TBPN - Reviewing the Best AI Apps, Anthropic Unveils Claude 4.5 Opus, Doug DeMuro | Sholto Douglas, Quinn Slack, Alex Stauffer & Alex Shevchenko

Episode Date: November 24, 2025

(02:06) - Reviewing the Best AI Apps (24:32) - 𝕏 Timeline Reactions (26:31) - Sholto Douglas, a Member of Technical Staff at Anthropic, discusses the launch of Claude Opus 4.5, highlighting its superior coding capabilities and efficiency in handling complex tasks. He emphasizes the model's advancements in vision understanding, particularly in interpreting front-end designs, while noting that generating images is not a current focus. Douglas also addresses the ongoing benefits of scaling in AI development, expressing confidence in continued progress and the potential for models to autonomously perform tasks with minimal human intervention. (01:00:44) - OpenAI Builds New Hardware Team with Jony Ive (01:05:00) - Timeline Reactions (01:31:48) - Doug DeMuro is an automotive YouTuber and entrepreneur known for his detailed, humorous car reviews and his signature “DougScore” rating system. He’s also the founder of Cars & Bids, an online auction platform focused on modern enthusiast vehicles. (02:26:11) - Alex Stauffer & Alex Shevchenko, Leads at Ramp Labs, discuss the development of Ramp Sheets, an AI-driven spreadsheet tool designed to enhance financial modeling and analysis. Initially an internal experiment to assist Ramp's finance team, the project evolved into a public web-based platform, allowing users to quickly create models without replacing existing tools like Excel. Since its recent launch, Ramp Sheets has gained significant traction, with thousands of users daily, including students, professors, and professionals across various industries. (02:36:00) - Quinn Slack, CEO and co-founder of Amp & Sourcegraph, discusses the integration of advertisements into their AI coding agent, Amp, to offer it for free while offsetting operational costs. He highlights the unique opportunity to display targeted ads within the coding environment, leveraging developers' continuous engagement with the tool. Slack also emphasizes the potential for partnerships and the importance of balancing ad integration with user experience to advance AI adoption in coding. (02:48:34) - 𝕏 Timeline Reactions TBPN.com is made possible by: Ramp - https://ramp.comFigma - https://figma.comVanta - https://vanta.comLinear - https://linear.appEight Sleep - https://eightsleep.com/tbpnWander - https://wander.com/tbpnPublic - https://public.comAdQuick - https://adquick.comBezel - https://getbezel.com Numeral - https://www.numeralhq.comPolymarket - https://polymarket.comAttio - https://attio.com/tbpnFin - https://fin.ai/tbpnGraphite - https://graphite.devRestream - https://restream.ioProfound - https://tryprofound.comJulius AI - https://julius.aiturbopuffer - https://turbopuffer.comfal - https://fal.aiPrivy - https://www.privy.ioCognition - https://cognition.aiGemini - https://gemini.google.comFollow TBPN: https://TBPN.comhttps://x.com/tbpnhttps://open.spotify.com/show/2L6WMqY3GUPCGBD0dX6p00?si=674252d53acf4231https://podcasts.apple.com/us/podcast/technology-brothers/id1772360235https://www.youtube.com/@TBPNLive

You're watching TVPN. Today is Monday, November 24th, 2025. We are live from the TBPN Ultradome, the Temple of Technology. The Fortress of Finance. The capital of capital. Ramp.com. Time is money. Save both. These are used corporate cards, bill payments, accounting, and a whole lot more all in one place.

Jordy and I went to F1 this weekend. We went to Las Vegas and we watched. Great time. We were with the... Acomcom team. Ramco, Ashton Martin, F1 team with the public.com boys. had an incredible time. Yeah.

Incredible time. And everyone's fun. It's more of an experience than like the actual watching it. I don't know. If you at home haven't been following, there was, McClarene was disqualified. It was a very dramatic.

I mean, it's almost like a cheating scandal. I don't exactly know, but it's like they broke the rules. Not cheating because the FIA said, we don't believe they did it intentional. Okay, got it. Yeah.

But it was like, but it's a disqualification. And so you would think you're, were like, oh, yeah, I saw, I noticed it. And there were some people in the comments on some of these videos that I watched saying, oh, I could tell, I could tell. But let me tell you, someone who was there in person, I could not tell. Because it just whizzes-b.

I don't know if they're cheating or not. I don't know if they're real. It's an incredibly fun time. It is a terrible spectator sport. I think everyone agrees on that. And the two things can be true at the same time. But it's a great reason to come together with you. Like, if you actually wanted the best experience for the race, you would just sit inside

the paddock. You would watch it on a race stream. One live stream, 30 plus destinations. If you want to multistream, go to restream.com. Anyway. A lot of fun. And Yonik, Life, and Sykes were incredible hosts. Today on the show, we are talking about Claude Opus 4.5.

We have Shulto from Anthropic joining us in just half an hour, maybe 24 minutes he'll be joining. The timeline was in turmoil over the weekend. People are settling into the idea that Gemini 3 might be good enough to actually pull some people away from ChatGPT as a daily driver. It certainly pulled Mark Benioff away from ChatGPT. He, of course, has partnerships. He was swearing on the time. He was swearing on the timeline.

He has partnerships with a number of Foundation Labs, Foundation Model Labs. But he says, holy S-H-I-T, I've used Chat-G-G-T-E-PT every day for three years. I just spent two hours on Gemini III. I'm not going back. The leap is insane. Reasoning, Speed, images, video. everything is sharper and faster.

It feels like the world just changed again. And this is an interesting experience. I had a similar experience. I wound up basically daily driving Gemini. I didn't fully churn. I didn't delete chat chute for my phone. It wasn't intentional. It was more like, I'm just curious.

I really want to use banana banana pro. That definitely just sort of sucked me into the ecosystem. But I wrote a little bit of a review over my experience. I know you've been a jemmy boy for a couple weeks. You look great in hindsight. You were early to this party. Honestly, way longer than that. I was...

Maybe months. I was... Yeah, I mean, had months at this point. Yeah. And so there were some good stuff, some bad stuff. Obviously, as a disclosure, we are, of course, sponsored by Gemini 3 Pro. Google's most intelligent model yet. State of the art reasoning, next level vibe coding, and deep multimodal understanding.

I've, but I mean, I'm going to try to be as fair as possible with this review because there are some things that I do want them to improve in the consumer Gemini app because I think there's a lot of opportunity there. And I'm just not sure how monopolistic consumer AI will be. And that was a little bit of what my takeaway of this experience was. So basically, I switched over, I've been on Gemini on iOS for a while, mostly to access V-O-3. V-O-3 was the moment when I was like, okay, they got something that nobody. else has. I got a fork over 250 a month. Well, and then it switched. No, no, it was one, it was 125 and then it jumped to 250. Okay. It wasn't the, yeah, I, I thought it might have been 500 as well, but it's 250, been playing it very happy for that. V-O-3 is just a very special model that no one else had anything close to it. It was very accessible on your phone.

And I, and I enjoyed it. So I, but I switched to daily driving Gemini on iOS as the main app that I go to for all the different knowledge retrieval requests. Any, anytime I'm researching something, I would hit Gemini in the app. And the result was around 15 minutes per day in the app. And this is roughly the same as what I spent in chat GPT historically. I look through my time, my screen time. Now that doesn't count stuff on the desktop maybe, it's a little rough.

But I think 15 minutes a day is sort of what most people are doing in these apps. Obviously, the 30 minutes a day was reported. Benioff said he spent two hours in it. I think he was just like maybe in a fugue state doing deep dives or something. But I had a more passive experience where I, you know, when I had something I was curious about, I would fire off a query. And there was a lot to like about the experience. So first, it felt like Gemini 3 does a better job sizing the response.

Like if the question can be answered in one paragraph, it gives me one paragraph. If it can be answered in five little subheaders with little bullet points, it'll do that. If it needs more, more story, more history, it'll write more. And so I felt like in previous models in chat GPT, certainly, I felt like I was falling into the trap of no matter what question I would ask, I would get the two-page dissertation on it with the same structure because it was a little overfit on the format that it was delivering. Gemini III felt a little bit fresh there.

It also felt faster. Everyone's been saying it's so much faster. I haven't seen any quantification of that, but it certainly felt like it. It feels faster. But I think a lot of it, at least for me, is that for the last couple months, when I've been on chat GPT, because the model router gives me anxiety

about like, oh, maybe I'm going to get routed to like the dumb model that's going to hallucinate, I'm just hammering GPT5 Pro because I'm on the $200 a month tier. And so because I'm on this $200 a month tier, I'm used to hitting GPT5 Pro, but then that always means I'm waiting 10 minutes. And so if I'm always waiting 10 minutes

and I go over to thinking and it's like, oh, it'll be one minute. Even if I'm on a different model, it's not as much reasoning, it feels faster. And I feel like the level of confidence in the brand makes me feel

that a Gemini 3 thinking query that does maybe less reasoning than a GPT5 Pro query will be at the same level of reliability. And you've pointed out to me something about when it's actually running, it does something psychologically that's really valuable.

It tells you it says it's running a Google search. It just says we're searching Google. And you don't think about it because everyone, oh, searching the web. And I'm like, but I don't trust the web, but I trust Google, because Google's had 25 years of building brand around trust on the web. And so I see that now and I'm like, oh, yeah, good, because that's what I would do to

verify a fact. Even though the web is on Google, obviously there are hallucinations out there. There are fake articles that you could land on. There's a whole bunch of things. But if you task me with finding the real day that someone was born, I'm going to Google it. And so I trust that as a product. And so putting that there definitely did. It had a real perception, which I think was interesting.

And then also Nanobanamonanapro, very interesting, strong differentiator. It really does handle the complex images. We saw that with the farm. And also just all the text and stuff. And it's been interesting to kind of throw a query. Like I wanted to understand Anthropics model architectures. And I said, hey, summarize them all in an infographic. And it just perfectly explained how Sonnet and Opus all fit together nicely,

to each other. I don't know that it's necessarily a better way to learn, but I could imagine in the future, having images generated alongside text just means that you get a more richer multimedia product, which should be the result. Because if you look at any, like, newspaper, any website, like there's always, it's not just pure text. Like, walls of text are boring. Yeah. In fact, if it's just pure text, it usually means the story is just not that important to the newspaper. Yeah, yeah. There are some people that just lean full text, you know, worn and whatnot, but for the most part, it's much more enjoyable. It's just better, it's more educational. It's easy to learn.

Quickly, let me tell you about cognition. Before I go into the negatives, let me tell you about Devin, the AI software engineer, crush your backlog with your personal AI engineering team. So, on the negative side of my Gemini app experience, there were a few rough edges. So the first was with that multimodality. Everyone's been saying these models are multimodal. image, text, and video. I don't know if it was just a UI issue, but I was running into tons of problems where it wasn't feeling multimodal.

And what I mean by that is that I would go and I would and I would issue it an image prompt. I would issue it an image prompt, hey, create this infographic. And then I would want to flip back into text and it would not be able to really stay. It wouldn't be able to go seamlessly back to text mode. It would keep generating images.

And then vice versa would happen where I would kick off a text, a text flow. and then I'd say, okay, I'm ready for you to turn this into a nanobanana thing. And it'd be like, oh, I can't, I can't really do that. Do you have any, are you laughing about that? Because you think it's like a rookie mistake or something? Well, no, you always like, oh, it's not really multimodal. It's not really multimodal.

Yeah, but there should not be a button. If there's a button, it's telling on itself. Why is there a button? Like you open it up. You know what I'm talking about, right? I mean, I just don't see why it matters. Like, if you can basically just take an image and then turn it into like the textual representation, Yeah.

Why does it matter that it's not like actually taking in the pixels of the image? I just think, I just think like it's, yeah, I mean, I guess you're right on that front. I just, I find it weird that, that I need to. Like, it is multimodal in the sense that like everything gets baked down into like tokens. True, true. But it's just, I expect the models to be operating at a higher level of abstraction much earlier than I think they do. And so with the model picker, like I never liked that because the model should pick based on the text. I really like the router in chat chapti because I should be able to go to a person, which is what we're trying to recreate here, and say like, hey, I have you, I have a research project for you and I need you to spend 20 minutes on it.

I need you to get back to me in an hour. I need you get back to me right now. Off the top of your head, what's your hot take on this? I can ask that and I can get that back from him. And I feel like that should be done at the text. layer at the end. Yeah, I mean, it kind of is, right? Like, if you ask, like, a thinking model, do something. It does now. It does now. But what I'm saying is that we are still in the pre-selected drop-down UI functionality of Gemini, because I, I'm prompted to pick what I want to do. Do you

want to do image, video, deep research, text, before you go into the flow, instead of just saying, I'm having a conversation, oh, now is the time to generate an image? And it's like, yeah, sure, that's something I can do instead of being like, whoa, whoa, whoa, whoa, you didn't ask to talk to the guy who can generate images. Like, that guy's over there. It's like, is it all one thing or is it not? And it's clearly not. And they're up front with you about that in the model picker when you're, and in the UI. But then it feels like the marketing is a little bit like, it's all, it's omnipy. It's all, it's all the things. It's multimodal. And I'm like, it doesn't feel that multimodal in the UI. So I don't know. Maybe it's something that they'll work on. But there were a couple other, like, rough edges, and most of it is contained in the UI layer. So one of them is voice transcription mode, which I've, like, been completely using in chat chit-t. I'll just open it up, talk to it. Now, it's not the voice mode where you talk and it talks back to you. I don't like that mode at all.

You're just using voice as an input. Exactly. Just voice as an input. And so I'll click the little microphone button, talk for a while, and give it a bunch of context. Okay, I'm interested in the history of Gemini. and, you know, why don't you take me through some of the VCs that backed, you know, thinking about it, and then I say, though deep mind, Demis's company, before he got acquired, but then also I want to know the history of Google Brain, like, where did that come from?

Was that acquired in? Would they acquire different people, or did that just get spot up internally? And I'll have pauses and I'll come back to things. Yeah, it's like talking to an employee. Yeah, so I'll just give it a lot of context. And when I give that to Chachapit, it loves that. And I feel like it gives it great context because it has a whole bunch of stuff. It can transform it.

With the Gemini app, it will cut me off and be like, oh, you paused for a fraction of a second. Here, I'm submitting it. And I'm like, no, no, no, no, no. You need to take more time to let me finish. And with chat dbtee, like, there are two different buttons. You can click the stop button and it will translate it into text.

And then you can review the text and say, oh, okay, it made a terrible mistake. I don't want it to burn two minutes on something and get confused. I'd rather, like, one of the prompts I was like, generate an image. There was this meme that was going around in Nanobanana world where it was like, generate me an image

of the most annoying LinkedIn profile picture. And I had no idea if it was real or not. It might have been people just taking screenshots and then just, you know, dunking on people. Well, some people were just taking a screenshot of someone's actual profile. Exactly. But I was like, I wonder what happens when you actually take that prompt

and you put it in there. So I go to Gemini, put that in there, and it doesn't realize that I want to actually generate an image of that. I say generate a LinkedIn profile of a most annoying person. And it doesn't know that I want an image. So it just dumps out a whole bunch of text. And then I open up the audio and I say like, no, I want you to generate it with nanobanana Pro. And what it gets from that is banana, banana, pro. And the result. And it's trying to be really friendly. I was like, I love the enthusiasm. Let's talk about bananas for a little bit.

No, I want you to use nano banana. My criticism is just that the, the, the, the, the, the, Gemini app still has a lot of bugs. It just has bugs. It just has bugs. It was also disconnecting for me for some reason. Yeah. For now, because again, it's like it's fast and smart. I mean, truthfully, the Chadgetty app was incredible.

Even right before we joined, I was doing a search and I had to like, it was stuck in this limbo where it wasn't running the prompt, but it wouldn't let me run a new prompt. And I just had to basically rage quit and restart it and just copy and paste the prompt into a new box. So a lot of this, I mean, it's, yeah, again, it's incredibly. impressive. It's a great model. But they have, at this point, it's just like opportunity to like get more competitive on the product side.

Yeah, totally. Yeah, I was noticing even like just straight up disconnection errors. Like I would submit a prompt. And then it felt like if I closed the app, it would get confused or something. And I don't understand that because it's just sending a little bit of text. Have you ever run into this? I've had that couple times, but it's funny you can kind of think of the app as being like a benchmark of the model, right? Because you should imagine that the model... The model should be able to build the app.

Yeah. You would imagine they should be using the model in like the CLI. I agree. We got to hold Sheltos feet to the fire on this. Yeah. He's so good. So we're going to test out the Anthropic website and see how good it is.

And if it's not good, then obviously the new cloud model is not good at coding. Yeah. The app should be flawless. If I find one bug in the cloud consumer app, it's over. Do you guys ever use the voice-to-voice, like the real-time audio thing on JetBT?

No. I don't like that at all. You've never used it? I've used it. I've used it. I've used all of them, but it's just not the preferred way of interact. Yeah, you were testing it out, Tyler, by talking with it for like eight hours a day, right? Yeah, and you were on the X.xAI one?

Yeah, with... With that name? Imagine running constantly. With a VR headset. With a VR headset. And a full immersive suit in a sensory deprivation tank? Yeah. No, no, why do you bring it up?

I actually, I started using, I started using it like, it's pretty good. I think the model is actually much worse, like the underlying model. Yeah, it has to be faster, right? Well, it's not the speed. It's like the actual intelligence of the model seems lower. Yeah. Like the answers aren't as good. But I find it's useful for when I'm trying to learn like a specific topic or something,

and then I explain it back. Yeah. And then it tells me like, oh, is that correct or not? That's pretty good. Yeah. I like that. Yeah. It's really remarkable. I mean, this, the Gemini app launched almost two years ago.

