started getting wasted and, like, all the time. Because it used to be like you had hard liquor, and it was fine, because you just couldn't afford to drink enough for it to be a problem.
Speaker 1:Mhmm.
Speaker 5:And suddenly, that wasn't true, and it was a real problem. And we we've now developed cultural antibodies. Everyone knows doing these things is a sign something's wrong. They don't even necessarily have to know why. We just know that that's a sign that something was wrong.
Speaker 5:We need to develop pretty quick this time. This is gin spread relatively slowly because it was a long time ago. ChiangGBT is spreading very, very, very fast.
Speaker 1:But I would also I would also say that that we haven't developed those antibodies yet fully for social media. Like, it's normal No. Haven't. Teens will use TikTok for six hours a day. Yeah.
Speaker 1:Like that should be sending alarm bells as like Yes. You need more things going on in your life. Yep.
Speaker 5:You're using TikTok are you using TikTok before 8PM and after or after ten? Like are you using TikTok not with your friends? I don't I don't know what the rule is, right?
Speaker 1:But
Speaker 2:like Yeah. Yeah. Yeah. It's the same thing as having a drink in the morning.
Speaker 5:We need bright line rules for ourselves because it's too hard to try to figure it out.
Speaker 2:Yeah.
Speaker 5:And we need to
Speaker 2:develop the story in audience. We're starting to develop like the memes around this. Like the idea of brain rot is a powerful meme because it's a very negative term. You don't want to be suffering from brain rot even though it is as vague as alcoholism. What is alcoholism?
Speaker 2:Right. It's not necessarily quantitative, but we but we understand it.
Speaker 5:You need to pair with brain rot things like things like, oh, you're using
Speaker 2:Yes.
Speaker 5:You're using it before five.
Speaker 2:Yeah.
Speaker 5:Like, that's a that's a are you okay? Like, is
Speaker 2:everything Yeah.
Speaker 5:Yeah. Yeah. I'm not mad. Like, are you alright that you're using TikTok? It's, like, before 5PM.
Speaker 5:Like, that's not normal. Yeah.
Speaker 2:Or or you're or you're, like, scrolling in the middle of a conversation one on one with another person. You're actively getting brain rotted.
Speaker 5:Or you That's okay.
Speaker 2:Then, like, showing up to a work meeting with booze on your breath. That's the
Speaker 5:same Exactly. Fascinating. I don't know what they are exactly, but I know we need them stat because, like Yep. It's really obvious that, like, it can really get you. No.
Speaker 5:The the Especially the LLM's even worse than the The one two
Speaker 2:Yeah. The
Speaker 1:one two punches, generation grew up on social media, iPads, and then LLMs.
Speaker 2:And it's
Speaker 1:just Yeah. I understand. And before you develop the antibodies, and there's maybe a window of of of ten years where people just
Speaker 2:because it's all happening a lot faster because it's moving at the speed of this not a watch out of you.
Speaker 1:Anyways, we're we're way behind. I wish we had a full
Speaker 2:hour. Yeah. Yeah. Yeah. Yeah.
Speaker 2:Come back in a couple weeks. We'll we'll we'd love to talk to you again.
Speaker 1:Cheers. Bye.
Speaker 2:Bye. Next up, we have Brendan from Merkor coming in the studio, breaking it down.
Speaker 1:Ultra lightning round.
Speaker 2:Ultra lightning round. We're gonna rip through these really quickly. Thank you so much for your patience. Is. Brendan, great to hear from you.
Speaker 2:Start kick us off with a quick introduction of how things are going at Merkor because it seems like you've been on an absolute tear. Things just keep the dominoes keep falling in your direction. Give it to us. What's the latest update from you?
Speaker 9:Of course. Well, thanks, first of all, for having me back on. But it's been an exciting few months since we spoke. We're now working with six out of the mag, seven, all of the top five that they've done a lot.
Speaker 3:Oh, that's great.
Speaker 9:And had averaged 45% month over month growth for the last twelve months. And so it's
Speaker 2:Which of the mag seven are missing from the six out of seven? I wonder. Sorry. That's me, not you. Continue.
Speaker 1:It's been it's been
Speaker 9:an exciting time, especially in the wake of of all the news with
Speaker 2:Yep.
Speaker 9:Scale, of course, and seeing a huge amount of of customer interest after that.
Speaker 2:Yeah. Okay. We were talking to Clem about this earlier from Hugging Face. Where where is the biggest demand? People used to think of what you do as, and and just human data generation generally as, kind of work that anyone could do, mechanical Turk work, you know, just just, you know, checking for hallucinations and fact checking and, you know, is this a dangerous prompt?
Speaker 2:Is this saying a bad word? Now it's moved into PhD level work. Where is the state of the art? Where do you see it going?
Speaker 9:Totally. So I think the key development in the market is that reinforcement learning is becoming so effective that once we have evals for something, the models can saturate them. And so, really, anything that we want to create agents or LMs that are capable of doing, we need to build out evals and RL environments in those domains. And so, actually, while some of it is PhD level work, there's a huge push into all of the professional nonacademic domains moving away from things that that were academic of how do we find the consultants, doctors, lawyers, bankers that can evaluate and teach models how to do the things that we would want those professionals to do on the job. And so that's been a really exciting growth area that we've been leaning significantly into.
Speaker 2:What's the solution to booking a flight? Like, do we need to get travel agents to define the computer use workflows, create some sort of RL environment? It seems like I should be able to go to a chat app or Siri and just say, get me to New York tomorrow. Remember my preferences. I don't like connections.
Speaker 2:I'm willing to pay this much. I like to fly these times. And yet it's a very human problem that I don't think people are fully comfortable delegating to AI.
Speaker 9:Yeah. It's a fascinating dichotomy that we're simultaneously talking about PhD level reasoning and Olympian math, yet it can't do a lot of the very basic things like booking a flight. And I think the key is that, really, one of the primary barriers to research is anything that the model can't do, we need an effective way to measure. That way, we can experiment with all of the different datasets to help achieve those capabilities. So we need people that otherwise could book those flights to create evals for how agents can do that to measure what success versus failure looks like in in all of those cases.
Speaker 9:And there's just that huge build out happening across all of the hyperscalers with respect to everything from simple tool use and how we book things or buy things online all the way to super high complexity reasoning over how all of these things interact with really complex knowledge bases.
Speaker 2:What are you seeing in video specifically? And, like, is there a role for humans in the loop in the training cycle or data generation around, these video models? V o three is incredible. Feels like it's a beneficiary of YouTube. What are you seeing there?
Speaker 9:Yeah. I think expansion to multimodality has definitely been something that's, really exciting to see. I I think that, there's definitely a lot of rich content online for video. But, ultimately, how we measure what those good videos look like and sample that, has been really important. In particular, there's a significant amount of eval build out in retrieval over videos in long context of, like, how can we watch an hour long YouTube lecture and understand and, obviously, that's just an example for one video case and understand, like, what are the elements of that content that really matter to users and how can we more than just build useful models, build useful products, that all of these companies can distribute to everyone.
Speaker 2:Yeah. Yeah. That makes sense in this in the context of a lecture. I I I guess I'm wondering, like, for v o four v o three is incredible. The physics are remarkable, and I'm wondering if it's important to have a human in the loop describing I mean, I I generate a lot of, like, Michael Bay knockoff videos, honestly.
Speaker 2:And and there's things where it gets the physics perfectly, and I'm wondering if there's if that's a beneficiary of not just a YouTube video of a Michael Bay trailer, but actually having someone sit there and describe exactly what's happening in the Michael Bay trailer in plain text that then can be fed into the system. And if we need to go further with the tagging and additional, like, metadata or transcription, not just transcription of what's said, but actually what's happening in the in the actual videos? Is that an important step to get us to v o four where there's even less hallucinations?
Speaker 9:Yeah. It we're definitely seeing a meaningful amount of that in that people want to create all of the tags using a combination of LMs and human expert data around what's happening in the video so that they can effectively work backwards from there to use those tags for for generation. And especially for all of the really high complexity stuff at the frontier of what the models can't yet do or or might require a significant amount of reasoning. But we tend to do all of all of that highest complexity work, the stuff that is very difficult and and high skilled to produce.
Speaker 1:Mhmm. Anything else, Jordy? No. This is great.
Speaker 2:We're going through a lightning round. I know it was quick. We'll have you back soon to talk more. I'm sure there will be more news. It's the, it's the hottest industry in the
Speaker 1:wanna see the actual revenue ramp chart.
Speaker 2:It's gonna be dropping soon.
Speaker 1:You can trust you can trust me with the revenue ramp
Speaker 2:chart. Yeah. I won't tweet it out
Speaker 1:before you
Speaker 2:no. We will we will respect embargoes if you have an upcoming milestone. I'm sure there's something coming up. So congratulations
Speaker 7:on all success.