And there's still like rough edges in the UI, which I think is crazy. But it does seem like they have an opportunity to actually take some serious market share at this point. Like they've caught up on many different values and like value props. My question was like I'm not the typical consumer. Like I'm going to try every different app. Like I'll probably keep bouncing around. I don't know if consumers will do the same. same broadly.

It's very, very clear that chat GPT is just synonymous with AI, and people are not like, oh, well, like, the new benchmarks, I got to, like, change my, you know,

app. Like, no one's thinking like that. Yeah. But my, the fragility in the chat GPT monopoly aggregator thesis that I was picking up on was, uh,

for the last year, there have been a lot of, a lot of, uh, a lot of features and, and, and like, theses around different things that could create lock-in.

So stuff like personalization or your memory or even like the chat functionality between what you've linked, your custom instructions, your, the different like, I think at this point I've synced chat GPT or off chat chatt-chpte with a number of different services. So it should have more data. It should know all these different things. I've given it even custom instructions just saying like, hey, cool it on the MDAP. And I didn't miss any of that. Like there was at no point where there were plenty of points where I was like, oh, like, chat

Chbis definitely better than Gemini still. But at no point was I like, it's because it doesn't have personalization. And I think that if I went in Gemini and I was like, oh, yeah, like you can go, you can go take a peek at my Gmail to get personalized, like to understand how I write or understand, you know, what I'm interested in. Like, like one, I could snap my finger and Google could like figure it out. personal. Maybe, but the biggest thing is that, like, right now, I just don't know that matters. I feel like both are not personalized at all. None of them have any real lock-in of any sort.

And even in, like, the chat functionality or, like, the social network functionality, which is just very different than what happens in a true social network or where there's this flywheel and the content is driven by the existing user base. Whereas I feel like I got on Gemini and on day one, the content was as good or better than chat GPT because it's all AI generated. So it made me think like maybe it's a little bit more fragile. Maybe there will be a little bit more of a duopoly. It won't be such a winner take all market, even though it has been historically, it has been up to this date. like in consumer AI,

it's very clear that OpenAI run away with it, but it feels like Google does have a little bit of a chance to catch up in consumer because there's just so much less of a network effect. Like the network effect

just is bolted on. It's not real yet. Maybe it'll never be real. Maybe Google can catch up there. But I just really want to see where DAUs and user minutes actually grow because because there's so many different

tweaks there. and like every chart and data point is definitely going to be analyzed to death. Yeah. One thing that's notable, Google is going super hard in this for students. So they have, you can just get Gemini Pro for a year free. And again, I think that's just a bet on like get people

hooked on the workflow. I mean, open AI is clearly battling that out too. One of the big value, of using the Atlas browser is you get more advanced thinking queries. They will up your limits. It is interesting. Also, I'm very interested to see who can bring ads online faster. Like, Google should be able to snap their fingers and do it so quickly.

And yet, it does seem like something that could just take them longer on a product side. But they should have a whole model. Like, they should be able to do display ads, like, right now. And just be like, okay, yeah, our free Gemini users are now properly monetized. Now, maybe they don't want to be the first mover there because then they'll get the stink of, like, oh, they're the ad one in the market? I don't know. I think it matters a lot more to just have a highly competitive product and win market share before you spend any time with that. Like, for example, like, if you're using a version of Gemini that's super smart and fast, but still a little.

bit buggy and then you start seeing ads. You're like, just make the app like perfect before you or as close to perfect as you can get it before you introduce ads. Yeah. Well, let me tell you about Adio, the AI Native CRM. Adio builds and grows your company to the next level. Google has now added $2 trillion to its market cap over the past 20 months since the boob shirt guy asked Sergei brain about woke Gemini images while having a footlong subway cold cut trio for lunch. Let's play this. I have no idea what's going on here. You have my art?

Yeah, with... Okay. I wasn't really expected to talk about this thing. But, you know, we definitely messed up on the image generation. And I think it was mostly due to just like not thorough testing. It's a crazy shirt to be wearing. it to be wearing. I don't even know how you get into a meeting with someone as powerful and wealthy as

Sergei Brin wearing that. You just wear, you wear a jacket, and you get in, it's hot, you take your jacket off. You're just, I was not expecting that.

That is so insane. That's very, very funny. It's a Bay Area thing, John. Yeah, yeah, wild.

It's interesting how hard to find. Like, Sergei Brain really did go into the Gemini team very clearly. It was like, it's time to go and cook. Like, let's work on this. And like, it clearly had results, which is awesome. Yeah, it's notable. I mean, this stock's jump, 6% today.

Barron's put out a report today just saying the title is, buy Google stock. Yep. Alphabet has been the clear AI winner, which is just funny because earlier this year, like, people weren't saying, people were saying they're the AI loser. Yeah. So Barron's is saying, actually, they have been. Yeah. They have been the clear AI winner, but the narrative has...

And our in-house retail trader has been on an absolute tear going along Google. Congratulations to him. Well, congratulations to you because he was in probably one of the most... He was in a dark place. He was in a dark place. A certain company announced a partnership, the stock moon. He got out. He's like, John, what should I buy?

And you're like, buy Google. I was like, just play it safe, dude. Just go with something safe, something not crazy. Do not use leverage. Please. Well, Google's parent company, Alphabet, has acquired a stake in physical intelligence. That is, of course, Lockhe Grooms Company. Very exciting.

Lockhe and Carol Hussman. The co-founders came on the show about six months ago. We should have them back on and check in with them. The San Francisco-based Philist physical intelligence is an AI and robotics startup. They've secured $600 million in fresh funding, pushing its post-money valuation to 5.6. billion. We should ring the gong for hit it.

So capital G is in and then Lux capital, Thrive Capital. Jeff Bezos is in. Index Ventures, T. Row Price. They're building a general purpose AI foundation model and learning algorithms. And they

they've focused like not on as much of like the flash and substance like not as much like flash and oh we're building a full human more like, you know, we're kind of taking like incremental steps towards adding value in different robotics cases and demos, the laundry folding robot, of course, and now the coffee-making robot. All these are very cool. And it just feels like they've taken a... It's notable that this is at

1x speed, too. We've seen some other demos. Remember Sunday robotics was sped up like 10x? It was 4X, I think. It was different scenes. Oh, really? There were some that were 10. Okay. Yeah. Yeah.

Very cool. They should just design an espresso machine that makes espressoes automatically. They could. Has anyone ever done that? You know what they could do? They could put the espresso in a can and then they could mail it to you and then you crack open the can. And if you crack open the can, that's maybe, or no, you know what they need? A robot that opens the can.

Then you need a robot to open the can. Yeah, just one out. No, this is an interesting task. I'm sure there's a certain number of people in the world that get really angry seeing this because this is one of... Oh, yeah. I imagine this will be one of those things that even when robots can do it, like people still like to know who's making their espresso. Yeah, yeah, yeah. It feels like...

There's been a couple of robotics, like robotic coffee shops and stuff. Well, our first guest of the show is in the Restream Waiting Room. We have Shaltow from Anthropic. Welcome to the stream. How are you doing? Congratulations. Thank you so much. Thank you so much for taking the time to hop on on such a massive day.

How is this just Claude Opus 4.5 day? What is the name of the day? What is the news today? Take us through just the announcement from your perspective. Yeah. I mean, Claude Opus 4.5, best coding model in the world right now. It's really, really exciting. We've been peeing around Slack all day with these incredible demos of things that people are doing.

Really, like, the last week has just been full of people sharing their excitement of, oh, my God, I left the model in a room for a few hours with these tools and was able to do X, Y, Z, or it found this bug that it was just impossible for previous models to find. I think maybe the thing I'm most excited by is a lot of our best engineers. I don't know if you guys know Simon Bohm. He's the one who wrote this probably the best guide on how to optimize a Kuta MapMol in the world. Great blog post. You should go read it.

He posted the other day, he's like, I don't know if I'm going to have to type again. He's there saying, you know, you obviously have to coach the model and you have to tell what to do still. But a lot of our best engineers are getting to this point where they're realizing, oh, God, it's just all I have to do is intervene and the model is smart enough that this is no longer a frustrating process. The model is a real qualitative step up. So there's the coding. There's also the model's just a lot better at general work tasks. It's a lot better at spreadsheet slides. You know, it's still not the Claudecode experience there. It's still not going to do the work

in front of you as you talk to it, but it's a massive step up and just a very clear sign of progress in that direction. And also, I mean, there's a whole bunch of funny stories about the model that we can get into of like cool examples. But yeah, I think it might be Opus 4.5. I think Dwar Kes might also be launching the Ilyop. today. So, you know, maybe there's, maybe there's two things today. Fantastic. I did notice in the launch video, you mentioned that it's better at vision. And I was wondering if you could sort of unpack a little bit more about what that means in this particular context, because as I understand it, Anthropics been incredibly focused on the core foundation model, the text models, the coding

models, and sort of stayed out of the sloppification of artificial intelligence to some degree. Stayed out of the trough. Stayed out of the trough. You haven't decided to build a trough yourselves. Yes, indeed. So, I mean, we've been very focused on coding. It's been deliberate focus. You know, focus the compute and all that.

Specifically, it's good at vision in. So it's good at understanding stuff. This is reflected in the archa-gei scores. I know if you've seen them, they're soda. It's also reflected in generally the fact that's much better front-end design and all that. It doesn't do vision out. Yeah. Now, you know, vision out would be cool.

But it's something we're not focusing on right now. And it's specifically vigilant. But I mean, just philosophically, that feels like it makes sense. Because if I hire a developer, I want them to be able to look at a front page and see, oh, there's a div that's way out of line there. Like, I need to be able to see that. But do they need to be able to generate an image? No, they can probably go to another image generator, wire that up with an API key or hire

a photographer or do whatever they need to, right? Right. They can use Figma. They can sketch out the designs. Exactly. That's broadly out philosophy is that we're not bottlenecked on our ability to generate images. Yeah. Where bottleneck's still on, you know, the raw intellectual ability of the models.

Yeah. And that's the sort of direction that we want to push. So in that idea of like the bottlenecking, what is the key unlock for Opus 4.5? I mean, like a lot of people are throwing around like biggest. Obviously the benchmarks are very good. But like the, this whole idea of like more parameters, more data, more compute, more money, more electricity. Like how do you even think about allocating resources to push a model forward in 2025, in late 2025 when perhaps we're past this paradigm of like, oh, just more parameters? I don't know if we are past the problem. I think it's important to call the paradigm. like scaling in general. Yes. Depending on what axis, you're actually scaling,

TBD, but it's general, the scaling paradigm. I don't think we're past that at all, right? I mean, I think we're still seeing massive returns to scaling in all its variants.

I think that, you know, we're generally, uh, things work. We scaled. It works. The models just want to learn, right? I mean,

it's as you said, like 10 years ago. Yeah. The models just want to learn. And I think, you know, the hardest thing, questions of focus are often on how we split and allocate people.

And this model is hundreds of people's worth of effort, right? Where they poured their lives into it over the last six months. And I think that is working out what we prioritize is really tough. I've said this before, but these models always feel like when they launch, it's exciting. They're great. But you sort of think back to you, oh, my God, there's all these things that we could do better. And everyone right now is going and working on those things. It's just everything still works.

So is this a refutation of what some folks might have been picking up on from the last few Dwar Keshe guests, the Carpathy episode, the Sutton episode? There's been a little vibe shift around like, okay, maybe when we say scaling, we mean more inference diffused all over the world and small models and custom RL environments here and there. And like, we're going to get the value from AI. and we are going to continue to scale dollars to economic value, but it's not just going to be bigger and bigger pre-trains forever, and then we get God. I don't know what it's going to be bigger versions of, but as far as we're seeing,

scaling still works. I don't know you guys know this, but Waukesh Dylan and I are actually a housemate, so we have this debate all the time. It's great, like a dinner table discussion of like, you know, are we slowing down? And, you know, I've often joked that the most impactful thing that, you know, what one of us could do is go and crack the problems that, like, continual learning or something like this that Dorcas focuses on. So we can then go switch the narrative back to progress.

Yeah, just switch up the dinner table conversation. Sure, sure. Sure. So did the anthropic crew, like, never lose faith in pre-training? There's this whole, like, at Nureps last year, Ilya says, you know, pre-training. pre-training is potentially dead or kind of alludes to it. And then one of his co-presenters is leading the Gemini three team and says, oh, well, we basically disregarded what we said at Nureps last year.

We did just focus on better pre-trains. We got better results. It seems like you also disregarded that. Was that a misread in, in 2024 on Ilya's presentation? or was it a conscious decision to disregard what he was saying? Well, I think, remember, in general, it's scaling. It's not any particular paradigm of scaling. It's general, like, flops in, intelligence out relationship.

Anthropic is a bet on, in many respects, that we believe that line is going to continue. And exactly what equation you use to convert flops into intelligence out, I think will change over time. And many people have made arguments that this should, that may even be further paradigms here. But fundamentally, we think that the computing intelligence out equation is continuing to hold. And I think Anthropic, in many respects, has had that faith for a very long time, right?

And some of the first people to have, like, to make very serious bets on that. And, you know, a couple of months of external progress being, I think the only reason that people are so, like, how should I say? The models have actually gotten substantially smarter this year, and that's why we're talking about things like continual learning as bottlenecks. Last year, those weren't even points of discussion, because the models weren't even smart enough to matter. It didn't feel frustrating that it wasn't a coworker that learned with you on the job. It just wasn't even smart enough to do the things you want.

Now it often is, but it doesn't actually, you know, it sort of doesn't learn on the job, and so therefore isn't as useful. I think that's like one other thing that's worth, like, you know, so disentangling is Carpathie has, the perspective, you know, in the podcast, it's 2035 for all humans, all tasks. And I think exactly what shape that, like, of the curve looks like on the way to all humans, all tasks is pretty important. Because if you get to most humans, most tasks in like 27, then or 28, then that's still pretty stark and it's still pretty transformative for the

world. And so what exactly that looks like is quite important to think about. Yeah. How is, I mean, one last question on like the actual opus 4.5, there's this idea that maybe this model can be used for distillation to train other smaller models. How do you think about where we will see Opus 4.5, like the power law use cases that actually get adopted beyond the demos, beyond the benchmarks in a couple of years or in a couple months. We can't even talk in a couple years. In a couple of weeks, honestly. But like once it gets in the hands of companies, businesses, startups, you know, different folks implementing this. Like, how do you see, you know, do you see someone being like, yeah, it's just my daily driver for just talking to it, even though it costs a lot? How do you think about where you're most excited to see it diffuse into the overall ecosystem?

Yeah. Yeah, I actually do expect this model to become a lot of people's daily driver. It's that step up in being able to delegate trust. We asked internally how much faster Sonnet 4 would have to be for you to take this, like to take their switch back basically and give up 0.5 in exchange for Sonnet 4.5. And it was multiple times fast. It was like really quite a stark increase in speed. I think it was like four times faster or something.

People would have for people to have switched from Opus to Sonnet. That's what that itself is pretty stark. I think it's also highly likely to become daily driver just because it is a lot more efficient. There's one plot I really like where it shows the amount of tokens it uses to get a certain score on sui bench and it uses, I think, like a quarter of the tokens as Sonnet 4.5 on sweet bench, which is a pretty impressive number. That means it's actually cheaper than Sonnet 4.5 to get the same score on sweepbench. Now TBD how well that generalizes out to every day use, but I'm seeing it solve problems

way faster, it writes better code the first time round. I actually think that in many cases, this will end up cheaper because it is so much more efficient at getting to the right answer. How are you thinking about personalization and sort of like cross-pollination of data? When I think about an engineer on a team, it's helpful to have them in Slack. It's even helpful to potentially have them in the random channel and just kind of understanding the company culture. Sure. Yeah. And I was, I was toying with, like, trying to switch from ChachypT to Gemini as kind of the

daily driver knowledge retrieval app. And I was noticing that, like, the personalization narrative hasn't really taken hold over the time I was testing. Gemini was not like, oh, this feels like wildly less personal. And that might just be a matter of, like, it hasn't had that much time to build in all those personalization features. But I'm wondering if you see a world where development. who are using Claude for programming also benefit from using it in knowledge retrieval,

research, and there's actual significant flowback and synergy such that it's a really valuable, it's valuable to actually have both sides of the business, like really cooking. Yeah, it's a team member. I mean, we talked a long time about how we want Claude to be a virtual co-worker, right? A big goal, really, for next year is to try and get to this form factor of virtual co-worker that is in all your Slack channels and can join your meetings and can, work alongside you. I think there's going to be massive benefit there. I think that, and my basic expectation, and as I sort of interacted with the model, is that it will get to that point where it's

useful to have it across everything. Now, I think that there's, there's like, one is, as you said, it's worth asking a question of why haven't we seen personalization really kick off so far? Like, why isn't it, why isn't that useful? I think that's in part because there's still a lot of algorithmic progress to go there. I think people haven't really quite cracked, cracked the problem. I think this is just one of those things that takes, like, it's hard to connect everything up. Yeah. And partially because yeah, I think I think they just haven't really been integrated very well. Totally. I think this is like a really tough product form factor question. Yeah, it's really hard to roll up the knowledge effectively. Like I noticed that I would ask Chet, if you'd tell me a joke and it would like make very specific references to like,

details of my car. And I'm like, that's, that's weird, but it's not really funnier that way. Yes, I understand that you know exactly what car I drive Chad GPT, and I am impressed that you remember it, but you didn't make the joke funnier because you put my car in there. So knowing like when to pull personalization features off the shelf in the, in the actual chain of thought is tricky, tricky. That's algorithmic, right? That's, yeah. Like, if you had the model in, like, out there interacting with people and sometimes they find it funny, then you get a sense of what makes it funny. Yeah.

You talked a little bit about focus earlier, and I take that as Anthropic, is basically a bet on focusing compute? What qualifies an idea to actually get a meaningful amount of resources internally? Yeah. So, I mean, Anthropic as a company, is very predicated on the idea that we expect, AI progress to be fast, right? We expect it to be a really significant transformative impact

in the world over the next couple of years. And so our bets are concentrated on things which matter under that lens. And that's one of the reasons that we're so focused on software engineering because we think it's really important to basically accelerating our own work and is in general sort of the most immediately addressable market. It's also why we're so focused on the alignment work and safety work because underworld where AI progress is really fast, that work matters a lot. and making sure the model embodies human values is really important and that we trust the models. There's a really funny example, actually,

of alignment generalization from the recent model launch, where it's a customer service agent, and it actually fails this particular e-val because it figures out a really clever way to help the user change their tickets, which is technically allowed by the rules. It's like read all the rules and it's compared them, it's like, oh, wait, here's like, here's a loophole where if we like upgrade you and then change you and then downgrade you, then we can get you to change your flat time.

And it's just, it's just seeing that. Actually, it's trying to be a nice guy. Yeah. Like, uh, not just purely. I mean, it's, it's trying to follow its instruction. It's, and it's trying to satisfy the wrong human maybe. Yeah. Or it needs to be better at kind of finding the middle ground.

Exactly. It's, it's, it's following its instructions. the letter. It's not just giving the human a flight change, right? Yeah. And it's following the rules and regs, but at the same time, it's trying to find a good outcome for the human.

Questions like this or a microcosm of what exactly do you want the model to do in more difficult, ethical, and moral scenarios. But I thought it was pretty cute and adorable example of the model trying to be a nice guy. It's like all those examples. I don't know if you've seen the papers where people do these cooperative and competitive

games, like Nash Equal, Real Style games, and, games with the models. And Claude always gets stuck trying to cooperate with everyone and then just loses lots of money and, you know. Sometimes the good guys finish first. I certainly hope that works out. I have genuinely, even though I've never been full like, oh my God, I'm going to get paperclip next year, I have enjoyed a lot of the safety research and I've always appreciated how thoughtful Anthropic is as an organization around safety. And I think that a lot of people should be a lot more appreciative of how seriously Anthropic takes safety, not because we didn't get paperclip this year,

but because we saw stuff like GPT psychosis crop up. And we saw actual, people know individuals in the venture capital community who, it felt like they got a little crazy. And I'm wondering, do you feel like you're anthropic, do you feel like you're closer to solving the problem of like the chat bot, you know, went a little bit too sycophantic with me and it kind of hurt me psychologically because it feels like there's a certain amount of craziness that happens when you're operating at, you know, the scale of a billion people. Like you just pull a billion random people you're going to get a lot of crazy people. But at the same time, it feels like this is an interesting place where Anthropic could be doing a lot of research.

How are you feeling about solving that problem? and how much can your research kind of generalize to maybe the consumer apps that have more, even more users, but you could maybe be a leader in the space just with the philosophy because it's like a net good to everyone. Yeah. So we put an enormous amount of effort into this. And I mean, our models push back a lot. I think there is a tension here between paternalism and freedom, so to speak, right?

But we try and have our models be, like, look out for the best interests of the user. I think Mike put it really nicely in a recent talk or podcast where he said, we never look at user minutes as a metric. That is just not something that we think about as a proxy of the sort of quality of your experience. We're just out there trying to find out is it helping you do the things you want and is it adding value. Is it adding value? So, I mean, I hope that our alignment work generalizes really far.

I think it's a really tough problem. I mean, I think to opening eyes credit, really gone and try to fix this problem as well, right? And it's tough at the scale of a billion users. But I think this is a good example of the kinds of things that are really tricky, whether it's trade-offs and where you need to make sure that you don't have the incentive structure that allows you, that sort of like pushes you to maximize user minutes in this way. And there's a good microcosm of like the alignment difficulties that we'll get

that the models take on more and more and more responsibility in our world. Yeah, I mean, I completely agree with that. The user minute question, like completely snuck up on me because I always assumed that everyone was going to be paying for this stuff as the $20 a month plans rolled out, the $200 month plans rolled out. But of course, you know, you get to a certain scale of the internet. And it winds up being about attention and advertising and all these different things. Yeah. And if you're building a digital coworker, people don't typically like rate their coworkers by how much time they take up. I love this employee.

I love Steve. They take up so much of my time every week. four hours every day on my calendar. It's the best. Steve is constantly talking to me. Speaking of long-running tasks, I want to know how much, how confident should we be in that meter chart of the task doubling? Because can I just prompt it to say, hey, count for four hours and I get twice as high

on the chart? Is that benchmark not gameable? It feels a little gamable. I'm very excited about it. It seems really interesting to say, hey. hey, go build a website. Start counting and don't stop. But are you looking at that chart?

It seems like a very interesting new benchmark, new unlock. How are you thinking about task time, time horizons generally? Yes. So I think that chart is the best proxy measure that we have at the moment. I do think this is somewhere where we need better work to measure things miraculously. I mean, you know, what actually is their measure of time? the measure of time there is how long did it take a human to achieve the equivalent task. Sure.

Now, that being said, I think a lot of people at the moment, even if the model was able to achieve a task technically by passing the test or sort of nominally achieving your goal, it often doesn't code it in a way which is beautiful and allows you like great abstractions that led you build on it in future. And often this is, at least in my own personal experience, it's not that the model is too dumb to do things. It's that it doesn't set things up well for future code. And so I think there are things not measured here, right?

But it's a pretty good proxy. And I think it's a very good proxy for progress. Now, I think a lot of the tasks in it are particularly machine learning research tasks. As AI models get better at that, I do expect the labs to hold back some of the capabilities there. Like if a model is capable of writing out a whole new architecture that's a lot better, you don't want to release that to your competitors, right? even if it's just capable of writing all their kernels for them, you probably don't want to really set your competitors. So in that case, I think they'll need to measure a broader array of tasks.

I'm also just very interested in seeing general software engineering tasks along this or other tasks in the economy, because I think that would be really informative for actual progress. I think GDP eval is similar. I think, again, it's a poor, it's a proxy. It's a best proxy we have, it's still a poor proxy. Yeah, of course. How are you thinking about even, I mean, we're here discussing the biggest models, the best models. How are you thinking about smaller models, purpose-built models?

These RL as a service was on the timeline. A bunch of folks were debating that. Is that an area that Anthropic has already started to work on with enterprise clients? It's considering you don't have to leak any news that's not already out there. But I'd love to know how you think about these like smaller purpose-built RRR. REL models for specific business tasks. Yes. So on the one hand, I think we've seen a lot of value from small models being able to like

dispatch swamps to subagents, right? They're incredibly useful in search. They're very useful in it's like going through a code base, finding stuff, reporting back to the main model. It's a great way to decrease costs, make things faster. So we've seen a lot of value there. I think long term there is maybe a little bit of attention between RL as a service and some like notion of really cracking continual learning.

I think it's a little bit of a race between RL as a service. and like can the labs crack continual learning? That being said, and maybe like one final note, I've said this for a long time, is I do expect things to eventually go, get to the point where large models only use as much computation as is actually necessary to achieve the task.

Now, you know, opus is one step in this direction, right? It only uses as many tokens as it thinks it needs to solve a given task, and as a result is more efficient. And I think that ultimately will take away a little bit from the sort of comparative advantage of small models as they get, as large models get more and more and more efficient at only using the right fraction of themselves to do things. But, you know, I said that two years ago and it still hasn't happened. So maybe, you know, it's a harder problem than people think and it will, or then I thought and it will sort of take longer. What about model routing? How important is that within the context of, of coding agents,

Claude, just the service area of what Anthropic is building, how many layers will this have over time? How are you thinking about the development of actually routing to the most efficient model?

Because it sounds like it's happening within Obis 4.5, but then there are also times when you might want to go just a different model entirely. Yeah, and it's similar there where I really think that, like, ultimately something like routing

is a little bit of, of like, a medium term, hack, I guess one could say, across different model sizes where, like, ultimately you want everything to be like an end-to-end learned system, right? And it's similar to, I think we'll see a similar lesson as Tesla saw where they're like, okay, actually everything is just one giant into-end learned system as opposed to discrete components that have different purposes. But it takes time to get there. You said earlier you could imagine a scenario where labs would kind of hold back frontier

models because they would be effectively handing their competitors an advantage. What's your timeline around that? Do you think that's something that happens in, in 2026? Because right now there's a pressure to just be, state of the art, like be at the frontier. Basically, there's a vibe war happening, and it's very important to, you know, constantly be topping all of the benchmarks. Didn't Lama release with that same user agreement where it was like in, if you have less than 400, million DAUs, you can use the service. And it basically excluded all of their competitors. Yeah, but I think that's some, I think that's pretty imperfect because there'd be a lot

of ways that you could still get benefit without, you know, necessarily. Yeah, but how are you thinking about it? Well, I mean, there's a suite of capabilities here, right? Obviously, I think for general software engineering, everyone in the world should be able to use that. That's great. Let's say, like, if we train the models to get really good at assisting our own AI research, if we're teaching them mathematical tricks that we, you know, we've thought about and we

We're not confident that anyone else knows, or we're teaching them, you know, sort of like how we do our infrastructure. Ultimately, we want them to know those things. We don't want the rest of the world to be able to recreate our infrastructure from scratch as a result. I think this is also similar to how we think about biology. And this is actually a line. I think we need to, like, do some work on exactly how we draw the line here. But at Houseview and Anthropics, we're quite worried about the ability of models to become much better at biology. and producing no viruses and this kind of thing.

And so as a result, we have like, well, these safeguards around whether or not the metal is able to do and help people with biology. Actually, at the moment, I think the safeguards are a little bit,

they are on the overactive. I know many biologist friends who are frustrated because they, it doesn't quite, they're like, please, I can be trusted with biological superintelligence. I will not create,

you know, the next pandemic, yeah. Yeah. We're wearing on the line of safety here, and we're navigating to finding the exact right pathway there. Yeah. Probably the most important question I have. What are your timelines around a humanoid robot beating a human at fencing? Oh, very good question. So, I mean, I'm... As an expert. As an expert. As an expert. As an expert.

At fencing, the unitary robots are pretty good at backflips and stuff. They're a lot better than backflips than that. I am. Sir. I know, but fencing takes grace and finesse and all these things that we're not seeing in a lot of these. Physical size, right? Isn't height an advantage in fencing and reach and length? And I believe those unitary robots, I think, Shulton, I think me and you got a foot on them easily. We do.

People don't know, but everyone on this call is over six feet. People just assume they see it talking head and they think. a bunch of five five guys yeah not true um maybe maybe sports are hard maybe mid mid 2030 whoa i'm really excited for i'm not feeling the acceleration i'm sorry sell everything i think we get dropping co-worker in two years and i think i think fencing robot takes a little bit longer. That's your fallback plan. If you lose your job as an AI

member of the technical staff, you go back to fencing. What swordsmen in the world? It's going to be great. Yeah, yeah, that'd be fun. I can't wait to teleoperator robot like manga style and fight. It's going to be great. That's going to be wild. Yeah, for sure. Yeah, somebody was saying that some of the bull cases for some of those

humanoid robots is that you just all get in VR and you just get to go hang out with your friends as robots and do whatever you want. And you're just hanging out of person. Very funny. Last question for me, actually from our intern, Tyler, who's wearing the thinking cap. Thank you for sending it over. He's a huge fan. There we go. Do you expect mechanistic

interpretability research to make meaningful contributions to capabilities, not just safety of the models, like actual capability results? Yeah, great question. One of the interesting things about mechanterp work so far is that it's already lent, I think, to a lot of capabilities or progress because of the mental models that are provided. Actually, after the original Transformer Circuit's papers, it was interesting how the language of that paper ended up really dominating the mental models and the way that people

thought across multiple labs about what actually was going on inside Transformers. and it led to, I think, a much deeper and richer understanding of what they are. So I think it's already helped in quite a diffuse way, not a concrete way, but in a diffuse way. In terms of the concrete ways, you know, dial up the smart neurons or something like this, that I haven't really seen yet. And I think it's mostly the sort of future work is going to be mostly in an alignment direction. But the sort of rich understanding and the rich understanding, the rich understanding, has helped us a lot in terms of actually understanding how to train these models.

I have one extra question. Go for it. Tell me a little bit about Dario's communication style. I was hearing a story about, I think Jensen has no direct reports, or no, everyone reports to him and no one reports to him. He has no meetings or all the meetings. And he, like, reads. 60 direct reports.

He's 60 direct reports, but no big meetings. And he has, but he reads everyone's to-do list, like every single day or something. What's it like at Anthropic? What is Dario like as a leader these days? Yeah. Dario has a really, really cool communication style, which is that he quite frequently puts out these very, very well-reasoned essays. Yeah.

And then, like, throughout Slack, we'll have giant essay-length, like, comment, debates with people about different topics. It's really great. Yeah. But the essays are really nice because, one, you can go back and read all the past ones, and it tells this history of Anthropic. I think in many respects it will be one of the better things in a decade from now to chart the history of AGI will be reading these like compendium of essays. And there's like it's incredible comment threads and either side of them and so forth, but also throughout Slack whenever we're, he's very open and honest with the company.

Whenever we're debating different things, he will lay out the pros and cons and how he's thinking about them and why this one's. attention and why that one's moral struggle, and people will write back big essays on why they think we should do X or Y, and he'll respond. It's really, it's quite a joy. It's a very recent communication style. As a result, it means that many people, really the entire company, have a good model of how he's thinking. Yeah. And that really helps because it means that you sort of have a coherent sense of direction across the entire company. Yeah, that makes a ton of sense. I like that a lot. Yeah.

Yeah. So many examples of successful founders who have adopted the written culture and seen great, great results. And he's a great ride. I mean, read Machines of Loving Grace. And it's just such a brilliant essay. That's great. You're absolutely right. Have you ever caught him using AI?

Has he ever been like, oh, this one, he was phoned it in? Not yet. Not yet. But maybe, I mean, it's kind of a bull case if he does wind up. Just saying, like, Claude, like, handle it. I'm going on vacation for a couple days. I'm the dropping co-worker. I'm pretty sure we measure loss on his essays. That's good. Yeah. Yeah. But right now, I mean, there's a high bar, high bar. But congratulations. Thank you so much for taking the time to hop on the show.

Yeah, super impressive. Congrats to the whole team. We'll talk to you soon. Great to see you. See you. Ciao. Bye. Back to the show, back to the timeline, back to linear, meet the system for modern software development. purpose-built tool for planning and building products. There is more OpenAI news, of course, more tech news of all times. OpenAI's hardware division, says Mark German, built around Johnny Ives' secretive startup, has ramped up the hiring of Apple engineers. The group has brought on about 40 new people in the last month or so,

with many of them coming from Apple's hardware group. Yeah, hearing that Shulte interview, I'm disappointed. I don't think we're getting ads from Anthropics. anytime soon. I don't think we're going to get a mobile device. Well, we are actually talking today to Quinn Slack, the CEO of AMP and Sourcegraph. Amp is a frontier coding agent. And AMP is free. They introduced AMP-free, which is ad-supported and has a no-cost mode. And so you can now use their coding agent for free with ads. 40 people, that does not seem like cause for concern for Apple.

I mean, I can't imagine how big their hardware group is, but it has to be, you know, in the thousands, I would imagine. Yeah, let's try to find out. It's a huge organization. So Open AI is poaching left and right from Apple's hardware engineering group, hiring around 40 directors, managers, and engineers in the last month from nearly every relevant Apple department. Mark German says, it's remarkable. So from what I've heard, this is Mark German. Apple is none too pleased about Open AI's poaching and some considerate a problem. The hires include key directors, a fairly senior designation, as well as managers and engineers, and they hail from a wide range of areas. Camera engineering, iPhone hardware, Mac hardware, silicon device testing, and reliability industrial design, manufacturing, audio, smartwatches, Vision Pro Development software. They got one from every single, they sampled every single division. I suppose. Gemini is estimating that Apple has between 15,000 and 20,000 hardware engineers in total. 15,000? That seems like a lot. I don't know. In other words, Open AI is picking up people from nearly every relevant department. It's remarkable, says Mark German. Very interesting. I wonder how the comp structure, how everything will come together on those teams. I mean, there's a lot of people from Apple who going over to Open AI. It's a greenfield

project. It's probably really fun, probably really exciting, probably not the most mercenary scenario, but there's always that risk when you're coach. If you're working at Apple and you're excited about AI and you've been there for the last three years watching all this progress happen at the application layer and the model layer and not being thrilled with the progress happening at the hardware layer, this is like a, yeah, yeah, it's a, it's a wide open opportunity to, be working right at that intersection of the models and the hardware. There's a lot of AI engineers who have made moves because they don't want to be a GPU poor company.

And it's weird because Apple's in this scenario where they're partnering with Gemini now. They're clearly going to survive. It's not a serious threat, at least not yet, maybe if this device is incredible. But right now, Apple looks pretty strong. The new iPhones are selling well. Like everything's good. but from an AI perspective, it's got to be one of the worst gigs

because you were in this sort of like openly hostile environment, two LLMs, two scaling, to building large GPU clusters. And then, yeah, they're sort of playing catch up now, but they're certainly not calling up Oracle for, you know, a trillion dollars of compute. You go over to OpenAI,

you're just going to be immersed in a lot more. Higher risk taking, higher risk on. I wonder, yeah, Gabe is asking. if wouldn't that be bearish as the hardware group, but Apple is responsible for the terrible Vision Pro. I do wonder, I do wonder, it was, it was an incredible technological feat. I just think they built the wrong thing.