Speaker 2:Fantastic positioning in the market. Get some sleep. We'll talk to you soon.
Speaker 9:I'm Azaz.
Speaker 1:See you guys.
Speaker 2:Have a great one. Bye. Up next, we have our next guest in the studio already. Sam from Superdial. Welcome to the stream.
Speaker 2:How are you doing, Sam? I gotta get this ready. I think I think we got some news.
Speaker 1:Let's give it up for Sam.
Speaker 2:We got a couple. We got a couple today. Couple big milestones happened. How are you doing, Sam? Good to meet you.
Speaker 10:How's it going? Good to meet you guys.
Speaker 2:What's new in your world? Can you kick us off with a little bit of introduction on yourself and your company and what's what's top of mind?
Speaker 10:Yeah. I'm Sam Schwager, cofounder and CEO of Superdial. We're a voice AI company focused on automating phone calls for health care. So we're we're specifically RCM nerds. If you're familiar with the, you know, medical medical billing space, we used to be an RCM company actually called Superbill.
Speaker 10:Oh. And then we got into to automating our own phone calls, brought that to the to the broader RCM market and converted the entire business to Superdial. So that was about a year and a half ago. And today, we're excited to announce that we have raised our series a from SignalFire.
Speaker 1:Here we go. Clean hit. It's a clean hit. Nice work, John.
Speaker 2:How much did you raise? 15. Yeah. 15. 15.
Speaker 2:There we go. Do you do you run into that weird case where, when people realize they're talking to a voice AI, they just keep it on the phone and and chat with it like a like a friend for two hours? Does that ever happen?
Speaker 1:Running up your your your cloud bills?
Speaker 10:We we don't because we're focused on these b to b
Speaker 6:B to calls. Yeah. Yeah.
Speaker 10:We're we're calling a we're calling a line. I know I think people kinda you can get a kick out of it that they're talking to a voice AI because, you know, these are like someone taking phone calls all day Yeah. To answer questions about, you know, claims benefits or prior auths.
Speaker 2:So Yeah.
Speaker 10:Pretty pretty repetitive transactions. So, yeah, when they get, hey, I'm Billy, you know, like I'm I'm calling to check on the status of a claim and
Speaker 7:it sounds Yeah.
Speaker 10:Kinda robotic, then I think that that can be kinda fun for folks. It's very polite to the point consistent.
Speaker 6:Yeah.
Speaker 10:No. We we haven't Billy has not served as a as a therapist. Sure.
Speaker 1:Sure. How how how competitive is this market right now? I imagine you guys have sharp elbows winning deals clearly. Are are a lot of people attacking this market going after very specific verticals? Yeah.
Speaker 1:You guys are relatively broad. Maybe you get the ability to do that because you had been in the business at a high level for for quite a while. But what does the go to market motion look like right now? And how do you how do you expect the market to evolve over the next few years?
Speaker 7:Yeah. Well, the the core of
Speaker 10:our business is actually pretty specific. So we we handle phone calls between RCM companies or like the billing team that sits within a provider org. Those are outbound phone calls into the payer, the health insurance company. So it's a you know, there there are, you know, billions of these calls by many estimates. So huge high volume use case.
Speaker 10:But in terms of companies that are, you know, addressing that specific type of outbound call, there are fewer. I think if you look at, you know, appointment scheduling or, you know, different, you know, types of phone based use cases, customer support obviously, then the, you know, set of companies in the space explodes. But, yeah, when you kind of, you know, overlap enough Venn diagrams, you get down to a set of a few companies. But, I mean, there are definitely others, you know, where, yeah, I'm right now at the I I didn't just dress to match you guys. I'm out of conference.
Speaker 2:I love the suit I was gonna say. It looks fantastic.
Speaker 1:You should've you should've told us. Thank you. You
Speaker 2:should've just should've Hey. Just for us.
Speaker 10:We're in the Superdial Quarter Zip, you know, but There
Speaker 2:you go.
Speaker 10:We're at the Healthcare Financial Management Association conference, the annual meeting in
Speaker 2:I'm it's going crazy right now. You walk in. Man. We we yearn for
Speaker 1:the floor. Yeah. It's out there. It's fun out there.
Speaker 2:Floor. Yeah. It's great. Last question from from my side. I mean, you're wearing the suit, obviously, building in the application layer.
Speaker 2:At what level of the stack in AI are you integrating? Like, are you building on top of a voice model that's a provider? Is it all is it all other startups, or have you had to fine tune stuff and and do kind of your own models at different layers? What are you building on top of?
Speaker 10:Yeah. Yeah. I mean, we we have a text to speech partner, speech to text in LLM. And then we, yeah, we we use some cloud computing Sure.
Speaker 1:Better, of course.
Speaker 10:But, otherwise, in terms of, you know, orchestration, like Yeah. Phone call orchestration, we use, you know, open source. We're not Yep. Using kind of one of the, you know, full service, voice agent platforms. And then, yeah, I mean, it really comes down to, like, perfecting the workflow and the integration.
Speaker 10:Like, you gotta know, like, what phone number to call and how to get through these phone trees.
Speaker 1:That's
Speaker 10:Yeah. That's actually pretty tricky.
Speaker 2:Yep.
Speaker 10:You're you know, you have to in in, like, your system prompts and, like,
Speaker 7:you know, the
Speaker 10:static ones and then, like, in the, you know, dynamic bits that you inject at call time, you need to have, like, sufficient domain context, like a bunch of, like, cycle management jargon or else the AI agent's gonna get lost on the call. So that's what we're focused on at at Superdial.
Speaker 2:But Yeah. Yeah. That makes ton of sense. Focus on, like, creating that value in the b two b context and and owning the customer relationship. And, yeah, this isn't something that's on the road map for Eleven Labs.
Speaker 2:So just partner with them. I don't know if you're working with them, but probably someone like them. Anyway, this has been fantastic. Thanks so much for stopping by.
Speaker 7:Good luck
Speaker 1:out there.
Speaker 2:Next time there
Speaker 1:is Send us
Speaker 2:some photos.
Speaker 1:If you close some deals
Speaker 2:Yeah.
Speaker 1:Let us know. Shake some hands. We'll send a video of a gong hit to you. Fantastic.
Speaker 10:Absolutely. Alright.
Speaker 1:We'll talk to you Cheers. Have a
Speaker 2:great rest of your day. Next up, we have Zach from Warp coming into the studio. Zach Lloyd. We will welcome him to the studio. It's been a good show so far.
Speaker 2:We got three more folks in this
Speaker 4:lightning round.
Speaker 2:Welcome to the stream. How are you doing?
Speaker 3:I'm doing great. Thank you both for having me here.
Speaker 2:Thanks for hopping on. Would you mind kicking us off in the introduction on yourself and the company?
Speaker 3:Yeah. So I am Zach Lloyd. I'm the founder and CEO of Warp. As of today, Warp is an agentic development environment. We are a sort of in the AI developer tool space.
Speaker 3:The sort of general thing our product does is it lets you prompt something that looks like a terminal interface to run developer agents, and they code. They can debug production issues. They can basically do any kind of development task.
Speaker 2:You said as of today, what what what what has been the key change?
Speaker 1:Warp two point o.
Speaker 3:Yeah. So today today, we launched Warp two point o. Congrats. I mean, the the the backstory of the company is we Congratulations. We we started off with the vision of kinda reimagining the terminal, which is this very, very old school developer tool, you know, black screen, green text.
Speaker 3:And as the world has changed with more AI, we realized actually the the terminal interface, the form factor of that is kind of awesome for deploying agents. Running an agent is very similar to running a command. And so as of today, we're kinda, like, fully in the game with a we have, like, a state of the art coding agent. It's it's towards the top of the list on the SuiteBench eval. It's the number one on the terminal bench eval, and so we're we're kind of in in the game of, like, automated software production at this point.
Speaker 2:Is the, how do you see, like, kind of the modern software developer workflow over the next year? You know, folks using Cursor, Windsurf in their IDEs or GitHub Copilot. Then there's these agents that you can
Speaker 1:Codex, Devic.
Speaker 2:Yeah. You can kick off Codex from inside ChatGPT, which feels like a very different flow than the than the terminal. Is it a certain type of developer that gets the most value out of this? Do you see it more of as, an ensemble approach? What what are you thinking?
Speaker 3:Here here's what I see happening. So I I think we are, we're moving from a world where most development has been done traditionally by hand. Mhmm. What I mean by that is, like, if you're a developer, and I've been a developer forever, my everyday workflow is I would come in. I'd be like, okay.
Speaker 3:I wanna work on this feature. I wanna work on this bug. I would open up my code editor, and I would find some files, and I would type some code. And then I would go to the terminal, and I would type some commands to build that code. And I think that the big shift that's happening right now is that instead of, like, doing all this work by hand, developers are gonna be starting most of their tasks with a prompt, and the prompt is gonna launch an agent.
Speaker 3:Mhmm. And the agent will do some, maybe all of the task depending on how complex it is. And so I think I think that's the workflow, the job of an engineer is gonna increasingly look like how do I multitask? How do I multithread myself by being able to, like, have one agent fixing a bug, another one debugging a production issue? So I I think that's what's happening for the next year.
Speaker 3:Beyond that, I think maybe it looks a little different, but that's that's what's now possible, and it's pretty cool.
Speaker 2:Yeah. I am. I'm I I wanna get to, like, the minority report style interface or, like, some sort of completely new UI paradigm, but I guess the terminal is undefeated. It's extremely Lindy at this point. But, I mean, do you see people using, like, you know, 25 multiple, you know, terminal tabs and kind of switching through all of them as different agents?
Speaker 2:They pop stuff off to just Yeah.
Speaker 3:I mean out their armies. That's the state today. I think when you when you look at where you can do this, because you asked that. So you can do this in an IDE. And if you look at, like, Cursor, WinSurf, they have a sort of chat panel Mhmm.
Speaker 3:That, like it does something kinda similar to the terminal. Honestly, it's like a log of, like, what an agent is doing. You could do this with something like Cloud Code, which is pretty cool, and it's actually kinda proving that this workflow is is is, like, achievable. Yeah. But the the disadvantage of something like Cloud Code is it, like, it's a terminal app, and so it's not the whole platform.
Speaker 3:It's like a thing you run within the terminal, and that kinda limits pretty severely what the user experience could look like. Or you could do it in something, you know, with what we're trying to do, which is, like, basically create an interface so that you can do this kind of multithreaded development, and it's got a very natural interface. And, actually, just, like, the interface of the command line is basically what you need. You need a way of telling the computer what to do and, like, splitting off tabs and panes and watching it do it. Honestly, you might not even need to watch it do it a lot of the time.
Speaker 3:I I think that that's coming pretty soon too. But, like, that's the interface. It's like, how do I tell the computer what to do on multiple tasks, have it check-in with me when it needs my help, be able to, like, hop in and actually edit, and so you can do that in Warp now. And so but, like, you know, it's gonna do a lot of the work under your supervision.
Speaker 2:Yeah. Is open source important to your business? Do you have a horse in the race? Do you care about, you know, how llama five turns out, for example, or what's going on with DeepSeek? Or do you just wanna pass through the actual inference cost to the user, let them pick?
Speaker 2:What what what's the strategy there?
Speaker 3:Super interesting question that I'm thinking about a lot right now. I mean, so so right now, I think it's very clear to us that we're gonna give just the best experience to our users, and that is using today, that's using Anthropix models, just to be totally frank.
Speaker 2:Yeah.
Speaker 1:I mean,
Speaker 2:that's what people like. Right?
Speaker 3:What's that?
Speaker 2:Yeah. I mean, that's just, like, the popular one that we hear about all the time and Yeah.
Speaker 3:And, like kind of cursor default. Have sort of evals that that test all these models, and, like, there are interesting models from OpenAI that are competitive. Gemini from Google has some interesting aspects, has a very long context window. Yep. But, like, our attitude on this right now, least, is, like, let's pick the thing that provides the best user experience.
Speaker 3:And at the moment, that you know, I don't think I'm saying anything groundbreaking. If you look at the other benchmarks
Speaker 2:Yeah.
Speaker 3:It's it's like the Sonnet. It's like Cloud four and Sonnet and Opus. Yeah. I do think it would be awesome if the what what I want is, like, a healthy competitive model provider landscape, to be totally honest. Yep.
Speaker 3:I don't really want one model provider running away. I would like to see these model providers continue to compete with each other to
Speaker 2:provide Price down for you. Right?
Speaker 3:It it it it brings the price down. It brings the quality up. Yeah. And if at some point there is an open source model that is at the same level, that's super interesting. Because because where we're adding the most value right now is, like, it's in it's in packaging the entire user experience into something that takes what's best out of the magic of these models and makes it so that users can actually, like, you know, easily use them to accomplish their development task.
Speaker 1:Yep. Anything else, Corey? How how are you thinking about, you know, do you think that in the long run having open source models get really good is gonna be key to the durability of the business over the next decade? Imagine, and, you know, the the dynamic with Anthropic obviously cares a lot about code gen Yeah. Developer experiences.
Speaker 1:You know, this is an area that they're clearly going to generate, you know, would like to generate, I imagine, tens of billions of dollars a year. And and so it has to be you know, it sounds like a great relationship right now, but how do you kind of, you know, continue
Speaker 3:to evolve? A tricky one because they're they're we're both their customer and their competitor. And so I don't know how that's gonna evolve. Like, optionality us having optionality does seem like a potentially valuable thing. But, you know, right right now, the way it is is, like, we're working great with Anthropic.
Speaker 3:They're, a close partner of ours. And, you know, they seem very, very committed to doing the API and doing the end user product. Not too different from, you know, a place like Google or AWS who has, like, cloud infrastructure and also builds things on top of it. Mhmm. I don't think it's, like, a necessary condition for us to have a working business model that open source models, like, catch up as long as there's competition amongst the the foundation model providers.
Speaker 3:If we were in a world where there was just sort of, like, one provider who won, I think that that's, like that's a tricky spot, but I think that's a tricky spot for anyone who's building at the application layer right now.
Speaker 1:Yeah. Yeah. We I have a portfolio company that was kicked off one of the the the the big lab that will go unnamed that their business was entirely dependent on. And over you know, within basically twenty four hours, he had to, like, completely you know, he ended up surviving and thriving. Yep.
Speaker 1:But it was tumultuous. But but, yeah, sounds like a great relationship for now, and and hopefully, we'll continue to be.
Speaker 2:Yeah. That's great. Thank you so much for stopping We'll we'll talk to you soon.
Speaker 3:This was awesome. Congrats on
Speaker 1:the lunch.
Speaker 2:Talk to you soon. Cheers. Bye. Next up, we have Dara from Delphi coming in the studio. I think we got
Speaker 1:Maybe who knows? Will it be the real Dara?
Speaker 2:Yeah. I was about to say, is this a clone? Give it to us straight.
Speaker 1:It's a clone. It's a clone. What's our safe word, Dara? What's our safe word? Just kidding.
Speaker 1:What's going on? News. What's up guys?
Speaker 2:How's it going?
Speaker 1:It's great.
Speaker 2:Welcome to the stream.
Speaker 1:It's great to see you. We we invited you on many months ago.
Speaker 2:We did.
Speaker 1:Well, but you're late. You were cooking. We were cooking. And trenches. In the kitchen.
Speaker 2:Yeah. In the trenches.
Speaker 1:Yeah. Well, do you have some news? John's got got the hammer ready. I'm
Speaker 11:ready. We do have some news. Today, we announced our series a with Sequoia.
Speaker 2:Let's go. How much?
Speaker 11:16,000,000, but look You guys are really investors.
Speaker 1:Early investors. Fantastic. Proud.
Speaker 2:We love to see it.
Speaker 1:Give us give us the state of by by our friends over at Sequoia.
Speaker 2:Yeah. Give us the state of the union. How would you, how are you pitching the product these days?
Speaker 11:Yeah. So Delphi lets people create a digital version of their mind to scale their expertise. And you can kinda think about it like a new form of If you write a book or create a blog post or create a YouTube video, you can reach millions of people, but it lacks the personalization of one on one conversation. So now your mind, which usually has a cap, can be in multiple places at a time in the way that you would be in conversation. And so that kinda spans across many different verticals.
Speaker 11:We have coaches using it to scale their client practice. We have authors using it to make their books accessible in an interactive way. We have CEOs scaling themselves internally in their companies, and we have some people using this as, like, an interactive resume or LinkedIn.
Speaker 2:How are you thinking about the economic model? I I take me through what's what's happening now, but, we had this interesting discussion with Derek Thompson, the the author of Abundance, and he was saying, paradoxically, his TV hits sold more books for the book that he was selling than the long podcasts. Because if he goes on a podcast and he does two hours of content, people are like, okay. I got enough abundance. I don't need to read the book because I got the whole Joe Rogan or the or the Lex Friedman deep dive.
Speaker 2:But if he does a three minute hit on NBC or ABC, people are like, oh, that was interesting. I'd love to dive into that book. So I can imagine there's this situation where if I'm an author or a thought leader or something and I set up a Delphi clone and someone interacts with it for five hours a day, like, they that's a good substitute for the rest of my content. And so I'd wanna monetize that to the max. So I wanna hear about monetization now and where it's going.
Speaker 11:Yeah. So we have some people monetizing their Delphi and making multiple 7 figures in revenue, and this is, like, insane.
Speaker 2:Woah. That's a lot.
Speaker 11:And and the the consumer phone calls with these Delphines can span up to six hours.