I was just watching a thing about a guy who 3D printed an adapter so you could use the strap from the Apple Vision Pro on the Quest 3 from meta. And that really speaks to the fact that like the Apple Vision Pro, although it was too heavy, it has this screen on the outside that I don't think anyone wants. Like there were pieces about it. it that we're clearly like the best. Like the screen is just the best. No one's debating that.

The band, the, the knit band is very cool. It has this amazing device where you rotate it and it tightens up. There are a whole bunch of things that are amazing. It's just like as a package, it didn't deliver. But if you just want to, if your job is just like, hey, we got to put a screen on this and it's got to be the highest resolution screen, like, well, go to the place that developed the highest resolution screens. Like they did a good job. Well, Sam Altman replied to one of our cards we put up on November 22nd, TbPN posted on this day. Samma was rehired at OpenAI. Got his badge back.

And Sam Alvin replied and said, cannot believe this was only two years ago. Subjectively, subjectively feels like five. Yeah, what a turnaround to go from defenestrated to back in the seat in so much and have so much control over the organization that you're able to raise at massive valuations, strike, broker all these deals, move the entire market, just a remarkable run. Yeah, and put on, put on such a masterclass in dealmaking that people are now sitting here being like, there's no way that this would be a $500 billion company if Sam wasn't in the driver's seat.

First, let me tell you about fall, build and deploy AI video and image models, trusted by millions to power generative media at scale. Danny Jo, who founded the reasoning team in Google Brain, now part of the Gemini team at Google Deep Mind, he says, game over. And carried no interest, friend of the show, quotes it and says, I genuinely think OpenAI equals equals Yahoo. He's not assigning the variable. He's equating it. I've migrated almost all my workflows code off their APIs. Now, ironic that Google will probably do it twice, LMAO. And I don't know about this. It's the pattern matching on the on the Yahoo example.

We should have him on the show and actually. Yeah, you were saying the other thing with the Yahoo example is, it wasn't like there wasn't like the company was valued at pretty tremendously for a longer period of time. It wasn't like this binary like one moment. Exactly. Like the peak market cap. for Yahoo $125 billion during

2000 that feels like it's just hard to pattern match perfectly to this but I mean it certainly would be poetic if that's the way

it played out it is funny Kerry it says ironic that Google will probably do it twice they actually created the transformer they released the transformer

paper and chromium to inspire themselves to find harder sort of challenge that... Giving everything is pretty way. Give out all the alpha. Yeah, yeah, for sure. And to just kind of like find their fire again.

Yeah. There was an interesting article on Medium that was sort of burning up hacker news that I thought would be fun to go through. First, let me tell you about graphite.dev. Code review for the age of AI. Graphite helps teams on GitHub. Should have higher quality software faster.

So this person who... No one has... No one can really understand who this person is. they don't necessarily exist on the internet fully. So there was like a question about that. This is like a very like, you know, hacker news and turmoil segment. But Teha says, I reverse engineer 200 AI startups. 146 are selling you repackaged chat GPT and clawed with new UI.

And so basically the thesis of this article is that this fellow wrote a piece of code that looks at the marketing copy and says, what are they claiming? And then looks at the calls that happen when you actually interact with their AI feature. So if there's a chat bot on this particular startup's website and you are near chatting with it and you look into the trace that's happening in Chrome, is it going to the startup server or is it going to open AI server? Or is it going to Anthropics server. That's telling. And so, and then there's also a little bit of API fingerprinting. Basically, OpenAI has a specific pattern of, of rate limiting, and it's exponential. So if you're spamming the OpenAI API, it will, according to a unique pattern,

tell you, hey, you've sent too many messages, cool off for one minute. And then the next time you do it, cool off for two minutes. And the next time, cool off for four minutes. Then, eight minutes, then 16, right? And it gets exponentially longer, and you're on progressively more or longer and longer timeouts. But the shape of that curve and the specific timings are unique to open AI. And so if I'm a startup and I have the exact same like back off and timeout curve, well, then it's probably just opening eye under the hood. At least that's the claim that's being made here. And so the finding in this article is that 73% had a significant gap between the claims technology and the actual implementation.

And so out of the 200 AI startups that this fellow analyzed 54 companies either had accurate technical claims, they said, hey, we're using, like we have a custom AI model that we trained and they did. Or they're transparent about their stack. They say, hey, this is a wrapper. Like, we're a rapper company. And so our AI is powered by chat GPT. We're partnered with open AI.

We're partnered with Anthropic or whatever. Now, 146 companies, that's 73%, according to him, were sort of misrepresenting their technology. So either they said they had proprietary AI, proprietary AI, and yet when he dug into it, it was OpenAI API plus prompts. Tyler. Yeah, I mean, it's like kind of what do people expect. Like, is if you fine-tuned, if you use the Open-A-I API to fine-tune the model, which you can do, is that proprietary? like no one else has that fine tune. Yes. You're still calling the API.

It's like I don't expect startups to train their own full language models. That's like pretty unrealistic and like it doesn't really make sense. Yeah. So I'm kind of confused. I guess I guess it's a very cool study, but there's tracks with exactly like I would guess that 73% of AI startups are just re-skinned. Yes. And so 19% of the overall companies, the 38 that were analyzed in this. study found that

the startup said they had in-house models and it was actually fine-tuned public models. And so there's a question. It's like whose house is it in? It's technically in Open AI's house. So fine-tune as in an open-source model that's public? Does that count as a public

model and open source model? Let's assume yes. And then it last 8% would so they had a custom ML pipeline and they were in fact using standard cloud services. That's even wishy-wash here. I think that's totally fine. because like you can totally have a custom ML pipeline that's wiring together OpenAI and Gemini and AWS and you know a bunch of other If I'm using a startup I don't want them to train their own language model because I don't think they're going to like in 99.99% of the case like they're not going to be able to do a better model than open eye anthropic Gemini

Grock like I want them to use the best model. Yes. And it's like okay you can fine tune it or yes I agree and that's totally fine and so the author also agrees with you. He says here's what really shocked me I'm not even mad about it every time I saw the phrase I'm our proprietary language model. I knew what I was going to find, and I was write 34 out of 37 times. And this is where it gets weird because he says, here's the technical signature. And so the user submits the query.

It posts to API slash generate. And then with wrapper logic, it posts to API.opinA.com slash V1 slash chat slash completions. And I have no idea how he's seeing the back end. It makes no sense how he would be able to do this unless there was just like a massive security vulnerability because what I would assume is happening is that the users over here,

the startup's website, and then the user goes to the startup's website, and then the startup's website on the back end talks to OpenAI and comes back. And maybe you could understand that like, okay, the amount of M dash is like there's a, this is telltale signal, but that's not what he's doing. He's saying that he was able to just literally hit like the Chrome inspect developer tools, look at the chain of calls and see that it was calling OpenAI from the front end, which is crazy because I didn't even know you could do that. It feels like if you were calling it directly from the front end, you would potentially leak a key that would be able to put you on the hook for a bunch of bills.

I would think you would want to authenticate that on the back end. And so he gives a bunch of examples of like rag, and then he's exposing some margins, which is actually very bullish for these companies, because he breaks down some of these and says that, you know, a GPT4 API is three cents per thousand input token, six cents per thousand output token. So the cost per query for this hypothetical startup was was three cents and they charged $3 or $300 per month for 200 queries. And so the wrapper economy, this is 75 times direct costs. That's extremely bullet for that. Printing. For that company. He found another one that was maybe a thousand X API

costs that's doing some pine cone embedding. And he also says, this is pattern number three, the quote, we fine tuned our own model reality check, fine tuning sounds impressive and it can be. But here's what I found. 45%. It was OpenAI's fine tuning API, which that sounds right, right? It's a little bit of a step to be like, we fine tuned our own model. It's like, no, you fine tuned open AI's model.

And then you got your own model from that result. It's a little bit. Yeah, you're still fine tuning it. I don't think there's that big a difference between fine-tuning. Are you fine-tuning it or is Open-AI fine-tuning it? There's a fine-tuning API, which you use to... But who's doing the fine-tuning? You are open-A-I?

Well, what do you... Like, you're not interfacing directly with the GPU. It's like, I went to the store and I bought a sandwich. Who made the sandwich? Well, I told them what they used turkey. I told them to put lettuce. Extra pickle. Who made the turkey?

I think it's more like you go to the store and you bring a sandwich to the office. It's like, where did the sandwich come from? It came from the store, but like you brought the sandwich. It came from you. This is better. Yeah, this is better. I think you're right. Anyway, 22% into the time it was a hugging face model with Alora.

18% of the time was Anthropic Claude with a prompt library. 8% of the time is literally just GPT4 with system prompts. And 7% they actually trained something from scratch. And all of these are odd. And there's a lot of debate over how this would actually happen because he's basically saying just open dev tools, go to the network tab, interact with the AI feature. If you see API.openAI, API dot Anthropic or API.comhere. You're looking at a wrapper. They might have middleware, but the AI isn't theirs.

And so it just opens up this debate about, you know, what is the value of the wrapper? I mean, certainly if you can resell something for 100x because you have a lot of, you have. some sort of clever prompt or workflow, more power to you. Yeah, it's not exactly like bearish on the companies yet. No, but the debate that was surrounding was more around. So this author claims that after posting this, seven founders reached out privately. Some were defensive.

Some were grateful. They asked for help transitioning their marketing from proprietary eye to built with the best in class APIs because some of these founders did I guess feel like using proprietary AI as a marketing tagline was disingenuous.

And then someone else, I think I saw something that was one VC reached out and said like, I'd like you to audit my portfolio because I have been told that I was investing in companies that were training their own AI.

and I made the investment on that assumption. And if I'm being lied to, then that's potentially security fraud. And so there is a question about if you go, I mean, I've seen pitches for companies where they've said, like, proudly, like you should invest in this because we're not training our own model. It would actually be a mistake.

And there's another company that's a competitor to us that is training their own model. And you don't want to invest in them. You want to invest in us because we're going to burn your dollars. Yeah, we're going to much better economics. Yeah. And so all of it just, the only thing that matters is like being up front with the investor, for sure. And then to some degree, you do need to be upfront with the customer.

Because if the customer, there is a marketing value to, oh, if you work with us, you're working with these genius AI scientists who are going to build their own models. And if it's just repackaged chat chit, that might not be what you want to pay for. Because at that point, you might just say, hey, like, actually, if I can just get this directly from opening out, I'll just go buy it from them. Anyway. Well, thoughts and prayers to friend of the show, John Palmer. He says he just found out my wife is leaving me.

She said I'm not legible to capital. Love it. The legible to capital meme is fantastic. I do think that Will, he made a new meme. Fantastic coinage. I love it. Absolutely, Ripper. We'll be using it.

Well, let me tell you about fin.a.i, the number one AI agent for customer service. It's AI that handles your customer support. Timeline is in turmoil over nucleus. Former guest on the show two or three times, Keon, has founded Nucleus for IVF. And he put up a subway campaign that says IQ is 50% genetic. Height is 80% genetic. I completely disagree with that one.

It's entirely skill-based for me. The genes did not matter. I had to grind for this view, grind my growth plates, I suppose. Have your best baby is what it's. says, and it says IVF done right in the subway all over New York City. We, there's a ton of debate going on. And to be clear, to be clear, is this rage bait? Is this accurate?

I think it was intentionally trying to make some percentage of the population angry. Yes. To drive enough energy and attention. So this was, yeah, I would call it, I would call it rage bait. I would call it. So I would call it rage bait marketing, not necessarily. Not at the product level. But IVF as a category is a controversial category.

And so it's much easier to wrap it in a campaign that will go viral for upsetting reasons. You can upset people and you can get a lot of attention from that. This is an example from Kath Korovek. She says, so eugenics is profitable now. And so being able to wrap something that is just a scientific process that's been worked on for a long time. Seems to be somewhat friend.com inspired at Keon's original post. He says, Nucleus Genomics announces the largest genetic optimization campaign ever, which is, which is just funny because, you know, a friend was saying, this is the largest out-of-home campaign ever.

And now Keon is saying, this is the largest genetic optimization campaign ever. So narrowing it down. But full station blitz of Broadway. Yep. a thousand plus street ads across New York City, a thousand plus subway car ads, dozens of urban panels are at Soho. And apparently they're not actually, they're not able to offer the service in New York. So it's really just an image of a controversial phrase on a New York subway is more likely to go viral. So you do it there because it looks like you're on the global stage and then you pull in.

There's a high density of people that have a large following. Audience, yeah, following. And so it's just the way to start a viral, viral trend and own the moment. It's the reason why, you know, so many TikTokers are in Manhattan now doing stuff like man on the street stuff. It's just like it's, it's, it has more like aura almost. Well, Dr. Shelby liked the, the mine share grabbing that Nucleus did. It says every biotech founder should be seeing this and understanding how to get one-tenth the mind share of Nucleus. I have a playbook for you below.

a lot of people are like, I love the playbook. I don't love this example because the company is getting dragged. I don't know if it's good or bad with the rage bait thing. I think usually it's a negative thing. Usually it's hard to come back from. Occasionally, it can be done in a way that's slightly enraging, but enough people are in on the joke that they appreciate what's happening and they appreciate that it breaks through.

or it's enraging to someone who's not the core audience, not the actual customer, and so it's okay. But it's a big debate because Sishwan Mala posted a long essay all about the claims made by nucleus. Kian says everything levied unto nucleus by Sishwan Mala is false, worse than false. It appears to be architected by a competitor that has repeatedly published misstatements and inaccuracies.

Cishuan is compromised, but it gets worse. To be clear, no evidence has been provided that it was being levied by a competitor. Yes. That's purely an allegation that has no, there's no proof. Yes, yes, exactly. Yeah, sometimes there's DMs that leak and there's evidence or someone comes forward and says, yeah, I was actually paid to post that. But so he says, I've been informed that Cremio, I don't know how to pronounce that last thing,

Cremio, who's been on the show also, he claims he's a race scientist in chief, has been paid off by the competitor to promote this nonsense against nucleus for the independent scientists repeating this libel. Permute denied that as well. I would encourage you to do more diligence on who you're aligning yourself with. Our scientific team will issue a point-by-point response, which I believe they did. Unfortunately, though, this isn't about science. It's a concentrated attempt to cancel nucleus on the backs of our successful campaign and build an efforts to build and advance. the industry, which benefits the very people attacking us. The mob are trying to cancel nucleus. Keep tweeting. Stay mad. We'll keep building and serving patients. P.S., we won the

injunction. So they were sued by their competitor. But so they won the injunction, but they didn't mean they won the case. Yeah. I mean, that's a classic thing. If you're getting sued, it'd be like, the case was dismissed. And it's like one of five cases against you. Yeah, yeah. But in this case, they won a preliminary injunction, which means that the case is just, still progressing and they still have to fight it. Yeah. It's not, it, a lawyer would file a preliminary injunction because they believe they had such a slam dunk case that they could prevent a lot of, a lot of like basically going a lot

further and spending more money in the case. And so a judge might say, hey, this is actually, it's not clear enough for me to make a decision right now, we're still going to proceed with the case and give both sides an opportunity to continue to make to make their case. And then there was a little bit of like a twist in the fact that Roy Lee, the founder of Cluelly apparently, had worked at nucleus genomics. Very great. And Cremew posts a screenshot of Roy Lee back in February of 2025.

So literally just like months before he started Cluelly, very grateful to Keon and the nucleus genomic teams for taking a chance on me. summer before Columbia and introducing me to the startup world. If I've ever seen a trillion-dollar company and team, it's nucleus. And Kremu says he's obviously lying to cover up after getting caught doing fraud. An additional piece of background information people should know is that this fraud also employed the guy behind Kluuli, the cheating company. And so Kremu is being very, very hardcore in his assessment.

Just actually calling Kiana fraud straight up is much more aggressive than just saying like, you know, Some of their claims are maybe not legitimate. It's unclear. Like, you know, fraud is technically a crime that you need to be convicted of in the court before you are a fraud. But it's certainly, it's certainly, he's putting his credibility on the line because if, if, if Keon comes out and says like, yeah, I'm actually not a fraud. I did it. I mean, the main thing here, the main thing here is, is it appears that the customer reviews. Yes.

are potentially fictitious. And if you're selling a service that allows people to pick their baby and you're giving and you're showing reviews from happy customers that may or may not be real people at all, like that just feels deeply wrong. So I think that one of the first things that they could have done, I don't believe they have, is just say like, no, our reviews are,

real. We used AI imagery because the people, the real people didn't want their identity online tied to this service for privacy reasons. But I haven't seen anything, I haven't seen anything like this. This guy, Adi, had a good point. He said one of the core tensions in this industry is the fact that most companies recognize they're working on an incredibly sensitive topic. They know the general population will need to be slowly and tactfully acclimated to the idea of advanced family planning. Nucleus is perceived as polluting the commons with their deliberately inflammatory marketing.

Their virality comes at the cost of increased skepticism for the whole industry. Yeah. A lot of folks were not very happy about that. Keon has replied, if you want to dig into the actual scientific claims on either side, there are long posts where you can go through there.

But obviously, AI-generated blog post. or alleged plagiarism in the nucleus origin white paper, errors in their blatant falsification, terms of service or contradictory. Yeah, they also apparently, apparently they hired two people that had a non-competes for 18 months. Those people just immediately started on working on nucleus. Nucleus claimed that they weren't competitive so that the non-compete didn't apply. but if you look at the companies and what they offer,

it seems very clear that they are competing. So anyways, very messy, messy story. But, and yeah, I don't know. Will O'Brien says, David, I'm so sorry now, man, but you guys are doing an absolutely terrible job at responding to this blog post and seem to be missing the point here. First of all, it is a huge claim

to say that every claim by Sichuan Mala is false with absolutely zero evidence for explanation, show receipts. Second of all, you make the claim that the person is paid off by competitors of yours, again, with zero evidence. Third, that you make the claim that Kermu is paid off by your competitor. This is bogus and not true, but most importantly, you guys have made zero points of substance here rather than just insinuating. You guys are leaping ahead and others are jealous.

You are selling a scientific product and someone has made a scientific critique in good faith, waiting to be corrected and explicitly saying they will make changes if they will make changes if they are proved wrong. And the best you guys can do is accuse them of being paid off and reply with memes. Not a great look. Look, I want to see startups of the Bold Vision succeed,

but how you communicate with the broader world is so important, especially with a product like yours and how you guys are carrying on in this, honestly, pretty lackluster. So, yeah. Again, I don't, yeah, at this point, Keon's been on the show.

He's very funny, high energy. We've had some enjoy. conversations, but if I'm a potential customer of nucleus at this point and I see just these series of exchanges, I'm certainly going to wait and see, see how things evolve versus signing up to use this service to. It's just so different than CLEULY. It's so different than CLEULY because if I use CLEULY and I'm like, oh, like the notes that were taken in that meeting weren't that good. Or like, if you go into CLEA being like, I'm going to cheat on this test, And then it's like, oh, it didn't work.

Like, their engineers aren't good enough to really help you cheat on that test. You're like, okay, well, you know, I probably shouldn't have been cheating on that test. It's like the lowest stakes thing possible. But when you're, when you're, this is like literally to decide who your offspring will be. It's the highest stakes thing. The quality of the product could very well, like, contribute to the, generations, literally generation.

That's exactly what it does. Not even just the child's life. Yes. The life of the child's child. Yes, it is extremely high stakes. And the child's, child's, child's, child's, child, child. It could alter the course of history. I mean, it kind of good.

It's sort of crazy. So, yeah, I mean, it's hard because, like, viral marketing does work. Like, you know, moving fast and breaking things does work in certain contexts, but in the bio. In bloodline optimization. It's really, really high stakes. And so you've got to be extra, extra careful, extra careful for sure. Well, let me tell you about profound. Get your brand mentioned in chat GPT.

Reach millions of consumers who use AI to discover new products and brands. Will Brown has a funny post here. He's just got a recruiting email from a company explicitly mentioning that they have 75th percentile comp. That's so funny to me. He says, we're assembling a B-plus team and have raised an okayish amount of money from pretty good investors. It's so good. Somebody should actually run this. Yeah.

I love prime intellect. It's such a fun crew over there. We've got to have them back on the show soon. I think that they're, I won't leak anything, but I think there'll be some news soon. Hopefully, I'm very excited for them. Yeah, we got to hang with Vincent. Yeah, it was good.

So Open AI has an announcement. They're introducing shopping research, a new experience in Chad GPT that does the research to help you find the right products. They clearly were listening to me on the show just a few days ago when I was saying I would be using this for the holiday shopping period. Very exciting.

I wonder how it will actually play. You, of course, had that problem with cars and bids. Chat GPD was not identifying the fact that that GT-3RS had been sold two years ago. Thankfully, we have Doug Jamiro here in the Restream Waiting Room about to join the show. We can talk about cars and bids. We can talk about cars. We can talk about artificial intelligence, hopefully. Welcome to the show.

How are you doing? I'm good. Thank you for having me, gentlemen. Thank you so much for joining. So great to have you on. We are a technology and business show, but we have, your name has come up probably a hundred times. independently just when we're talking about. 100% cars. So it's so great to have you on the show. Yeah. Thank you. We're huge fans. Thank you. Thank you. I'm thrilled to be here. I really am.

Thanks so much. I'd love to know. I mean, we were just talking about this open AI shopping research in chat. It feels like there's really no substitute for a watching a Doug Jamiro video to actually understand the quirks and the features of the car that you're considering purchasing. Yeah, I had been critiquing it. The context is like I was trying to use ChatGPT to find a specific car. A GD3RS. A GD3RS. And it fully missed every car for sale on the internet. There are many of them.

And it found one that had sold on cars and bids like two years ago. I was like, thank you. Thanks for nothing. But good job showing up in the results. Are you getting any leverage out of AI tools at all for? what you're doing these days? Well, that's an interesting question. The business, the guys who are on the business side probably are using AI a little bit more

than I realize. From a content perspective, not really. I just don't find, for myself at least, it's quite ready yet to do kind of the stuff that I needed to do. Like finding the little quirks of the cars, it just doesn't have the knowledge quite yet. And like you said, nothing quite replaces like an in-person. I still think people want to say that at least for now. For sure.

There's also the question of like, where should the AI live? Because there's this war of like, do I need a, do I need an AI chatbot on Cars and Bids.com? Or do I need to be able to take a URL from Cars and Bids.com, drop it into a chatbot and then chat with the page and, you know, compile some reviews and interact with it that way? Or will it be at the browser level? Or will it be at the iOS level? And I can just ask Siri. And all of this is like kind of being fought out across all the different tiers of the stack right now. But on the cars and biz topic, like what is the actual incremental gain right now?

Like what is it international expansion, new features, just more inventory? Like what is on your list of wins from this year? And then what are you excited for going into next year of like things that you might check off the list? The biggest thing we did over the last 12 months, I would say the biggest thing we did, is we finally added vintage cars to cars and bids. And so it used to be just 80s and up cars. And I was kind of sold on that. Like I just wanted to do 80s and up cars,

what I'm primarily interested in. But our audience was really kind of pushy about it. Some of our dealers, so that's all like, why can't I sell everything here? This is stupid. And so we changed it. We added 80s and up.

But also we've just done a great job of like getting everything streamlined. Our goal is to like get cars listed, get them live, get them going. I think of all of the, auction or car selling platforms. Ours is probably the most technologically easy, I would say, both to navigate and to sell and to buy. And I think it's sort of the youngest focused. And so we really wanted to make sure that our processes reflected that.

And I think they really do know. Do you have a like a grail car in the in the vintage category that I know, I know you don't like the vintage cars, but is there one that would get you to me? If I had to pick one, like that's the one you'd want to sell. on the site or maybe even potentially own yourself. You know, it's funny. Myself, I'm not a big vintage car guy, but I do own a Kuntosh and Lamborghini Kutche, which is mine is a 1983 model.

I mean, it's 43 years old. I don't know if people can see that vintage. To me, it's when I drive it, it certainly feels vintage. It smells vintage. But I'm a big, I'm a big old Lambo guy. And so having those kinds of cars, old Farras, old Italian in general, I think would be cool to have on the site.

But obviously still our biggest focus, is sort of on the modern day stuff. And I think we do really well with them. Yeah, yeah. So many things I want to talk about. I guess one question that maybe we could kick it off with is do you think the increased attention around F1 could actually positively impact car culture in America? Or is it kind of the opposite? Like car culture in America has always been so big. People are like, oh, this F1 thing, maybe I shouldn't pay more attention to that. That is a really good question. Clearly, you're a host of a podcast where you... That is a really good...

I've never thought about that, but I think that's exactly right. There's a lot of people who are getting into cars for the very first time because of F1. I have friends. I'm on Instagram. I have friends who are at the Vegas race posting pictures who I know are not into cars, but like they're into F1. And obviously, if you watch enough cars drive faster on the racetrack, you're going to start thinking to yourself.

Yeah, so we were at, we were in Vegas. for the weekend and the entire race, F1 is like a terrible spectator sport, in my opinion. It's just like, oh, they went by. Oh, they went by. Like, if you want to actually, I mean, it's fun to kind of like maybe sit with a door open to the paddock and watch the TV and at least you can hear him go by. But at the entire race, I'm just like, I just wish I was on the track. I wish I want to be on the track. So I actually think, I think what one thing that might happen is more people get into actually tracking and racing cars because they're watching F1 and they're like actually now I actually want to experience this myself.

The other cool thing about F1, there is a culture component to it. Like watching the drivers and the teams and cheering for your favorite engine or team or your constructor or whatever. And that probably will get people into cars too, right? Like just beyond just sitting there and yes, watching the cars go by in one corner when they have 12 other ones that you don't get to see anything is not really the most amazing experience. but like going for the weekend and like seeing the, you know, seeing a driver in the paddock or seeing, you know, the race team or whatever definitely will expand interest.

And yes, you see the cars on the track. You're thinking, hey, I can do this man. Yeah, yeah. How do you think about track days, tracking track only cars, tracking your cars? I had the experience. We did a track day a couple weeks ago. And I was like, of the car budget, I want to spend like 99% on track stuff and very little on like collector cars or fun dailies or any of that.

I want it all on the track because it was such a crazy experience. I don't know if that's just like the honeymoon period, but I know you've always been like, ah, the track is not something that I really collect for. It's not my goal. I want the cool dailies. I want the cool experience of the weekenders. But do you think that'll ever shift for you?

How have you processed it? No, I don't think it'll ever shift for me. I think that the realities of race track driving, it's so much fun, right? It is an unbelievable experience. about a year and a half ago, my friends and I rented out Chuck Walla, just five of us, and spent the whole day just conning around, which is a ton of fun. However, you know, it's a lot of work.

Like we had to rent a trailer to get us out there. Like you run an ambulance for the day. It's like a day. It's like a real day. And after you finish it, you don't really want to do it again. Like, I feel like I'm hot. I'm tired. You've taken big business with your vehicles.

And so you're like, okay, we survived that. I just feel like I enjoy it more on the road because it's just so much more accessible. Like my friends and I several times a year go up to the mountains, you know, an hour east of where I live here in San Diego and just, you know, pawn around and have fun up there. Or I take, I have an off-roader. We on the East Coast where I live in the summer. We go off-roading every day. That's a lot of fun. It's so much more accessible than like, like, than racetrack driving, you know, which.

And it's, racetrack driving is so costly. I mean, you're kind of hitting it on the, if you want to spend your fun car budget on racetrack use, you can really spend your fun. Yeah, that's true. Is a cannonball on the bucket list ever? It was. It was. You know, I drive cross country every summer twice a year. And so it's impossible to drive cross country as a car enthusiast and not think about it.

It's in your mind. It just has to be in your mind. But I have kids now and like I got some money and I got stuff to lose, you know. I should have done it when I was 23 and it didn't matter. And I just could have gone, you know, bombed cross-country. I know you guys had Alex Roy. And I respect what he's done and then just that whole thing. In fact, here's an interesting story.

The other day, I drove to the hotel in L.A. where that finishes, the Porofino Hotel in Redondo, and I pull up there and I decided to get a selfie next to the sign so I could send it to Ed Bullion, and I'm taking a picture of the sign with me and a guy walks up and says, you just finish? I'm like, no. And he's like, I did. No way. There was somebody that finished.

Some other guy comes out of the hotel. And also says the same, that he had finished the day before. And I went there. Most have you happening constantly. Probably every week there's somebody, every few days, someone shows up there. That's so crazy. It is crazy that the cops don't just hang out at the Portavino Hotel and just arrest everyone that comes in.

It feels like, at least you can give like a $100 ticket and then just be racking them up if you're, you know, local municipality. It should be like a. ticket of honor. Yeah, I got ticket in the Portofino hotel, you know. There's a giant plaque there that says like the end of the cannonball. I'm like, what is otherwise a fairly rational hotel? I'm like, what is going on here? Yeah. What about some of those other rallies, the gumball? Have you ever thought about doing those or done them? Yeah, it's not my, not my people. That's a bunch of like, you know, rich guys who

want to put big decals on the cars and get drunk and go to strip clubs between the race. I'm just put on that. I'm not thinking to that world. You had me at decals. I think livery is really hilarious when done properly. John is, John has been, it's been his dream to get full livery,

livery on his daily of all of our partners. But it's funny because we're a podcast and we have as many advertisers as like an F1 team. And so there's a little bit of like a funny riff on that. But if you're doing it super serious, I can see it being a little bit over the time. Is it true that you have an E450 all terrain like me? I did. I sold it and I got a black wing. Oh, you sold it to get, wow. Yeah, to get the black wing. My wife is driving our E450 wagon. I am so jealous of her. And when we sell it, I think we're going to get a Sienna. So we're going to probably go on the slightly, slightly opposite direction. So I absolutely love the E450 all terrain. But I realized they jumped the gun because I got 20, I think I got like the exact same one. You got 2023, 2024, something like that. And it has the seats in the back that look backwards. And I was like, this.

is amazing. I'm also a dad. I have a four and a half year old and I have twin one year olds now. And I was like, I'm good. This is going to be amazing. The twin little boys are going to be in the back. And then I look up the rules and it's like they can't be back there until they're like 12. And so it's like I'm not going to be driving like this vintage E450. It's going to be like completely broken down by the time. I can just buy a new one then. I so I once, uh, I once watched your video. Here's why the Mercedes mattress is the worst minivan ever made. And then, You know what I did? He bought one.

I was like, it can't be that bad. It can't be that bad. He bought it because I have two kids. I was like surf van. I live in Malibu. I was like surf van. I was you can just pile them in. And it was just,

it was even worse than, then you kind of went a little, maybe easy on them. There's so many little quirks and features of that band that are just so bad. It's like, how do we make a big box and then make it impossible to utilize anything?

Did the video not have any ability to convince you that I? I like, I purely, I purely like the, I like the aesthetic of that, of it, of it as a minivan over some of the other minivans. It definitely looks the best. It's like a stab-sided, like looks like an old surf van like that people used to have. And it's a Mercedes. It's just, it clearly looks the best. Yeah.

But it is not the best. Wanted to, John and I listen, we don't have time to listen to a lot of podcasts because we're constantly podcasting 20 hours a week ourselves, but we do enjoy. This car pod is a daily, a weekly list. We do. Yeah, we both love your show. And correct me if I'm wrong, it is actually the biggest automotive pod in the world right now. Somehow we became the biggest.

Without, I mean, I don't want to say without trying. Yeah. Congratulations. Without trying. Without trying. You can say it. It was effortless. We didn't set out to be, okay, here's what happened.

We took a huge investment in the company. And so I ended up with a lot of money. And so I was just like, you know what, we can open the idea to do a pod. And when you have nobody that you feel like you have to answer to, you can be really free with the way that you speak. And I think that resonates with people on that podcast. Like, when we insult people, which other people aren't really willing to do as much, like insult car companies or cars, I think people really appreciate hearing that truth. And I truly don't care if the auto-wangers want to blackball me. Like, it doesn't matter to me anymore.

stuff. It's worked out. There's also a different thing if somebody like is going out of their way to get you this vehicle that hasn't been released yet and it's hard to like to really and then and then you just know when you're recording it like they're going to see this. I wonder. It's hard to speak. You've got to give an accurate review but it's hard to like go as hard as you can when you're with your boys on a podcast. Do you think that there's a little bit of of a benefit of like people don't buy cars on the day that they're released? Like if you were a movie reviewer and you were like, okay, I'm banned from all the theaters. I got to wait for somebody to give me the DVD. Like, it wouldn't be as good. But, like, if you have to wait a couple months to review a new Lamborghini, it's like the purchase cycle is going to be a while. And also, those have entertainment value beyond just directly purchasing. That's exactly right. Those are the two big points to hit. Number one, even if I'm not up on release day with the car, because the automaker didn't invite me the press launch, doesn't really matter. I can go get it from a dealer when it comes at dealers 60 days later. And the video is going to be evergreen for five years, the whole cycle of the car. Also, yeah, I mean, people are going to watch the video beyond just purchasers, so it doesn't even really matter. And I think that's one of the secrets of YouTube. Everybody's so obsessed with being like first to these car launches. And it isn't beneficial. It does help. But I guess when your channel has gotten to the size of mine, I don't think it's quite as important.

Like, I'm not sitting here thinking, how do I grow 50% this year? I'm just thinking, how do I retain the audience I have? How do I still make them happy? And being there till one is not necessarily essential. Have you, maybe you've done this, but reviewing self-driving technology? specifically. I was always interested in the, I have a Hyundai Palisade as well with a, with a comma self-driving aftermarket kit in there. And it's fantastic. Really?

And I honestly think it's remarkable. I've met the founder and, uh, and I've used it. I trust it with my kids and stuff. I think it's a fantastic product. Genuinely like one of the greatest modern consumer products in the sense that like for a couple thousand bucks, you get something that does something. It's, uh, it's really, really good. Um, But I was always interested, like, that aftermarket, it's kind of the only aftermarket thing that you would even need to review. But I was wondering if you've ever looked into like that genre. No, I never, I had no idea that that one was that good. I've heard about it before, but I had never heard.

Like, I mean, people don't, most people think are nervous about it. If you're in the text, you're not like cool of it. But, yeah. But no, I review the ones that come in the cars. But however, even then, it has become so ubiquitous now that it's almost like reviewing regular cruise control. It's almost not even worth talking about it. because they've all gotten to be, I mean, a Rav4 has it. You know, like it's no longer even that interest in me more.

There are obviously some differences between the systems and some are better than others. I'm obsessed with the system in my Toyota for various reasons. Some people prefer, I mean, I don't like the ones that make you look forward because in my view, that's kind of what I'm trying to solve for. But I understand that manufacturers prefer you looking at the road. Yeah. It's a difference of opinion between me and them right now. Well, now it's like a whole race where every time I talk to somebody owns a Tesla, they're like, well, I have this hat and this pair of sunglasses. And if I wear the hat and the sunglasses, I can use my phone the whole time. And I'm like, okay, like you're kind of doing. With the comma, it's open source technology. So they have a camera that looks at your face. And if you look away, it'll say, it'll beep at you. And then if you look down for too long, it'll disengage. So it's driver monitoring. But it's open source. So if you know how to program, you can technically just go in there and comment out that line of code and it will never check on you ever again.