Speaker 2:Six hours on the
Speaker 11:six hours, but you're choosing your own adventure.
Speaker 1:Yeah. Yeah. Yeah.
Speaker 11:Station. So course completion rate is an all time low. All of our course creators are like, it's a course winner. People's attention spans are declining. They want more guidance.
Speaker 2:Course winter. Interesting.
Speaker 11:Monetizing your Delphi directly is an option. But, alternatively, a lot of people make it for free, and it increases the strength of the parasocial relationship between the user and the creator, and they actually end up wanting to buy their products even more. Yeah. Or even two step further, analytics dashboard, in the last week, give me the most profitable opportunities given the conversations my audience is having. Content ideas, products, people I should prioritize.
Speaker 11:So the analytics actually ends up being very valuable.
Speaker 1:That's awesome. Yeah. What what what does the next year look like? You guy I mean, I I feel like I have so much context on the business. But what what does the next year look like for you?
Speaker 1:I feel like you guys have had competitors already kind of come and go. How are you thinking
Speaker 2:about market?
Speaker 1:How are you thinking about the market, and, what are you guys gonna be focused on?
Speaker 6:Yeah. I mean, when we
Speaker 11:started the company, Jordy, you know, the first year, things were just not working. And most people, would pivot away, but I think when you're building a product where people have to trust you with their data and likeness, quality and brand and trust are super, super important. So I think past that first year, we got that quality and trust. Now we have thousands of customers who have created a digital mind of themselves. And the next step is how do we bring on the next million people?
Speaker 11:How do we make it easy for someone who isn't Tony Robbins to create a digital mind and actually get value from it? And the the common use case we see is twofold. One is a new kind of personal website for the Internet where if someone wants to pick your brain, they can talk to your Delphi, and it's gonna notify you if you should talk with them in person.
Speaker 4:And a lot
Speaker 5:of people
Speaker 11:actually end up talking to themselves. So that's on the consumer side. And on the actual creator side, so people are monetizing their Delphi, but what the consumers actually want is guidance. So the next version of a course or a book isn't just, like, open ended conversation with the author, but rather more of a guided experience towards a specific goal, which we call a journey.
Speaker 1:That's awesome. And the beauty of Delphi is as the models get better, context windows get longer, Delphi's product experience get better. Is that right?
Speaker 11:Yeah. And the service area with the product experience is kind of threefold. One is our digital mind architecture, which we're actually trying to capture your mind and worldview and how that changes over time. Two is the actual experience for creating your mind, getting the insights, monetizing it. And then there's the consumer experience for calling, messaging, and video calling.
Speaker 1:That's great. What, what what does success look like over the next year? I I I know you're very ambitious. You've already onboarded, you know, massive number of, kind of a listers from different categories? What are what are you looking to prove over the next twelve months?
Speaker 11:I think twofold. We want multiple Delphi millionaires making money, versus just a couple. It's like a new field of monetizing your expertise, and we want a 100,000 to a million people using their Delphi as their personal page on the Internet. I think that'd a very good outcome.
Speaker 2:Talk to me about some of the weirder potential Delphis that you could potentially interact with in the future. Character AI for a while would let you talk to Joseph Stalin. I tried to debate Stalin on the merits of capitalism versus communism. And the the model that they rolled out was kind of too RLHF. So it kept saying that communism was bad.
Speaker 2:And I was like, I don't think Stalin would be making those points. This feels like an American elm. But, have you thought about offering a product for historical figures or talk to your late grandmother or historical records or, like, talk to your dog or some other sort of, like, persona that's kind of outside of just an individual creator wanting to offer their knowledge.
Speaker 11:Yeah. I think our main focus is people scaling themselves. And we have some historical figures on our website as a demo like Socrates and others. We don't allow people to create a digital version of anyone who's not themselves or that they don't have permission. So we couldn't do Steve Jobs right now unless we have, like, the Steve Jobs Foundation's permission.
Speaker 11:Interesting. And in in terms of legacy, you guys know the idea of the company was inspired by me wanting to talk to my grandfather. Yeah. And I think pretty early on, we decided we didn't wanna be associated as a company that's trying to profit off of grief. Yeah.
Speaker 6:So I
Speaker 11:think the best world is eventually we can just have a free tier for people to do that with their parents and grandparents, and that's not really a business model for us, but rather just like a good thing people want to remember their families.
Speaker 1:Yeah. How do you how do you think about, just like data capture and and creating the the training data for these individual clones? I think, we're in a unique situation in that we we record three to four hours of live video content a day.
Speaker 2:A lot of people are saying like it's not enough.
Speaker 1:Yeah. A lot of people are saying
Speaker 2:that they need to get eight hours
Speaker 1:Eighteen hours
Speaker 3:so that
Speaker 1:it can always be on as well. Yeah. With
Speaker 2:BCI, just pipe it directly in. Two x. They need forty eight hours a day. Yeah. But, yeah, data capture.
Speaker 2:What what what are the what are the richest sources of of content that actually
Speaker 1:drive value? Other thing that like an interesting challenge is, you know, people have certain views that they want expressed online and then other views that you could capture that they don't want expressed online. Right? If you wanted to if if an investor, you know, wanted to clone themselves, they might wanna talk positively about certain companies and not share their real views on other companies that they may or may not be invested in. So I feel like it's a hard challenge to figure out what information should actually be made public and and actually trained on.
Speaker 11:Yeah. Representing a human being in every facet of their life is is very difficult. But, we have a couple ways we've thought about this. In terms of data sources, YouTube, social media, Notion, Google Drive, and it creates feeds so that it stays up to date whenever you have a new tweet or a new YouTube video. It just updates itself.
Speaker 11:And for those who don't have a lot of data, we have this mind quality score that tells you how good your mind is at representing you and can ask you questions to incentivize data entry to make it better. And then like you said, who you are at work is different than who you are with your friends. You may behave differently. You may share different information. So now we have this ability, to separate different versions of your mind that has different access to data.
Speaker 11:Right now, have to manually do that. So I I have one for eventually my great grandkids that has more intimate Dara data. But, ideally, it can be an intelligent, and it knows what you would say depending on who the person is.
Speaker 1:That makes sense. How are you thinking about working with the underlying labs? I know you've had some problems historically, with with different vendors, but what what's your updated thinking? Are you guys leveraging open source in a big way these days or or or have you found, good ground with with, some of the core labs?
Speaker 11:Yeah. We're kinda both. We have some of the core labs. We have open source. Given our experience in the past, we are trying to get to a world where we are fully model agnostic.
Speaker 11:And the true architecture and innovation that we're focused on is what what is that representation of the mind that can be such that your mind and my mind, you can look at them, and they're clearly two different entities. And the models we use on top of that can be agnostic.
Speaker 1:Makes sense.
Speaker 2:I wanna know about, like, the the the the different, like, touch points here that you could potentially go after. Remember there's this Do you remember this trend where social media influencers would be like I leaked my phone number. Do you remember this? Yeah. Or they'd like textwasthatcommunitycom?
Speaker 2:Yeah. Yeah. And so there's a world where somebody that like I I think that was basically just like ingest point for something that was essentially like a mailing list for promotion. And it was like text me and then you'll be opted into my email blasts. But you could imagine that being interaction point for essentially, I mean, get to be an influencer at scale.
Speaker 2:There might be it's almost like customer service. And you're interacting with basically like an FAQ. Like, hey, I wanna know where to get a TBPN hat. Like, can you tell me? And in just being able to DM your personal profile on x and interact with that or text you, that's one kind of modality that people could interact with.
Speaker 2:There's also like the big social media companies could put a button there that allows you to chat with that user directly. And so how do you see the market playing out? What are you worried about? How are you counter positioning against the different players or different Yeah. I remember
Speaker 1:I remember the exact day that Meta launched digital clones Yeah. In their own way. And I think Dara went through the cycle that every entrepreneur does when they get when somebody else, you know, attacks
Speaker 2:their Step one, all the VCs text you, have you seen this?
Speaker 1:Have you
Speaker 2:seen this?
Speaker 1:Yes. Step two, just enraged. Step three, realize that, you know, it's not actually going to end you Yep. And and it just validates your
Speaker 2:preexisting grow faster because maybe because more people are aware that this is a thing, and they might want a better version of it. But, yes, yeah, just general understanding of the marketplace.
Speaker 11:Yeah. I think there were a couple of core decisions we made early on that have set us up for success. One is that while most big companies and AI companies kind of are using your data to try to create AGI because that's where 99% of their revenue is gonna come from in the future, we're saying you own your data. You own your digital identity. We're not training models on it.
Speaker 11:And that has actually gotten us a ton of customers, people who wanna own their data, and they don't want these big model companies to rip them off. Number two is we think if you're gonna go out of the way to create a really good digital version of yourself, it should work everywhere. So a lot of big tech companies are walled gardens. Like, it's only in those platforms where Delphi, you can create your digital mind, SMS, WhatsApp, Zoom, Slack, Discord, Telegram, all these different sources, and it's a central source of of that. And then three, I I just think that it's such a new habit.
Speaker 11:Delphi isn't really meant to be deceptive. Like, it doesn't reply to emails for you or respond to DMs. It's more of just like a new form factor of consuming someone's content. So on Instagram, it's a platform that promotes people who are very aesthetically pleasing, where Delphi promotes people who are pleasing for very different reasons. And so it's just a different incentive structure in terms of the network and who gets up at the top.
Speaker 1:Totally. What are you guys hiring for?
Speaker 2:Yeah. What does it take to earn a $100,000,000 at Delphi? Yeah.
Speaker 1:Are you guys hiring for any AI research research any 9 figure signing bonuses?
Speaker 11:We should foundation models right now,
Speaker 1:but we
Speaker 11:are looking for AI people and a lot of design engineers. I think the interface for the nine mind needs to be beautiful. We're honoring human beings, and people need to feel special when they're creating their digital mind. Eventually, when we have the consumer platform that we call the modern day library of Alexandria that every mind is going to be on today and five hundred years from now, It needs to be beautiful. So design and design engineering.
Speaker 2:Well, we'd love to recommend Figma and Linear for you and your team. If you're not already using them, they are sponsors of this show. They make it possible.
Speaker 5:I'll check
Speaker 7:them out.
Speaker 2:Don't think I'm gonnaeverdo.com. You heard it here first. It's a name brand Right. In the Akoya backed world.
Speaker 1:To I I know this round happened a while ago. Yeah. But congratulations to you and Sam and Spencer and the whole How
Speaker 2:much was it?
Speaker 1:Hit it again, John. There we go.
Speaker 2:Production team's getting better. They can cut to the gong cam when the gong hit when the gong hit happens. It's here for the production team.
Speaker 1:It's great to see you, Dara. Congratulations on the milestone. Come back. There's more news. When you have more news.
Speaker 2:We'll talk
Speaker 1:You're the man. Cheers. In
Speaker 2:other news, friend of the show, Bozz, not really a friend, but yet we hope he will be a friend. We covered his beautiful profile of his house in the mansion section. He followed us. So we're gonna get him on the show soon. We gotta
Speaker 1:make it happen.
Speaker 2:I'm the biggest fan of this guy. He posted that the Meta Quest three s Xbox edition is now available. Everything you love about Quest and Xbox in one sweet package. It's sleek. It's awesome.
Speaker 2:And it's limited edition. And you know what else? I already ordered one. It's
Speaker 1:coming in
Speaker 2:the studio.
Speaker 1:There we go.
Speaker 2:Yeah. I'm very excited. And and so I I mean, it doesn't seem like this is like the next
Speaker 1:Uh-oh.
Speaker 2:What?
Speaker 1:What? We said this is a real risk factor. Oh, it is. Is. Addiction.
Speaker 2:I'm not addicted to alcohol. I'm not addicted to gambling. I'm addicted to video games. Seriously. If I ever go down a real bender, I will
Speaker 1:on because I've never seen you play a video game. Oh, yeah.
Speaker 2:The entire time we've been in the show, haven't logged an hour of video games.
Speaker 5:I haven't played
Speaker 2:them in a long time. Oh, man. But you get me the GTA six drops?
Speaker 1:I'm just ripping this thing
Speaker 2:off face. Using Adelphi here in my
Speaker 1:We gotta go live.
Speaker 2:Yeah. But I have been, so so this is not the Quest four. This is not a a proper hardware iteration where we're getting the screen from the Apple Vision Pro pulled in, which I think will be in the next version, which will be a major, major development because there will truly be no screen door effect and you will like like, the the resolution will be remarkable because Yeah. That's what the Apple Vision Pro did so well. That technical hardware now exists in the world.
Speaker 2:Meta's clearly gonna go get that and bake that into the next Quest four. But the Quest three s has already been a pretty solid product. But what they're doing here is they're leaning into video games that are not VR specifically. And I think that this is 100% the correct move because so much of VR content is this chicken and egg problem. No game developer wants to go and spend I mean, how long have they been working on GTA six?
Speaker 2:Like, a decade? Like and it's cost them, a billion dollars. No one wants to do that if no one has the VR headsets. No one one wants to buy the VR headsets if there's no GTA VI on it. What do you do?
Speaker 2:With Meta Xbox, you get GTA VI on the Xbox, and you just play it on a big screen in VR, and you're able to sit on your couch. And instead of having to mess with a projector that might be a few thousand dollars or a TV that's a couple thousand bucks, you throw on this headset that's a couple $100, and you get the movie theater experience in your apartment, on a train, on a plane, wherever you are, you'll be able to stream Xbox games. So I'm very excited about this partnership. Excited for them go go deeper. And Microsoft early Meta Investor, actually.
Speaker 1:There you go. So Well
Speaker 2:hand washes the
Speaker 1:We have Amjad in the studio. Get the hammer. Get the hammer ready.
Speaker 2:Oh, yes. Yes. Yes. Give us the number. What's the news?
Speaker 1:What's the new revenue? We've been waiting
Speaker 2:for this.
Speaker 5:100,000,000.
Speaker 2:100 Boom.
Speaker 1:Let's go. Incredible.
Speaker 2:Congratulations. I
Speaker 12:thought this would be a 100 a 100 bangs.
Speaker 1:100 bangs.
Speaker 2:A 100 bangs.
Speaker 1:We'll have to do it after the show but I will give you
Speaker 2:one of these. Success.
Speaker 1:Classic overnight success. Replit. There we go. Alright, John.
Speaker 2:Broke the hammer.
Speaker 1:He broke the hammer. Replies going so fast, the hammer broke.
Speaker 2:The hammer hammer breaking growth. Congratulations. How'd you do it? What's the what's been
Speaker 1:What's the one simple trick?
Speaker 2:Yeah. Yeah. What's the secret to success?
Speaker 12:It's it's been easy. You know? Just a little bit of work. Not not too many years. You know, I started working on on Replit back in college.
Speaker 12:Like, I had the idea for it back in college, and, actually, the first implementation of it was open source, and it went viral on Hacker News in 2011. And and then I sort of we we left it for a while. It didn't work much on it. And then, actually, my wife, Haya, who's who's my cofounder, kind of revived the project. In 2016, we started as a company, and it was it was brutal.
Speaker 12:We barely were able to to raise money. We try to sort of, like, bootstrap it initially. During the pandemic, we sort of took off, especially as, you know, we were the only sort of collaborative editor in the cloud. And when when AI came on the scene, we just knew this is the future of our company because our mission is to make programming more accessible so that anyone can do it. And with the launch of agents back in September, we created a category basically of prompt to application, And and and that really sort of finally went from something that a lot of people use and love but don't pay for to something that's also commercially successful.
Speaker 2:Yeah. I feel like your company is now maybe the only AI company that's appropriately hyped. Like, it's not overhyped. You haven't raised the $10,000,000,000 round and you're paying the $100,000,000 salaries, but also the growth's been really serious and the business is really real, which is which is great to see. How have the, the crazy news of the $100,000,000 salaries been affecting you?
Speaker 2:Is it harder to hire these days? I know you've kind of changed structure of the team, relocated the team, I believe. So Yeah. Walk me through the current hiring market for, the type of engineers you're trying to hire.
Speaker 12:Yeah. So a few things that's, sort of different about Ruplit. One is that team is relatively small. We're still around the seventy, seventy five people.
Speaker 1:Mhmm.
Speaker 12:We haven't been growing all that fast. We are not in San Francisco, which makes it a little harder to hire. We're in Foster City. Probably the only startup, perhaps maybe there's a couple others in in Foster City. And, you know, we just take a different approach to the culture, and we, like, expect more commitment and want people to come to Repla and work for a long time.
Speaker 12:Especially since, you know, I've been working on this for so long, my mindset is that it's just gonna take a long time. And, like, San Francisco, Silicon Valley culture has been increasingly about, you know, I'm gonna go spend twelve months here, eighteen months there. And and so, yeah, we're we're building it more methodically. In terms of hiring, definitely, it's been it's been very competitive. And, you know, I love OpenAI and and Sam, and, you know, he he's talking about how Zuck is is doing that to them, but, obviously, OpenAI is doing it to the rest of us.
Speaker 2:And so it's just
Speaker 12:a different scale. You know? They're trying to poach their employees for 100,000,000. They try to poach everyone for 10,000,000. And so just a factor of, like, how much revenue, I guess, you're making.
Speaker 12:But looking
Speaker 1:But as part of part of from a hiring standpoint, because you're keeping the team small, now revenue per employee is Over
Speaker 2:a million dollars per employee.
Speaker 1:New evaluations at some point will allow you to sort of have comp you know, be competitive from a comp standpoint.