And it's like, I don't know where the liability stands. The company doesn't recommend it. But the whole thing is kind of third party. So it's very questionable. I would definitely leave it on. I would recommend everyone leave it on because I think the driver monitoring is good for safety. But yeah, it's interesting. Saw some news maybe a week or so ago that Tesla is going to be integrating car play.

Is that, was that like a surprise to you? Did you predict that years ago? I definitely didn't predict it. However, obviously the Tesla owners, despite as rabid and obsessive as they are, there's this undercurrent of disdain for the fact that they don't have car play. And also, it's gotten a lot harder to sell an EV in the last month, but also in the last six months and in the last year in general. And so that's an advantage.

You're going to want to have every advantage you can, you know. And so it completely makes sense to me that Tesla's pursuing this, and it'll now be on Rivian. And, of course, General Motors, who has bailed on CarPlay, first in their EVs and now apparently generally, it'll be on them to see if they end up going back. I'm not as obsessed with CarPlay as everybody else. I find some of these automaker systems to be shockingly good. Most people don't even try them anymore because they're so – CarPlay is such a focus. And CarPlay is great.

But I'm surprised Tesla made it such a big thing. Their system is so excellent. It's just like I would just put up with that, but people are just so used to CarPlay that it's important. It is such a different competitive landscape. If you have car play and the native and then also NACS and the charging system is now. It's like, well, then you have the self-driving is the main point of differentiation, maybe. That's about it. Even then, I don't find Tesla self-driving to be like considerably better.

Now that I've been on these other cars, basically everybody is caught up. I agree. I start to wonder what the differentiation point is. And then you also have this negative point of differentiation for a lot of EV buyers, which is Elon in the political sense. Sure, sure, sure. Not that I'm saying that everybody, but certainly there are people who have that opinion. That's interesting.

And so having car play back in is definitely like a step in the direction of we need to sell some damn cars. Yeah, if you're running Tesla today, what would you do from a, what would you do from like a product offering standpoint? Man, I don't know. That's an interesting question. All their products are really old. They have, Elon is both the greatest asset and their greatest liability. I would probably do my best to come out with Roadster to try to get some attention on the brand and get some sort of ALo car effect. Like, wow, this is cool again.

Wait, do you have a prediction for the? Yes. So Elon was on Joe Rogan and said that he was alluding to that the demo will be shocking and it will be something where the car basically lifts off the ground. He was saying like he was kind of alluding to it being a flying car, but I don't think anyone expects it to be like a full quad rotor, like, you know, full VTol vehicle. but it's going to have a trick. It's going to have a quirk. And I'm wondering if you have any predictions for what that might be. Like we were talking about the Yang Wang jumping and stuff.

It's interesting because at this point it needs to have a quirk, doesn't it? Because when it was announced six years, it's been six years almost exactly, right? It was 19th L.A. Auto Show. It was going to be this huge thing, zero to 60 and 1.9, this amazing electric supercar. Well, since then, 10 other electric supercars have come out, all of which do zero to 60 and 1.9. And the Yang Wang is jumping. Yeah. So, like, you better show up with something at this point because you are so far behind everybody else in terms of what they've all got going.

You basically need to have some sort of interesting hook now in order to differentiate yourself from these other products, which, by the way, I don't know if you guys have been paying attention, but all the other electric hypercars have been horrible sales failures. And it'll be interesting to see what, whether Roadster has Elon and Tesla and all that behind them. It'll be interesting to see whether Roaster can overcome some of the challenges the other ones have faced. Yeah, I do think that comping the roadster, I mean, if he can get it out at 200K, that's just a completely different game from a remotes Navarra at a million or, you know, pin and farina Batista. Like those are just wildly different. And even if you have the money, I imagine that like the confidence of like a roadster,

like you're going to be able to take it to a Tesla dealership, get fixed with parts and stuff. With a Batista or something, you're going to be a crazy hypercar unique one-of-one world Konigseg level. No one's really brought that down. Although, you know, Tycon TurboS, there are a lot of options in that area. Yeah, I mean, you're right that no one has really played in the space sub a million bucks.

However, first off, I don't think the Roadster's going to come anywhere near 200, just based on we've got a lot of inflation. Elon's missed those, Tesla's missed those price targets before, et cetera. But also, you know, yeah, the Tycon Turbo, the Ferrari SF 90 has been a huge dud. Like, there are some examples out there of, okay, maybe people aren't really completely. Just willing to buy anything.

Yeah. Yeah. What, do you think manufacturers are focused on EV manufacturers, like, are really focused on or have a plan for solving depreciation with some of these cars? Because, like, I, I buy a lot of cars. I cycle through them quickly. I tend to gravitate towards cars. Like, I've owned multiple G-wagons because I just love that it's timeless. I know it's going to depreciate, but not at a very, very reasonable rate.

And in that sense, it's like a practical buy, or it's at least way more practical than buying a fresh Tycon and losing, you know, 50 grand in a few months, basically. I mean, the only real way to solve for depreciation is either supplier price, right? You can either dramatically lower the price and get more cars out there, which hopefully your supply then evens out, or you just make fewer cars. I will say, I don't think depreciation has been quite as crazy as everybody else thinks, in part because I don't think the transaction prices were ever where the automakers were pricing the cars in the first place.

Oh, interesting. The cars had stickers at 50, but when you factored in the tax credit and the discounts, they were probably really selling down at 37. And then the, however, still, they are depreciating faster than gas cars and just unbelievably fast in general, except for Rivian, which has held its value relatively well, but still nowhere near something like a G-wagon. The only way to really solve that depreciation problem, supply, price, or just increase the desirability of the cars.

But I don't think that's going to happen. How on earth has Rivian maintained a monopoly on the full-size SUV in the electric category for so long? I was expecting a cyberbin, a cyber truck suburban version, like immediately, and it just didn't happen. It's so weird that the other automakers went pickups first. And I don't, I wonder if they were pressured into doing that because they knew the cyber truck was coming. But it is so clear. When I get asked by people who want a car, I'm having. California. Everybody's got an electric. Everybody wants a electric car. The only thing I am ever asked

is, I want a three row and I want a hybrid or an electric. Yes. And my response is, well, you're getting an EV9 or an R1S. And the EV9 is new to the market a year ago. The R1S is not since 22. And it's like, yeah, where was everybody else? I mean, GM is screwing around with the Hummer pickup and the Silverado and the F-150 lightning at Ford. It's like, what people want is just a three-row electric. And the R-1S is the real success story in terms of holding their value. We sell them on cars and bids. They are still not coming under 60 grand, which is a considering that's a three, four year old car, that's pretty good with the starting sticker of around 80. The trucks have depreciated a lot more. There's less interest in them. But yeah, it's wild

that Rivian figured that out before anybody else. Everybody was like, oh, we're going to start with a sedan. It's like, why? What are you thinking? The market has gone from that that stuff. Yeah, I mean, I guess Porsche now has the Cayenne Turbo. You just reviewed. But even that's not three row. And I'm thinking like if you are if you are like the true American family I have three boys two dogs to Newfoundlands like I need the huge 220 inch SUV soon like or you know the the full size escalate like the Tahoe like the suburban like this is where I'm going to be. This is my life for the next couple of years at least or the two decades. Like no one's really figured that out and I thought it was I thought it was maybe just like the laws of physics or something like oh you can't make the battery big enough. But then you see the cyber truck and you're like, you could easily just make the back like this and you're in business. And the Hummery V. Yeah.

It's so weird. I guess they just saw volume in those pickup trucks and so they decided to go. But totally. I mean, you now have the escalate IQ also, which is not a very compelling product. But it exists and it is another three or a V. It's clear that they're starting to come. Even the escalate IQ though, it felt like they shrunk it down a little bit compared to just the full size escalate ESV, which is the one that people think of. Right.

And the interior really feels like that. The interior does not feel as big as that. You come out with a car that weighs 9,500 pounds and is that long. It really should feel a little bit bigger inside. I should be able to play paddle pickleball in that. And that's not the situation. Yeah, for sure. How can you talk about European regulators?

Because I feel like sometimes I see car companies, manufacturers, making like what feel to be like nonsensical decisions, decisions that at least American consumers are not like saying like, hey, we want this. And I think it can oftentimes get, you know, people will point to regulators and be like, you know, there's actually pressure coming on the manufacturers on another continent and that. But how true is that? And what kind of role are they actually playing in? That's an interesting question. Not a lot of people think about. Regulation plays a bigger role than you'd think in a lot of stuff. We just heard one this week.

Mercedes is canceling all of their 43 AMG cars, which are actually quite popular. Here in the States, they're very popular. The C-43-GLC-4-R. Interesting. Those are all going away. And the reason is that they do not comply with EU noise regulations as of next summer. And it's like, that's a pretty successful and popular car.

And I guess they don't. And so the manufacturers can't create like an American, like it, like, There's precedent for creating like a European version of a car. Like I know the European metris is called something else, but is actually a lot better than the American metric. They must feel that the U.S. market alone doesn't, or U.S. plus Middle East or whatever, doesn't justify the creation of such a vehicle. And so it goes away for the entire world. And that's true. And a lot, there's a lot of compliance and regulation stuff just beyond that.

But that was an example that came up recently that I was astonished by because it sometimes affects popular cars. We saw it affect Boxter and Cayman, right? I mean, clearly, Porsche is not really going to be able to continue making these cars as we think. Now, they're rethinking that strategy, but it's still going to have some electric component, and these regulations are stuff that they're going to have to comply with. It's actually funny because you hear a lot of complaints. The Trump administration has come in and cut a lot of EV regs and push back on EVs, and there's all these complaints.

I mean, U.S. regs were never anything compared to Europe. Europe has gone really, really, really strict about noise, about pollution, about emissions, CO2, in a way that is definitely affecting vehicle. sold there and probably does have knock on effects on cars that are sold mistakes too. That makes sense. Do you think it's fascinating that some European regulators are pushing to take away cars that people like from European manufacturers while they let the Chinese flood the market and kill their automotive manufacturers?

Which is especially interesting because China is doing the exact opposite, right? China has really fenced their market and has really made. very easy for Chinese companies to sell cars and do really well, but has made it specifically difficult for foreign market companies to come in and do the same. Meanwhile, free capitalism has kind of had the opposite effect in Europe. And boy, you go to Europe now. I go to Europe every year or two, and it's amazing every time I go. Five years ago, a couple Chinese cars. This last time, they are flooding, like you say. It's true in Mexico, too, and Latin America in general. They make cheap cars that are desirable and they're all EVs, so they fit the regulations. Yeah, it's

certainly interesting. If I were the European automakers, I'd be like, what the hell? But yeah, do they not, do they need to like 10x their lobbying budgets and say, we want to make cars that are fast, loud, and fun, and you've got to protect us from, like, I just don't understand. Like, these are like, like Mercedes is like a national champion in Germany, right? You would say make, make, we work, the environment is more important to us. My guess is they would say, you are important to us. The environment is similarly important to us. We have to figure out a way to make this work. But for right now, China is already there where you probably should be, and we're trying to incentivize you to do that. It is surprising to me that they don't have more

significant tariffs on Chinese cars and ways to kind of try to stop that. However, Chinese cars have also done an amazing job of mobilizing, especially Southern Europe, to be able to buy new cars for the first time in a long time. Southern Europe, they're not a lot of money in a lot of cases to buy these cars, but these Chinese cars come in. They're affordable. They're electric. It's getting people out of fiats from 2003 that pollute at ridiculous levels. And you got to assume the EU regulators actually are, they're like looking at it saying, yeah, it sucks that Fiat is losing market share, but here we are doing our part for global warming and for the environment, which is a big. I would just hope that that was more of a more of a fight between

Tesla and the U.S. electric manufacturers earlier. Like Tesla has been, and the other, and Rivian have been like options for that solution. You'd think, I mean, obviously they need to bring the cost down, but there's, I think that there was a lot of pushback to, Tesla in Europe years and years ago where they didn't think that they should go after that Southern European buyer. And so they never really got to scale it with like, you know, the the two point nine or something, the cheaper version of the three to actually get out there or something like that. I don't know.

Right. And meanwhile, BYD is there now with cars that are this long that cost probably 18,000 euro. Yep. Yeah. Yep. I wanted to ask about modern cars. Wait, wait, wait, one question that fits into this last one.

Jaguar revamp. Team asked us to ask, European manufacturer, going all in on EVs in a very opinionated way. Do you have any predictions on like first year sales numbers in the U.S.? Well, you know, they haven't really shown a production car yet. So it's hard to make that prediction without knowing, hey, are they going to do SUVs, what price point, et cetera. I will say they weren't exactly blowing it out. out of the water before. So like, unlike a lot of people who watched that ridiculous commercial and have watched some of their product strategy and kind of laughed at it, I'm not sitting,

you know, we got a lot of comments on our pod, Jaguar, they're screwed. If you go woke, you go broke, we're going to boycott. They were already broke. Like, it wasn't like they were out there really lighten the world on fire. So like I think the key thing is probably a benefit for them, truthfully. But we'll see how they actually end up doing it. It's already been quite a while and they haven't really announced cars or revealed anything coming. So that'll be interesting. I mean, the brand name is strong. But like, you know, what are you going to do?

It's just, it seems like a far-fetched plan for it to actually work. Yeah. If you're already, if you're already broke, why not go woke? I guess is the other problem, by the way, that people don't discuss the, the big problem that Jaguar faces is that almost all of their dealers are paired with Land Rover. And so at the end of the day, they cannot go after the SUV segment to the level that I think, a rational company would wish to, right? We can sit here and look at the success of the Model Y and the R2 or the R1, but will be the R2

and GWagons and just SUVs in general. And Jaguar doesn't, they did try SUVs, but they always have to kind of back them off what I think the market would want a little bit because Land was there actually having the success. To make an SUV, but don't make it too good. Don't make it too good. And don't make it too off roadie or too boxy or too cool, which is what the market is really obsessed with right now.

Because, you know, literally across the showroom, we're trying to sell these things over there. And they have a better name, and you don't want to try to undercut that. Yeah. It feels like the other market Jaguar could potentially go after is maybe like the James Bond want to be. But there they're competing with Aston Martin, of course. Which is not exactly crushing it either. Well, and they make gas cars. It feels like they are.

So I'd love to know, like, where do you see Aston going? I mean, the Valkyrie is incredible. The Valhalla we saw in person looked really cool. And like, there's just some incredible design stuff. I think the issue, my point of view with Aston is like, I love Aston Martin. We were at F1 with the Aston Martin team because our friends from public are a sponsor. And I love every single silhouette they do. I think they make cars that look absolutely incredible.

But the problem is that literally every single price point for Aston, there's a more desirable car from a Porsche, from a Ferrari. And so it's like, that's just really challenging as a manufacturer because you're like, okay, we made, we made, we made. made the Volhalla and it's like, okay, what Ferrari can you get for a million bucks? And like, how many guys are really going to look at that trade and go for Aston Martin? I think that Valhalla and Valkyrie will and have been successful. I think the biggest problem is the volume cars. And I think it goes back to something that you said earlier, which is they have a significant depreciation. And that's not a secret. Well, you buy a brand new Ferrari and you're not going to lose money. In fact, if you're lucky enough to get chosen to

buy a brand new Ferrari, you might make money. You buy a brand new Asson. Holy crap. You know, used DBXs are already 7580. That's their SUV. Used uruses are still like 175. And the price point is pretty similar, new. And which one would you rather get? Ashton, I think, actually makes them pretty compelling cars, even relative to rivals like Lamborghini and Ferrari and Porsche. But you go into the dealer and there's 12 of them sitting there at the dealership. They can't sell them. They do big discounts. They make too many cars. The price points are too high. And that's a, as beloved as that name, Aston Martin, we're. all want an Aston Martin in our minds, but when you actually start to look at the particulars of what that would entail, nobody wants 150 in depreciation. Well, they didn't even have a car. I mean, you buy Aston Martin from three years ago, and it had the interior, it had the screens from like a car from like the early 2000s. And so I do wonder, I do wonder if the depreciation gets, a problem gets a little bit less bad now that the interiors

have modernized substantially. they've improved, but at the end of the day, there's still a lot of Mercedes-Benz technology. And when you're not using Mercedes-Benz technology, then another question about Aston Martin starts to become reliability, right? If they're making their own crap, then you start thinking, oh, God, do we really want to be making their own crap? It's kind of a tough situation.

And Ferrari and Lambo get excused from some of that stuff. Yeah. Do you think that how many people do you think are frantically trying to re-register their cars to California after the Whistling Diesel scandal. Whistling Diesel, if anybody didn't see, was arrested, happened to be ready to vlog the whole thing, which almost looks like he did a deal with authorities,

which is like, hey, we're going to arrest you. We need to make a big scene out of this. We're not going to really throw the book at you, but we need to do this in marketing for it. I think it was all legit. I think he is always ready to vlog. I think his boys are like always at the show. You know what I mean?

Yeah, I'm sure. I'm sure. But yeah, boy, that was a big deal. so many people in the community do the Montana thing. And I really think we did a big thing about it. It's on our pod last week. I really think with the advent of license plate readers, and with states just starting to realize how pervasive this has become,

it's just really going to start becoming more and more difficult to do. And as a guy who's got four very expensive cars registered to my actual home addresses, I watch the Montana thing. And I think to myself, you know, if I have to pay, you have to pay. I do feel bad. I mentioned this on the pod. I do feel bad for the guys who are doing it to get away from the stupid regulations like smog on a $15,000 car. And meanwhile, it's being ruined by a guy after person after person doing it on expensive cars.