Speaker 12:Yeah. Exactly. And and, you know, the other thing about valuation is that I think start up employees should start thinking more about these things because a lot of companies can be profitable early on earlier on. And and so we're again, we're trying to build the business for the long term instead of, like, playing the the hype game. When you've been building a company for as long as as, you know, I've been, you know, you've you've bought you've been through so many ups and downs and hypes and lows that you just become very steady minded about it.
Speaker 2:Yeah. Totally. I I guess, like, the bigger not to not to no pun intended, the meta question here is is, are we in a new era of valuing the leverage that truly world class managers or AI engineers can have, the impact that they can have in reassessing the value. We've talked a lot about this on the stream that Apple's culture doesn't seem to be receptive to paying an NBA level salary. And Tim Cook, he makes $75,000,000?
Speaker 1:No. He doesn't make
Speaker 2:he doesn't even
Speaker 1:make 75. He only makes 74.6.
Speaker 2:It's a big number, but it's tiny considering their market cap and what if you could come into Apple and move the needle 1%, that's a $30,000,000,000 market cap move. Could you capture 1% of that? That's still $300,000,000 and yet they're not valuing talent in that way. It feels like there's a change in strategy at Meta. Do you think that that's a do you think that that's a unique case in where we are in the technological rollout of AI, or is it more that we're finally catching up to just understanding the economic dynamics and impacts like the NBA has for years?
Speaker 12:Yeah. So, you know, the the buzzword in Silicon Valley for a long time has been the 10 x engineer. Mhmm. About, you know, two two, three years ago and maybe in '22, I talked about the thousand x engineer. Yeah.
Speaker 12:Well, you know, highly kind of leveraged by autonomous agents. Mhmm. When you know, right now, people with replicable, cloud code, and others, they're spinning up multiple agents to work on their project as they're, you know, working on their main tasks, and there's a lot of other explorations and tasks and prototyping happening everywhere. And so you're really not just one person. You're a team of engineers.
Speaker 12:Right? And so, you know, we talked about engineering managers. I think every engineer is sort of a manager right now. And so, yeah, I mean, we're we're not at a thousand x yet, but we're we're going up. I think I think pretty soon the difference between a great engineer that knows how to use the AI tools and an okay or good one is gonna be a 100 x, I think, over the next six to twelve twelve months.
Speaker 12:So the compensation structure, I assume, would start to to change to match that.
Speaker 1:How do you think about Replit's potential today? It it seems like every day in the timeline, you know, you know, the obvious thing is everyone becomes an engineer and you know, it's being able to create software is fully democratized. But every day in the timeline, I see people realizing how much they spend on different SaaS products and thinking how much time would it really take for me to kind of make a version of this for myself that fits my very specific need like document signing
Speaker 2:Talk to the Clara CEO about this.
Speaker 1:Yeah. Some some something along those lines. And so is part of of the potential that you see CEOs and companies everywhere sort of running that sort of buy versus build calculus in terms of all sorts of tooling that they're using internally, externally, etcetera. And seeing, you know, trillions, you know, or or hundreds of billions of dollars of of market cap that that you guys could ultimately replace.
Speaker 12:Yeah. I mean, look. Right now, there is sort of information asymmetry. A lot of people in Silicon Valley and tech know how to use these tools. But a lot of people outside of here just aren't aware how much savings and how much productivity they can gain.
Speaker 12:I met someone from Australia the other day here in Silicon Valley. He stopped me on the street. I was like, oh, I love Revlet. He runs, a construction company. And he's like, you know, I'm the CEO now, and I'm the most productive engineer at the company.
Speaker 12:I, like, build all these tools and dashboards. I replace monday.com and all this other stuff with home built stuff that is act exactly bespoke and works exactly from a use case. Another story, the guy, his name is Ahmad George. He works at a skincare company, I think, in in DC, and he's an operations manager. He's responsible for, like, the ERP systems and things like that.
Speaker 12:They he had intuition that they could automate, I don't know, a big percentage of the work, 30% of the hours of human work. And he went and got a quote from their from NetSuite to build this piece of soft automation software. It was a $132,000 to build. So he went to Replit, spent $400, built it for their company, rolled it out, went to the CEO, told him, look at how much time and money I'm saving, and the CEO gave him $32,000 for
Speaker 7:the That's amazing.
Speaker 2:Wow. That's awesome.
Speaker 12:So so, yeah, it's it's you know, there's a period of time. At some point, you know, the market will, like, equalize as everyone realizes that you can do this stuff. But for now, there's a massive arbitrage opportunity.
Speaker 2:Yeah. As you build out agents and agentic workflows, do you have are you limited by data for RL pipelines? Are you are you buying any services from the Scalais of the world or the Mercos of the world? Or are you able to sit at a higher level of abstraction and kinda let the foundation labs get their hands dirty with all that?
Speaker 12:You know, having been a platform where millions of people coded over the years, actually, we could be a supplier of data.
Speaker 2:Oh, interesting.
Speaker 12:So so, yeah, we we have all the all the data we need. I think it's still early for us to try to exploit all of that, and we're hiring as fast as we can. So if you wanna come work on AI at Replit, I think we have a really exciting opportunities to to use data to kind of create amazing new capabilities.
Speaker 5:It's very
Speaker 2:cool. I have another question. Go for it. Are you feeling the acceleration in software development? I don't know.
Speaker 2:I I was thinking
Speaker 1:about this Pretty easy to when you refresh the Stripe dashboard and Yeah.
Speaker 2:No. No. No. No. I get that.
Speaker 2:I get that. But as a consumer, I was just thinking about like like the apps on my phone still haven't changed as fast as I've thought. And there's there's in my in my business, folks are using different tools. But it doesn't it doesn't feel like the rate of like churning for most consumers. Like, we like, we like, we're we're not rebuilding all the different consumer apps right now.
Speaker 2:It feels like we're very much in, this exploration phase for, like, what the next it feels like we're almost pre Cambrian explosion for consumer AI driven apps or uniquely enabled by AI apps. There's a lot of creation tools. So like, v o three is amazing. That got me to install the Gemini app and I paid for that and I enjoy generating my my paltry three video queries per day, sir. May I have some more Sundar, please?
Speaker 2:Maybe you can help me. Think don't you know him? Yeah. Help We're
Speaker 1:good partners.
Speaker 2:We my my TikTok feed. But yes.
Speaker 12:TikTok feed is a 100% AI now. Maybe because I like them so much. I've been into the Bigfoot
Speaker 2:Yeah. YETI. So maybe that's like the instantiation of the consumer adoption of AI is, like, you're a consumer through TikTok as opposed to through like, it's a very 2,010 mindset to think that my home screen apps would change, and instead the 2025 mindset is what is happening inside of those apps is changing. Is that the right framework?
Speaker 5:Yeah. I I think
Speaker 12:I think you're right. I I do think that Apple is probably holding back the ecosystem. I mean, Apple really has been a tax on startups, has been a tax on innovation. I love this company as much as anyone would love Apple, but but they they are they are sort of like a government in a way that there's, like, old and moving slow, and they just, like, put up so much barriers. Like, when I, you know, when I want to ask my phone I go to Proplexity, like, a 100 times a day.
Speaker 12:Like, I should be able to, like, press a button just, like, talk to it talking to me, but, you know, it's it's not working that way, and it is frustrating. But, look, I I think, you know, AI adoption is moving as fast faster than any technology adoption we've seen in the past, faster than a while, faster than mobile, but it still doesn't feel fast enough, like you say. And by the way, this is an interesting kind of data point on, you know, the the folks that are really worried about AGI and things like that. The limiting factor to adoption is really humans and corporations and governments, and all of these things are very, very slow.
Speaker 2:Yeah. This is kind of the Tyler Cowen take of, like, how AGI will be rolled out is there's a lot of sticky industries that just won't adopt it for a long time, and there will be lots of barriers. Is there anything else on the Apple side? My takeaway from WWDC was that on device inference could be a catalyst for an App Store explosion, an AI app explosion. Then there's the ruling about third party payment processing, things that that like, the tax is becoming less of a requirement.
Speaker 2:Maybe you can work around it. And so maybe the barriers are dropping ever so slightly with Yeah.
Speaker 1:And I feel like I feel like you would you would see this early. Totally. Yeah. That's why. Developers become really excited about this sort of effectively free inference where they can launch an app that goes viral without worrying about, you know, running up some bill with OpenAI.
Speaker 1:Yeah.
Speaker 12:I mean I mean, look. The Apple has held back the web quite a bit as well. Like, there's no reason why native apps should be much better performing than than web apps. Really, there's no fundamental physics reason for that. I've actually the head of product or upload, Jordan Walk, and and myself worked on React Native.