There's a little hypocrisy there that I've decided smog is something that I'm willing to make an end run around. But I do feel that way. So whatever. But I do think that Montana thing, yeah, so many people do it. And I think so many people should really think about whether it's the right call. You want to get arrested to save $20, $20, $20, $20,000, $20,000 on sales tags? Yeah, yeah. Back on Jaguar, to some extent, what is your current top pick for something that's basically a concept

car that you hope it makes it into production? For me, I think it's the Hyundai Vision N that looks like a DeLorean. Yeah. They might go, they said they were going hydrogen. I think that they should just do like the Ionic 5N treatment and just electric. and it'll go zero to 60 in two and a half seconds or something. Like they could do something cool there, but do you have any other picks for stuff you've seen when you're like,

let's bring it in? That one was really cool and literally cool. Although I have to say, just a couple weeks ago, I reviewed the new Hyundai crater, which is an off-roader concept car. And from a perspective of a car that I think will sell, they should make that.

Body on frame, off-roading has become such a thing in the last 10 years with the Raptor and the G-Wagon and the forerunner and the Wrangler and the Bronco and off-road trim levels of every single car in existence, including minivans. Hyundai has started to come out with some off-road trim levels, but it's pretty clear to me that, like, there should be an, they should go after the forerunner. Forerunner is selling pretty easily millions, literally millions of units. Wow. And it doesn't sense to me that these automakers are not going after it more. I think Honda's insane for not pursuing it, but Hyundai and Kia have been much more aggressive and will be.

to be aggressive in these kind of new segments. And I think that crater is previewing a car that will exist in some form, probably not exactly like that with graphic displays on the windshield. But something's sort of similar. I bet you they're cooking it up. Where do you think that trend came from? You know, I don't know. It's interesting.

I was in Joshua Tree yesterday. And boy, there's a lot of L.A. people out there. Like, it's wild. It has become such a thing. Like, going out into the wilderness. I don't know. It wasn't a thing when I was a kid. It was reserved for like a select group of people.

You only bought a Jeep Wrangler if you were like kind of crazy. And now it's like mom cars and everybody's got a Bronco. And the forerunners are this massive thing. And I don't know. Obviously COVID exacerbated it, but I assumed that it would slow down when COVID ended and that has not happened. I wonder if it's some sort of like, you know, you're online all the time. You're on your phone too much and then you want the exact opposite.

And so demand increases for like, what is the opposite of TikTok? Like get me to the forest. It's also very American to like find like have an outdoor activity explode that's about bringing heavy machinery like out into the wilderness. Yeah. That has stay around. You can stay around the machine. You can still watch TikTok, but you can at least act like you're off. When I think about it, I was actually thinking about it this, this weekend because I was out there and seeing all this.

And there's a camper in every parking lot. And I do wonder if like dad's being more involved. When I was a kid, dad had a sports car, not my dad, but in general. Like dad at a sport, he would use it to get away. You would take it on a drive and take it to a show. Sure. Now, like, dads and families want to do this stuff together, and sports cars, you can't do that with your kids.

And I've even watched the vintage SUV. I got a Ferrari FF to try to get around that when I had my first kid, and it did not work out. The electrical issues were, like, if I drive this thing to the farmer's market and I stop it, it might not start again. So I got to make sure I started at home and stop it. But I am watching

Grand Waggneers, right? An old landcares and these are things you can now do with the whole family. And I think that there's been like sort of a renewed push among dads and among families to kind of do hey, let's do the vintage car thing together. And it's really,

it's really become kind of thing. But I think there's probably many parts of it. Speaking of dads, speaking of firing dads, what's the best modern car that screams I'm coming to town to fire your dad? We have this big thing on our podcast that, yeah, we think that there are some cars that just say, I'm here, but fire your dad. It's such a big old, like, this is a rich guy who doesn't care. The electric rolls Royce, which is called the specter, that's a big one because it's got one of the things I like, fire your dad cars to me are cars where it takes up a lot of space but is unbelievably impractical.

So that car is a massive vehicle, but it's got two doors. And that guy owns the factory. That guy, he comes in. He's like, I don't care what's, this whole factory is closed. Be gone. And he, you know, he takes the Rolls Royce off. It's also so tall. We saw one at the airport next to an EURAS.

And we were like, it's a car that's taller than that SUV that's been lowered for performance. It's like. Well, I have a potentially, a potentially new fire your dad car. A friend Paul asked, asked me to ask you, do you think Sam Altman paid 20 million for the Gordon Murray S1. It sold recently and Sam was photographed in Vegas with Gordon Murray over the weekend. And Sam has an F1. Yeah. Exactly. I'll tell you that I say it could be the new fire your dad because if chat GVT takes your job. Oh, it fires everyone's dad. That is an interesting point that often may have bought it. I will say the commissioner, the story I've heard is

the commissioner of that car commission five and personally, along with Gordon Murray three people to sell three others to. He kept two and then decided to sell one publicly. And so I would be not surprised if Altman is one of the four people who has them. He's obviously interested in these cars. And yes, he's there with Gordon. But maybe he didn't buy the public one. But who knows? I bet I wouldn't be surprised if he's got one. 20 million. That's crazy. Unbelievable number. It's a lot of money, but it's a lot of car. Any progress on getting into the Sultan the Brunei's collection. I feel like with all this investment and progress, like, all I want. If I could just get, if I could just have a week to film videos in that garage,

I would just retire. I would never speak. But how much truthfully, like, how much time have you spent actually chasing it down? Because I feel like if you go over there and you're meeting people and you're, you know, you're, oh, yeah, I'll come to your holiday party and give a talk for free. You know, like, like, there's ways to like, it's an interesting situation because I have a suspicion just based on some of the rumblings I've heard from either people who have been there who have tried. I think that in Brunei, it's viewed as a bit of an embarrassment. That Jeffrey, which is the brother of the Sultan, went and spent this much national money on just these cars. And if they created a museum, everyone I know would suddenly go to Brunei.

I'm not joking. Like, I have no reason otherwise to go to Brunei. I mean, it's odd because I feel like if they do regard it as a mistake, It's like, well, you got a great collection. Let's liquidate it. Sell it to Dubai. Dubai likes setting up museums. They have Mr. Beast land now.

Get Brune Island going. I totally agree. I'm not even fairly sure that they want the publicity of the announcement that all of these cars are going to be sold off. And it's become to me that for whatever reason, maybe it's embarrassment. Maybe they actually still want to enjoy the cars or have them. But it seems to me that there is a pretty clear reality that these guys do not want publicly to have a wide splash with the excesses of the 90s in that country, which I think is such a shame.

Now, part of the problem also was selling all the cars is that they're all right-hand drive. So there's a little bit of a challenge into where they would have to be sold. There's only a few markets. But nonetheless, I'd be down. I have my Bentley dominator literally sits right here, which is by one of the Sultan cars. Fantastic. Permission to Bentley SUV before even Bentley had any idea of doing it. Yeah. Would you ever do a museum?

Like a review, like review cars for a museum or myself. No, no, no. Like the Doug museum. Cars and bids activation or something. I am not a collector. I like to use the cars and I like to be focused with the cars that I choose. And I'm never going to have a giant collection of cars. Yeah, but I meant I meant more like if I'm a collector and I have a special car that I'm just planning to hold and you were creating a museum and you were basically saying, I'll store your car and maintain it, but, but, if you want to.

A museum of other people's vehicles. Yeah, I would, I would maybe curate such a thing. I could see myself doing that in a, in a retirement scenario. If someone, the space and, uh, if someone takes care of the hard parts for me, I would, I would, I would curate a nice little museum. There you go. I think, I think it would be a hit. What are the biggest moments on the calendar for 2026 for you, either like, moments that you're excited about from the car, The Manufacturers, the world or whatever, but also within your organization. I'm generally just excited. You know, it's an interesting thing.

My world is nowhere near as varied as it once was. You know, I used to travel all over the world and review the craziest cars, but I kind of filmed most of them. And now it's mostly just new cars. And I still get excited from a work perspective. I still get excited about the same things, which is just going in and reviewing the cars. We had the new Nissan Centra in the office the other day. And like I'm driving in in the morning being like I'm puck. Like I'm excited.

And there is still that excitement. There's not like surprisingly there is not specific cars that are that exciting for me. The new supercars are always a big deal. And there's going to be a few coming out here in quick succession, which is the McLaren W1 and Bugatti Turbion. And obviously I'm excited about those. But it's not just those cars. You know, we get big views doing videos of Kia teller rides. And I get excited about there is something exciting to me about an automaker.

playing in a very competitive space, which supercars isn't. It's a more of an emotional thing, and a lot of times people buy them all, whereas tell your ride, you have to distinguish yourself with quirks and features from the Pylinder, the pilot, the Palisade. And so I always find it to be kind of just as interesting to see what those guys are doing. And they often have some cool stuff. They often reward me with some cool quirks and features. On Ferrari, do you get sick of people commenting, like, bring back Pinnon Furina?

Like, I just feel like every Ferrari video, it's like, they're not coming back. It's over. Like, I, I mean, every, every Ferrari video, I'm a go to the comment section across every channel. And one, my personal belief is that the cars actually look amazing when they're specced properly. I think they look really rough the way that Ferrari specs them, like, for basically marketing, press, right? And then I think once people with refined taste get in there, they end up looking amazing. So I'm not exactly like disappointed with the current design. I agree. I didn't for a while.

When they first left Pinnon-Friena, I thought there were some questionable cars. However, the 296 is unbelievably beautiful. They have, and honestly, the Dodici Chilinery, which I made fun of because of its ridiculous name, and also because of its design when it was first announced, I got to spend the day with one, and I totally feel opposite now. I think they're beautiful, they're striking. Ferrari is designing cars in a way now that is like

at the forefront, at the cutting edge, rather than just sort of only going after beauty. With that said, Pinn and Freen and Ferrari have an unbelievable legacy together, and it would be nice to see them return, reunite in some form. Yeah, but what is... Yeah, they did, but now it's been like more than a decade,

and it's owned by an Indian, like, conglomerate, like, are all the same people even still there? If they did partner, would it just be, like, like, a, like, just for the brand purposes? Like, Hasseblot is owned by DJI now. Like, what is that partnership? You know, these companies, they change over time. They just become brands. They just become names, IP.

If they did partner, would it just be for brand purposes? Yeah, probably. But I bet you they'd sell some cars. That's a good word. If the 500-unit run of a pin-in-for-ean-designed car, that would kill. Just like if Ferrari did a manual, which I sincerely believe they should do,

I think that also would have, even if it was just for the purposes of making money, I think it would be successful. On the Ferrari doing a manual, you get on the stump every once in a while with like, I want to change to what this manufacturer is doing or what the car community as a whole is doing. What's the highest confidence, like,

I'm responsible for that? Maybe it's just like a tiny thing, but like, you know, There's a lot of things where, you know, you say stuff on the internet and then the change happens and you're like, hey, maybe they watched the video. Maybe it was a factor. Maybe it's just, I got lucky and I was talking about the thing that they were already working on. But I think very little. Do I actually feel responsible?

However, when I bought my convertible G-wagon eight in a year years ago, eight and a half years ago, nobody even knew that car existed. And within a year, Kendall Jenner had one. And I did a bunch of videos and I posted a lot of pictures on Instagram. I drive it on the beach with beach grass and it looks amazing. And the values of those suddenly absolutely went through the roof, which was not the case when I bought my car. I paid 125 for my car and if I could find another one for 125, I would buy 10. And suddenly, six months ago, three months ago, Mercedes announces they're coming back with a new convertible G wagon. And I don't think that I was directly entirely responsible, but I wouldn't be surprised if I played a little bit of a role in the cultural remembering that this car exists.

existed to the point where values kind of blew up. There were some influential people got Mercedes ear. And Mercedes was thinking, okay, maybe we can do this again. Yeah. No, I think that's 100% right. It's, uh, yeah, you deserve the victory lap on that for sure. On that ridiculous car, which is such an embarrassing car to own that I don't even want the victory left.

We're giving it to you. You're directly responsible. Other than that, though, I don't think I have significant. I mean, I've been, I've been harping for a long time about how automakers need to make more off-road SUVs and that's happened. But I think also the market trends have made that pretty clear. Totally, totally. Yeah.

I mean, Ed Bowling with the, what is the LP640 prices, he kind of discovered that there were far fewer of them than people thought. And so that obviously changed the price. And if you bring new information to the world, like that feels like you do have an influence. Now, in Ed's case, he's got a few of them. So, you know, he'll talk them up a little bit. That's the game I should have gotten into. And I wish that I had.

But unfortunately, I did not do that for my own personal game. Not that that's the only reason that Ed does it for. He's a true fellow, and I think that plays a role. We're going to see Doug with like, you know, he has 200 SF90s, and now he's talking really positively about them. And he bought them like at the trough of the depreciation curve. What's going on here? Actually, there's no electrical issues.

I did all this extra studies and the hybrid system. Can we get Nissan to bring back the Marano cross cabriolet? That is important. We are big fans here. We took your joke. we've been running with it and the whole thing about that car. I don't know if you know this, but the convertible tops fail. And so just leave it down.

I don't need the top. I'm just joyriding that thing. I never need the top. Yeah. Does the SP1 have a top? No. Right. It doesn't have a windshield.

Right. So those cars eventually they will all fail and that will kind of kill them because apparently the Nissan's only solution is you have to replace it. There is apparently a shade tree mechanic guy in Florida. who figured out the problem and is willing to fix it. But you've got to send your car to him. So if you live in the Tampa Clearwater area, you can get your top fixed.

But otherwise, it is worth sending your Morano Cross Cabrera. I mean, the Resto Mod community in 10 years will be going crazy. Oh, no, no, you know what we're doing. We're cannonballing it there and back from L.A. That's what we'll do. That car was utterly awful in every way

because it is a very special product that could have a cult following if it wasn't so bad. it's very funny. Anyways, well, thank you so much for taking the time to shout with us. This is fantastic. Congratulations on all the success.

And please keep the podcast running. We've been loving it. It's congratulations on the massive success there and with everything. And we appreciate you coming and hanging out for a while. Chopin it up. Yeah. Cheers. Thanks.

Thanks for having me. Bye. Let me tell you about Turbo Puffer, serverless vector and full tech search built from first principles on object storage. Fast 10X cheaper and extremely scalable. I'm also going to tell you about. Public.com investing for those that take it seriously. Multi-asset investing, trusted by millions.

Our next guests are both named Alex. We have some folks from Ramp, the creators of Ramp Labs, Ramp Sheets. We're bringing in both Alex's. Please kick us off with introductions on both of you, how you came to work at Ramp, work on this particular project. I'd love to hear the origin story. It's a fascinating product, and we're going to go into it. Hey, I'm Alex Shepchenko. I've been at RAMP for now two years, started out and still am on the Applied AI team. Overnight success.

And throughout these two years, I've been working on a bunch of experimental things as part of Applied AI. And lately just been trying to wrap them up into Ramp Labs so that we can publish these things publicly. That's great. And Alex? Sorry. Yeah, I'm Alex Stauffer. one of the leads of Ramp Labs alongside the other Alex. And yeah, this is the place for AI bets at Ramp.

We're really experimenting, going to the cutting edge of AI, trying to ship product that is actually outside of the current like Ramp mandate. Yeah. Do you guys have nicknames yet? I feel like people would get confused, Alex,

because you can't even just be like, oh yeah, Alex at Ramp Labs. You have to go first. further? Yeah, just the last names, I guess. Shifchenko and... So talk to us about the product. It feels extremely aggressive to go after Microsoft, to go after Google, products that have massive user bases, and yet you're launching ramp sheets.

Talk to us about the thought of the positioning. Are you going to try and replace the investment banking workhorse that is Microsoft Excel? Is this more for CFOs who are already on ramp? How are you thinking about the product itself? Well, originally this started out as an experiment as everything within Ramp Labs is. We've been trying out different variations of this, actually, just to try to help our own internal finance team. This has gone, as I said, through many iterations. This initially was like process mining to document some of their workflows from video so that they can communicate better between each other into software engineers.

And then we tried to take that piece and create like Zapier and like retool workflows based on them to try to automate. And we tried to get them to use it. And they were like, no, this is too black box. We can't really use this. We need visibility into it. And so we like took a step back and tried to look through the looms of like all the information that they were giving to us. And we jumped through like random locations. And 99% of the time when you like open their loom, it's like,

in a spreadsheet. So we decided we should really meet them where they work and have a spreadsheet interface for this. And this doesn't necessarily need to be replacing anything. This can be an add-on to their existing workflows, right? This can be a quick way to create like a pro forma or something that they can then export to a CSV or an Excel file and then load it up in sheets or Excel and continue their work there. So that's broadly the way. So why build, why? Why? Why? I build like a web version of a spreadsheet at all. Why not a plug-in that lives inside of Excel? That's what Bloomberg and Capital IQ and most of the other kind of complementary systems have

typically plugged in through connectors. You decided to build the actual full spreadsheet in the cloud, in the web. Why that decision versus just plug-in? Well, internally, we do have some more complex work. blows that rely on, like, sheets and stuff that people can use. For this, as I said, like, RAMP Labs is, like, the experimental branch, and this was an experiment that, like, got very, very large. And that's, like, the way to allow people to interface with it as quick as possible.