Speaker 12:He's the inventor of React and React Native at at Facebook. And the the reason we worked on that is we felt at Facebook, we felt sort of, like, oppressed by Apple. Like, they controlled what we launched, when we launched it, and they would, like, you know, reject you would have a launch, and it was delayed by Apple. And so the the idea behind it is, like, can we ship over the air updates very, very quickly? And and, yes, you can.
Speaker 12:But, again, Apple kind of, like, makes it a lot harder. And so I think look. I think AI has an opportunity to create an alternative platform entirely because the form factor of computing is on the precipice of changing. I saw you guys talking about VR. AR is another one.
Speaker 12:Is another one. You know, we've seen all these toy companies like Rabbit and things like that come and go. But there's but there's, like, an intuition there that's correct, I think. Everyone's feeling that that the form factor needs innovating. And I think this is how you break out of the, like, Apple Jail.
Speaker 1:Wald Garden. Yeah.
Speaker 2:Yeah. It's great. You gotta come back on where you could yap about all different news stories. Do have anything else, Jordy, you wanna
Speaker 1:right now? We we gotta get you on regularly. Yeah. Yeah. This Well, that's
Speaker 7:why I was here. I got canceled.
Speaker 2:So What? We went that was the most viral clip of that month. I I think so.
Speaker 1:What what happened?
Speaker 2:I think well, I asked I asked, should people learn to code? And he gave kind of a controversial answer saying like, no, you shouldn't learn to code anymore. And it got like millions and millions of views. So I bet it drove I bet
Speaker 1:it drove some some sign ups. For sure. I bet it did.
Speaker 7:For sure.
Speaker 2:Learn to vibe code. You gave a very nuanced answer. I don't think there was anything that There's no crazy about
Speaker 1:it. The Internet does not understand nuance.
Speaker 2:Canceled and viral are two sides of the same coin.
Speaker 12:Like Doesn't matter anymore. Doesn't
Speaker 7:matter to yourself.
Speaker 12:You know? Yeah. Haters gonna hate.
Speaker 2:Yeah. Just don't get deplatformed from Gemini because, otherwise, the the the three VO prompts be zero, and it'll and your life will be miserable because you gotta prompt those.
Speaker 1:Thanks, guys. I do we do
Speaker 12:great work. Really appreciate the show.
Speaker 1:Yeah. Well, come come back on at two hundred. At this rate, it'll probably be next week.
Speaker 2:Next week. Let's do it. We're ready with the gongs ready to go.
Speaker 1:Yeah. We'll get a new hammer. Yeah. We'll be ready. Thanks, guys.
Speaker 2:Talk to you soon.
Speaker 1:Cheers. See you soon.
Speaker 2:Bye. See you. Next up, we have
Speaker 1:Dirt man.
Speaker 2:Dirt man. Another anonymous account. I don't know why he's anonymous, but
Speaker 1:Well, he's launching he launched, cheaprobotarm.com today.
Speaker 2:Very excited for this.
Speaker 1:He's making, cheap robot.
Speaker 2:Go buy them now.
Speaker 1:He we we cover he used to have a different name on the show. Yeah. Used to cover his post.
Speaker 2:I used to laugh Yes.
Speaker 1:Every time because
Speaker 2:It it was rude. So we just call him dirt man. Welcome to the stream.
Speaker 1:Dirt man. How are you doing, dirt man? Changed his name.
Speaker 2:Oh, he did?
Speaker 1:He dropped the Okay. Parentheses.
Speaker 2:Wow. Classing it up. Introduce yourself. Who
Speaker 7:are you? Yeah. Yeah. Hey, guys. My name is Angus.
Speaker 7:Yeah. I'm I'm just a I'm an engineer from Australia. Oh, cool. My background's, like, very much in aerospace, and, yeah, I've been, like, obsessed with, like, building these kind of robot robots for, I don't know, like, the last ten years or something. Yeah.
Speaker 7:And it's, you know, building building this mini one, it's really just it was really, like, a stepping stone for me on the way to building welding robotics. You know, I've got some larger robots where I'm just setting up a workshop currently to do basically strap on cameras, sensor systems, and drive the whole thing drive the whole thing via AI because, you know, it's basic we're basically at the point where there's, you know, we have the ability to drive, you know, these really complex, you know, and industrial scale robotic systems, like, with, you know, the current robotic foundation models, you know, given the right training data. And so this, like, this small robot that I've put that I've got on, you know, cheaprobotarm.com is
Speaker 2:Great job.
Speaker 7:Was really really yeah. I mean, I'm I'm surprised that it available. Was like, hell yeah. ..Com.
Speaker 2:That's great.
Speaker 7:But but it was really it really just a test bed. You know?
Speaker 1:I'm work I'm working with
Speaker 7:a few guys, you know, one guy, Julian out of Pennsylvania, you know, runs a runs a shop floor, like, a welding robots. And the other guy, Steven, you know, currently works for NASA. You know, he's built working on the vision system portal. But it's really a test bed for us to, you know you know, validate our algorithms, do small scale prototyping. You know, my the picture that I've got up there, you know, I don't have many demos because it was basically thrown together, like, you know, 2AM my time last night.
Speaker 7:You know? Got the shop Mhmm.
Speaker 1:It live.
Speaker 7:Store up and running. And
Speaker 1:Doing it live. So
Speaker 7:what do what
Speaker 1:do you want people to use these for? What are the use cases that you're imagining?
Speaker 7:Yeah. So very much the same as, I would say, those Li LaRobot ones. It's just a much more robust platform. So improved, like, improved power in the servos, improved, like, structural integrity. But the key like, one of the key differences is I've actually built a, like, an industrial quality, like, kinematic solver.
Speaker 7:And, basically, what all that what that's telling you is, you know, where the end where the end effector of the robot needs to be. There's usually, like, a whole bunch of equations that you're either solving numerically or you've got a, you know, a solver in a low dimensional space that you can work out really quickly. Like, I built a, like, I built a really robust one for for, six degrees of freedom. And then this is just more than you get with the with, I guess, what's typically being used in all of those, like, AI, like, hackathons. And so I've written written all the software so you can take all of the outputs from an AI model and just stream joint commands to it, you know, connect it to, like, things like NVIDIA's Xisax SIM.
Speaker 7:But you can also run it like a traditional industrial robot and, like, and do path planning for really, really high fidelity, you know, trajectories and controls. And the, like, the benefit of that is, like, all of that all of that information that you can stream or record from those joints as you use the high quality kinematic solver, is you can use that as the training data. So you see a lot of these guy like, lot of these robot systems have got, you know, one arm and then another one's controlled by a human with a little lever, and then it's, like, going and doing these pick and place or whatever. Mhmm. And that like, you're just streaming, this joint should be the joint angle of this other robot.
Speaker 7:And frankly, that's really crap data for training robots. Like, the like, you get it done, but you can see all the movements all janky. It's the same kind of movement that the human's doing with a robot. It's awkward. But you like, you know, industrial robots, they've been doing this for decades.
Speaker 7:Right? And they have a really nice trajectory. It's really smooth path path planning. But the bottleneck is, like, how do you get that high quality data? And, you know, with this software combined with this, you know, little robot package, that's what I can do.
Speaker 7:You know? That's what this can do for people. You know? Help them get really high quality data without and only having to buy one robot. You don't need to buy, like, the equipment for two.
Speaker 1:That's amazing. What how is the health of this sort of open source robotic hacker world right now? We had Clem on earlier from Hugging Face, and he said this was an area that he was most excited about on on on his platform. And and he, you know, he said they hosted a hackathon recently. It sounds like there's a lot exciting activity here, and you're contributing to it yourself.
Speaker 7:Yeah. That's it. Yeah. I basically just posted all the, you know, all the code on GitHub for free. You know, I think there's it's real it's like a really exciting time at the moment just because there's, like, there's so much activity and attention in it, and there's many more people really coming into the field and, you know, having a go and learning.
Speaker 7:Could you like, not just robotics, but on the hardware side. Right? It's, you know, it's like, my background's aerospace and and mechanical engineering, you know, was you know, built rocket engines once upon a time, and it's you know, you know, we kinda got left out in the cold, you know, by Silicon Valley a little bit. You know, you you sort of like you know, but but you see all the, you know, the major all the really standout companies, you know, are all hardware companies, you know, space like, SpaceX and Dural, you know, a whole range of them. And I think there's a lot of, like, a lot of attention, like, just on this area has really put a lot of energy, particularly into the open source space, which is awesome to see.
Speaker 2:How have sales been? Have you sold any yet?
Speaker 7:Yeah. Yeah. So I've woken up. Well, you know, we've got we've got three sales so far.
Speaker 1:We go. Congratulations. Thank you, guys. Thank you, guys.
Speaker 2:That's three sales. That's $1,500 in one day. You annualize that. I think we're in the millions.
Speaker 6:I think
Speaker 1:I think if you put together the right deck, maybe you take a trip to Sandhill Road. Yeah. You tell the right Hop
Speaker 2:on a quick flight. 100 on a To SFL. 200 on a billion. Easy. Get it done.
Speaker 7:Get it done. Yeah. Yeah. Look. You know, it's you know, I think, like, yeah, Yusine said, like, hey.
Speaker 7:My evaluations should should be the same as figured because I've never shipped more.
Speaker 2:Would would be perfect for the AI grant program from Nat Friedman. You know, this is going to be the enabling technology. This is gonna be deeper in the supply chain for the robot that picks up the leaves or, you know, one of these fun, delightful robots. I hope Nat can continue that program. He certainly has the resources for it, and it'd be very cool to see that partnered up with the medicine.
Speaker 1:I would like a robot there.
Speaker 7:Definitely came to head to The US. Like, We'll
Speaker 2:sit you here.
Speaker 7:Yeah. I about eighteen months ago, I stopped by the Gundo, you know, met, like, Augustus, met a bunch of other fellas. That's awesome. It's been awesome to see those guys. Like, you know, I think I you know, I went there when Valor Atomics was just a squat rack in an empty room.
Speaker 7:Jesus. It was just a squat rack. Yeah. With scraps.
Speaker 1:Yeah. Well, you seem like an alien of extraordinary Ability. Product abilities. Let's get you. Know what.
Speaker 1:My vote is we get you over here.
Speaker 2:Bring you in. Yeah. This is fantastic.
Speaker 7:Be there in a in a couple of months, I think.
Speaker 1:Great. Well, let's tune the robot up to crack open
Speaker 2:Energy drinks. That would be useful.
Speaker 1:Yeah. The desk. It'd be great if we just hit a button. Yeah. Because it's hard to reach over here and and and crack them ourselves.
Speaker 2:Yeah. Yeah. It's brutal.
Speaker 1:Congratulations on Exactly.
Speaker 2:Yeah. Thank you so much
Speaker 7:for fellas. And, yeah, love love the show. Been a big fan since the beginning. So Yeah. Remember.
Speaker 7:Appreciate
Speaker 1:it. I'll never forget your name.
Speaker 2:Yeah. It was the first time it came up in the deck. This is great. Amazing. We'll talk to you soon.
Speaker 2:Have a good one.
Speaker 1:Cheers, Angus. Alright.
Speaker 7:Should
Speaker 2:we go through some timeline? Get out of here. Move on. We got a big show tomorrow.
Speaker 1:Let's rip a little timeline. A little just a little timeline
Speaker 2:couldn't hurt. We got Will Menidas talking about there no there is no real index of wealth in the world other than the land owner.
Speaker 1:Land owner 100. Everybody should be aspiring
Speaker 2:to handle this. Acres. John Malone's got over 2,000,000. Ted Turner's got over 2,000,000. I like the Reed family here at 1,600,000.0.
Speaker 2:Just a
Speaker 1:picture of a bear.
Speaker 2:Just a picture of a bear. I don't know if that do they not have another picture that they could pull from? I have no idea.
Speaker 1:I think I think
Speaker 2:You gotta start the
Speaker 6:land back.
Speaker 1:Should just buy land. Just go buy. Start with an acre. Yeah. Could be anywhere.
Speaker 1:Could be out in the desert. Start with an acre.
Speaker 2:Just add just double it every day for a hundred days and you're good. Yeah. Just double.
Speaker 1:Getting the landowner 100.
Speaker 2:Doomer says how it feels to charge your phone off your MacBook Pro. It's refueling the b 70.
Speaker 1:Really does feel like bomber. Mean, this could have
Speaker 2:been more perfect. So accurate. I mean, a million views makes sense. Yeah.
Speaker 1:It's a perfect analogy. You know, you're you're you got some juice.
Speaker 2:It's fantastic.
Speaker 1:You're sharing it. We're both running
Speaker 2:out of We gotta talk about this one from nine to five Mac. Control center is now darker and much more blurred in iOS 26 beta two. Good news. Fantastic. Tyler, he's been struggling with iOS iOS 26 beta one, but beta two is now available.
Speaker 2:The contrast ratios have been fixed, improved. It's easier to read. We knew Apple would do this. We called this out that any of the criticism that you lever at at Apple's design chops, it's it's not gonna be an issue by the time they launch. Tyler, how has how has daily driving I o six iOS 26 been over the last few weeks?
Speaker 8:I mean, I think overall it's been very good. I I think most of the issues, it's just it's hard to tell if it's because my phone is just so
Speaker 2:old Yeah.
Speaker 1:The phone. IOS.
Speaker 8:Yeah. But I think overall, like, looks really nice. Okay. Yeah. I think it's great.
Speaker 2:But you're yeah. Okay. That's good to hear. That's good. I like Arvin's from
Speaker 1:I would have been surprised if he's like, I haven't been able to open my phone for weeks now. Since Broken brick. Fateful day.
Speaker 2:No. No. No. Yeah. It didn't break.
Speaker 1:Thank you for being a guinea pig for the world.
Speaker 2:Yes. Arvin from Perplexity is building in public. Building a competitor to Bloomberg in public. I love this. He's searching Perplexity.
Speaker 2:How much revenue does Bloomberg make yearly? He's been pretty open about wanting to build, like, a new version of the Bloomberg terminal. And so he's just searching how much money do they make. And he says Bloomberg makes 12,000,000,000 in annual revenue with 10,000,000,000 coming from the terminal. And then he's just like, throw this out there.
Speaker 2:Like, hey. Maybe I'll maybe I'll do this. Like, maybe you should invest in this. It's like Yeah. Yeah.
Speaker 2:It's a it's a hilarious thing to post.
Speaker 1:Well, I've been I've been a recent DAU of Perplexity.
Speaker 2:Oh, yeah? You like
Speaker 1:it? Lot of value out of it.
Speaker 2:I mean, it makes
Speaker 1:sense free on the free tier.
Speaker 2:It makes sense that
Speaker 1:But they're gonna get me soon.
Speaker 2:Yeah. It makes sense that that, an an LLM company, an AI company it's like they either it does feel like there's a fork in the road for perplexity. It's either team up with Apple and become the new default search and figure out something there or go into Bloomberg. Because, like, the the core product very
Speaker 1:different product to try to actually compete with terminal versus displaying data.
Speaker 2:Distribution is just so is so important. I mean, we saw this with DuckDuckGo. Like, a lot of people in Silicon Valley, a lot of power users preferred DuckDuckGo to Google. But DuckDuckGo never got to the scale that Google did just because of distribution. And so, you know, he's clearly looking at Bloomberg.
Speaker 2:There's rumors that he's talking to Apple. It's interesting to see where that goes. But he's cooking. Anyway, Pierre Rishelsen, from cal.com says, I think organizing LAN parties for 30 year olds could unironically be a banger. Warcraft three, Battlefield 1942.
Speaker 2:I played that game. Great. Age of Empires Dota. What else? I don't know why you gotta have thirty thirty year old plus.
Speaker 2:Bring the teal fellows in. Bring the bring the
Speaker 1:They didn't really play those games, John. You're dating yourself. Yeah. But they can
Speaker 2:pick it up. I mean,
Speaker 1:a lot
Speaker 2:of these games
Speaker 1:But it's about the nostalgia.
Speaker 2:Yeah. I went to a hackathon last year put on by a venture capital firm. Was fun. I had some Halo. Halo three was the popular one.
Speaker 2:But the new Halo, if you install it, they they just give you all the previous Halos. You can just play any of them in, like, the in, like, one disc, I guess.
Speaker 1:Because it's, like
Speaker 2:yeah. Buy the current one. It's like SaaS now. You buy the current one. You get you get all the previous all the previous ones.
Speaker 2:Should we close with Cow Tech startup? Yeah. This is exciting.
Speaker 1:Good on r for rock.
Speaker 2:Confirmed one month earlier.
Speaker 1:One month early.
Speaker 2:Cow Tech startup becomes New Zealand's latest unicorn in $100,000,000 fundraising. A rare New Zealand unicorn. I said this was very bullish.
Speaker 1:For beef? For cows. For cows.
Speaker 2:It's bullish. Knee slapper. Knee slapper.
Speaker 1:Knee slapper. Well, we could just keep going all day long here, But
Speaker 2:If you get a text message from Mark Zuckerberg or an email, do not assume that it's fake. He has taken over recruitment for the super super intelligence lab and is reaching out to hundreds of prospects personally. If you respond, this next step is an invitation to dinner. So he's he's in founder mode. He's grinding.
Speaker 2:He's he's putting together the Avengers. It's it's coming together nicely. I'm liking it. Liking it so far. It's fun.
Speaker 2:Anyway, I think that's enough for today. Thank you for watching. Leave us five stars in Apple Podcasts and Spotify, and we will see you tomorrow.
Speaker 1:We love you.
Speaker 2:Have a great
Speaker 1:day. Have a great evening. Bye. Cheers.