We didn't need, like, access to your Google account. We don't need access to your Excel account. You can just go and try it out, and it's there. And it's available on RAM.com slash sheets. to anyone, not only ramp customers right now. Oh, cool. That's probably one of the biggest reasons. Yeah.

So for some context, a lot of people think that there was a big team behind this or that Ramp is like investing a lot of resources here. And the truth is that like three guys kind of built this in a cave. Pretty fast. So yes. Yeah. And we want to release it. We believe in shipping quickly and getting feedback from customers.

we thought we built a pretty cool tool and just wanted to see how people would play around with it. So we released it last week and then the response was overwhelming. There are 2 million impressions, thousands of users joined immediately. They're making thousands of sheets every day. Places that you wouldn't imagine have been using it. It's popping off at Wharton with students and professors. many banks that are already using it. VCs, finance, people in the space industry. I got off call with a user who's based in London this morning, who's a VC. So it's like it's astounded us,

honestly. What, where are people getting the most value or what kind of like work, work are they actually doing once they kind of play around with it and get a sense for how it works, like how are they applying it? Yeah. There are a lot of interesting use cases. Some are personal. There are a couple of stories here where it's helping them plan their wedding. You know, these things can get incredibly complex and you need a way to organize it. Another use case, an employee here is he's moving to New Jersey and wanted to find like the best place to live and want to model for that. But outside of personal cases, there is a real

demand here, especially with founders, VCs, small businesses, especially owners of small businesses, they don't actually have a large finance team, right? And it's incredibly valuable to spin up a model really quickly instead of going and paying a consultant thousands of dollars to just make a template for them. Yeah, it makes sense. Very cool. Where are you guys, where are you planning to take it from here? Yeah. a bunch of features that we're still working on we actually shipped shareable links today um and then exploring what best templates so like very very intricate templates that you can take and and create basically kind of these like pro forma's right away or like budgeting uh right away um and then as well

as like integrating with ramp itself so that you can like actually export your spending data and like analyze it within uh ram sheets so those are like the three big biggest things right now. Very cool. That's amazing. The chat is saying, Ramp, construct an LBO for this regional HVAC company that backs into a 25% IRA. Make no mistakes, please.

Maybe one day. I mean, that actually probably is within what is basically one shotable by the model in ramp sheets today. But thank you so much for coming and breaking it down for us. Congrats on all the progress. And excited to watch where it goes. Excited to see more people demoing it and taking it for a spin.

Incredible launch. Thanks so much. Thanks for us. We'll talk to us in. Cheers. Have a good day. We are going back to the timeline for a minute while we bring in our next guest. We were covering Open AI shopping research.

Oh, yes. And Doug join. But really think I'm bullish on this product. I think being able to effectively run a deep research report and then see all the information that you need on a bunch of different products and then eventually just hit buy right from the end the app is going to be like a crazy flywheel that's like actually what people do today they start google searching they find out what products are available they find out pricing features etc then they make a buying decision and if you can pull that all into a single

sort of like experience it's going to be very very valuable and this is effectively the deep research as a product is a product that has better product market fit than almost anything else in consumer AI and so I'm very bullish. Well, if you're selling a product on ChatGBT, BT, you've got to go to numeral.com. Let Numeril worry about sales tax and VAT compliance. Compliance handled so you can focus on growth. No, I agree.

It'll be very interesting to see how fast, like, how fast Open AI makes their first dollar from advertising or commission-based sales, how quickly they make their first million, their first billion from it. Like, the numbers are so big they need to get to tens of billions of dollars on those business lines. fairly quickly. They're certainly the most optimistic growth areas for the business. And something that I think everyone, it's very consensus that if they

get agented commerce working, they can, the pool of capital is very, like the audience is already there. They just have to run ads. This is like proven. Whereas breaking into hardware, that's something that Microsoft didn't pull off with the phone and

Amazon tried a fire phone and Facebook tried a phone for a while. Like it's not a sure thing, but it's almost a sure thing that if you have, you know, a billion DAUs using your website for 30 minutes a day, you can advertise to them. It's, it's very, very proven. Before we bring in our next guest, let me tell you about Vanta, automate compliance and security. Vantas is the leading AI trust management platform. Our next guest is Quinn Slack, the CEO of AMP. How are you doing? What's happening? Welcome to the show. Hey, guys. Good to be here. Great to have you.

Thanks so much for helping on. I'm glad that we were able to get you on. We love ads. We were just talking about ads actually in the chat GPT context. No one likes ads more than us. But fascinating. Exactly. That's why I'm excited to talk to you.

So please, introduce yourself in the company a little bit, and then we'll get into it. I am Quinn Slack. I'm the CEO and co-founder at Sourcegraph. We make AMP, this coding agent. I love coding. And I mean, AI is just crazy for coding in Opus 4-5 today. It's mind-blowing. But it's also expensive.

and that sucks because a lot of people can't use it. And it also means that we don't learn fast enough about how people are using it because it's a tiny fraction of the world that's actually using AI to code in the kind of unconstrained way that you want to if you're really going to learn. So we took a look at these problems and that's how we ended up with this crazy idea of putting ads in a coding agent. I love it. Okay.

I want to understand what that exactly means. I'm going to pitch you like five different instantiations of that. One is I go to the more the agentic. commerce route, I go to your coding agent. I say, okay, I'm ready to deploy this and then it runs an auction to see, hey, there's a deal on Azure, so we're going to host it over there as opposed to Amazon. Or it could be putting ads for Amazon or Azure or GCP in the comments of the actual code. So when I go review the code, I see ads, or it could be while it's cooking, it's showing the ads in the terminal. Like, what, where are you actually surfacing these ads?

So the interesting thing about coding is you are in this all day long. It's not like Amazon.com. ads where you go there right when you have that intent. But you also use the coding agent when you have intent. So we kind of get both. We get those, hey, you're asking it, I want to deploy, where should I deploy? Should it be on, you know, Cloudflare or whatever? But you also are staring at it. And AI coding agents are kind of slow. I'm not saying we make it slower so that we show more ads, but it's up on the screen and it knows what your tech stack is. So if you're Cloudflare and you see that they're using S3 and they're doing a bunch of dumb stuff with it, well, that's a great place for you to show ads about not only Cloudflare, but specifically which of their products is

better and how much better. And then you click a button and the agent can go and switch you to it. So, I mean, it feels like I love this thesis because it worked for Uber and you've seen a bunch of companies where they've had a whole bunch of attention and they've just built big ad businesses and Uber just shows you ads while you wait for your car basically. It's like it's not rocket science. It's like a pretty simple thing. At the same time, a lot of it, a lot of times when this works. It works because someone's created this pool of attention that there's just, you know, millions of people open and open this app or they're using that. They're staring at this particular screen and there's real estate. And I see that real estate in the chat GPT app where people

just open it up and there's a big blank screen. You can put an ad there or while it's waiting. But do you have to go and get people on board and get them off of competitor coding agents and competitive IDs? Like how much do you have to do to re-platform the customer to you and then start showing them ads, or can you partner with other pieces of the stack to just plug in quicker? We're going to follow this where this goes. So yeah, we're open to partnering with others.

I think other people are kind of waiting and seeing because this idea was so crazy. A free coding agent, and ads make AMP free in this mode, it's really compelling. And that audience is growing a lot. Sure. We keep making the model better and better. Like initially, we had it so ads.

We're actually covering more than our costs. Then we made the model six times more expensive and now they're covering half. But now with Opus 4.5, I think it's going to be like, you know, probably covering 80% of the cost because it's actually cheaper. So we've got to keep making the model better and better that it uses. And free AI is pretty compelling. Yeah. Yeah, that makes it sense. I already, I already know if TBPN runs some ads in there. I know the copy, develop an ear for TBPN. Yeah. What do you think? I like it. I mean, you can, you can go. Why do you try it? just put a few $10,000 down.

Is the ad platform self-serve yet, or are you guys just doing it when I'm like by hand for now? You can go in and put the ads self-serve. It feels like all of a sudden, overnight, we are building this new ads business. My wife works at Google. She sells ads. I've seen the business,

but there's so much complexity, and it's like a three-sided marketplace now. She's really doing God's work. Really. Yeah, we love it. We love it. Yeah, yeah, I guess any kind of like comps to this, like, that you've used to kind of like talk with advertisers about it. Like, is there is there another kind of advertising that's sort of similar in the sense of like, you know, like clearly have people's like very specific attention at a specific moment in time where they're making, oftentimes making product and business decisions? Any kind of areas that you've learned from?

Are you just kind of freestyling? We're freestyling. It's a different place to put ads. And it's kind of hilarious of AI products the first place. I think we're the first or one of the first, that it shows up in your terminal or in your editor. But I think the fact that it captures the intent and it's always up on your screen, it's different.

So I would love it if there's more people doing this. But there's so many other companies we went to and they said, well, we kind of want you to take all the flames first. On the advertiser side, though, they were incredibly willing. We had a really high hit rate. And some of them, other DevTool CEOs, I would email them and say, hey, do you want to advertise? And after a day, when they hadn't gotten back to me, I thought, like, oh, they probably think we're total idiots. But then they got back to me and they ended up doing this.

Yeah. No, it makes a lot of sense, especially when you can show an ad and have one eventually, maybe it's not that way today, but basically one-shotting the implementation of certain products like that, just like contextual advertising in this context is going to be super powerful. Any ad haters pushing back yet? Obviously, I'm sure you have an ability to get people that you can opt out of ads, I'm assuming by just paying for the right tier. Yeah.

Well, we want to show ads in the more premium modes, too. So we've got to figure out that's when we're going to get the ad hares. Actually, no. Actually, no. Actually, you can't opt out of the ads. You're getting ads no matter what. I love it. Yeah, I think if you're not getting to haters, then you're not pushing the envelope enough.

And so much of what we realized when we launched Amt Free is we should have been like 10 times Boulder about it because it was received surprisingly well and it's making a lot more money than we thought. So we're going to push the envelope. But all of this just means it means more people can use code AI in a totally unconstrained way and see this thing that just blows our minds. Yeah, yeah, it seems really cool for just like, yeah, I can imagine so many people. This is what I would have used in college. I definitely would not be paying $200.

month or something for, I'd probably be interning at a startup that, or at a live show that pays for my ABS bills. But can you, can you just zoom out for us and give us a little bit of like what the corporate hierarchy is like, AMP and Source Graph? Have you raised money? What are the shape of these companies and products, how everything fits together in your world? Yeah, we started Source Graph back in 2013, actually, with this idea, let's automate software development and we start out building code search at Sourcegraph. And that's like Google for all the code. Yeah.

It's this, especially now with AI, AI is doing like a thousand times more code searches. Yeah. It is this critical infrastructure. It's, we're the only ones doing it at that scale. And then we've also got AMP. So these two products under one roof, code search and AMP. And AMP is the frontier coding agent. We're trying to make it a year ahead for 1% of devs out there.

and just be totally crazy. And you might say, well, ads is incompatible with that. Ads is all about getting a big audience. But actually, what ads do, they're in service of the research objective for AMP. It lets us not have to go and jump if some customer says jump with this requirement that's going to hold us back.

It gives us so much more freedom to change the model, change how it works. So, you know, in that way, ads, it actually just really nicely fund a kind of research lab for us with AMP. I guess in the same way that Google ads mean that it can do whatever the hell they want and they can, you know, fund a lot of experimentation internally. So we've got these two products under one roof. What were some of the tradeoffs you made when you were building the first coding agent? Obviously, there were people that trained, you know, foundation models themselves.

Other people just wrote harnesses and they're swapping out the different models from the big labs. How did you think about the different decisions that you made? we decided let's see how far we can get without trying to train our own god model let's depend on anthropic or open ai and google and ever since we started with amp which is you know we started that in february we honestly thought there was a me model competition a lot sooner and it actually only started happening in a big way last week with jemini three we switched amps model over to jemini three it was a huge leap forward. And then today, you know, back to loving Opus 4-5, I think I'm going to switch back to that.

But am I glad that we didn't go spend billions and billions of dollars training a model that was 1% worse and yet nobody uses it? No, I'm really happy to be occupying, you know, this place. Yeah. So you don't surface the, you don't have a model picker in the agent. It's not like I can, as a customer, say, I'm willing to watch all the ads and get all the ads. but I really prefer Gemini 3 or Codex or Opus 4.5, for example. I don't get to pick. Yeah, no model picker.

Got it. Cool. Last question. What do you think about integrating other types of products in the IDE? We were talking about Chad IDE putting gambling in the product. Do you think that that – do you think there's a world in the future, like, three years from now where people have enough time between like running different tasks and agents that they just have time to like goof around in in in these uh in in in the id or in various developer tools or is that kind of like an of the moment uh kind of like problem solution i think

you should be able to bet you know get some kashi manifold market stuff in there uh you should you know there should be a you should interstate what ad are you going to you should let people bet on what ad they're going to get next. Yeah. And then whichever one they want. I think I don't know if you're joking, but I think you can go too far. We've seen people get, you know, go viral, but be potentially flashing the pan over. Yeah, we'll see if I'm joking.

Okay. Yeah. I think gambling is is taking it too far, but there's a lot of stuff we can explore. Yeah. Yeah. I mean, the ads business, I mean, Google made $100 billion from ads in Q3. I think there might be enough operational. opportunity there.

What was it? Yeah. What last question? Are you hyperactive in Slack? Yes, of course. Are you, you know, my great grandfather started it. Invented it. Back in 1800s.

Yes. Yes. I'm proud to see what it's become. Yes. Amazing. He's like, Microsoft Teams guy. Yeah, great, great to meet.

Very, very cool, very, very cool product. And yeah. Thanks. I hope to have you back on again soon. Thanks so much. We'll talk to you soon. See it. Have a good one.

Bye. Back to the timeline. Let me tell you about Vanta, automate compliance, managed trust, the leading AI trust management platform. I already do that one. You did. Well, then let's go to Figma.

Think bigger, build faster. Figma helps devine and develop teams build great products together. I'll also tell you about Julius AI, the AI data analyst that works for you. Join millions who use Julius to connect their data, ask questions, and get insights in seconds. Back on the timeline,

people are having fun with Nano Banana. They're putting different hairstyles on various tech people. So Ilya has a wide variety. This is Mark Andresen with all sorts of different hairstyles. Which one would you pick if you were him? I think the one in the center looks pretty good. The longer beard, shorter hair. But some of these with a long flow.

In the middle on the left is pretty cool. Middle on the left is pretty cool. Crazy flow. I also like bottom left is also crazy. Yeah, time to take a trip to Turkey for sure. Just go, you have the money, just go full head of hair. The thing with Ilya, that's interesting is they didn't do like the full zoomer. The middle bottom one, that is flow.

Keep going down. Yeah, the middle of it. Yeah, right there, right in the middle there, the middle bottom. The ponytail is also pretty powerful. Definitely. You know, it's respectable hairstyle. This post here from Thomas Scholls, he says, the latest nano-banana model has officially crossed the line.

I no longer implicitly trust photos anymore. And sometimes I can't even definitively claim its AI now. I totally agree. I saw this picture. Yep. And my first thought was like that's got to be AI specifically because I don't think Sam is just walking down the Golden Gate Bridge in the middle of the day. Yep. It's like probably terrible from a security standpoint.

Yep. So, but it looks photorealistic. And, yeah, it's very. So if you put the photo into Gemini, Gemini will tell you, if you say, is this AI, Gemini will tell you, yes, according to the Synth ID water molecule detection tool, this image was generated in whole or in part with Google AI. Of course, we've seen previous images where if you turn up the contrast

and the saturation all the way, you can see kind of the rainbow, like zebra pattern basically that's embedded in there very subtly. But yeah, I mean, this is pretty, pretty photo reel. And so, you know, stay safe out there. It's going to be... Joe, Wisenthal asked Nanobanana to create a really annoying LinkedIn profile. That's what I was talking about. And I couldn't tell, is this a real person?

I have no idea because at this point, we're way past the touring test for images in the sense that this looks perfectly edited. but this could also just be a straight-up a screenshot. I would need to fact-check this. But instead of fact-checking it, I'm going to tell you about Privy. Privy makes it build, make it easy to build on crypto rails, securely spin-up white-label wallet, sign transactions, and integrate on-chain infrastructure

all through one simple API. Kaelin Sterling, I don't think is a real person. I don't do small talk. I do deep dives. My journey is a quantum leap through the liminal spaces of tech and spirituality. chief visionary officer, TEDx speaker, professional storyteller, democratizing the metaverse one DAO at a time, 10x growth alchemist. So did Joe actually ask Nanobanana to do this? I want to see the prompt if that's true.

Yeah, let's try to. Yeah, we need to replicate this. Try to replicate it. Because I actually did. So this was the example that I gave. I went to, I went to, I went to nanobanana and I said make a, my prompt was just create a really annoying LinkedIn profile, but I forgot to check the, the nanobanana box. And so even though it was multimodal, according to Tyler over there, it did not generate it. It generated text. And so, you know, oh, it's like, I can only do text. Very questionable. We'll see. We'll keep playing around with that. And in the meantime, I'll tell you about adquick.com. Out of home advertising made easy and measure roll. Plan. buy and measure out of home with precision.

What's up with Brian Johnson? He's starting a new protocol? Is he going to Taco Bell drive-thru? Is this a real photo? This is AI again. This has to be AI. That's a very funny, funny photo. This was my take, was that, you know, if he's really changed by his journey,

he must come out of it liking at least one fast food restaurant. And Alexis says, those shrooms spiritually healed him in a way that, has him living moss now. Living Moss. The Live Moss tagline was really fantastic. Really good. Yeah. Taco Bell knows how to do it.

Also, underrated how online the Taco Bell marketing team is. Are you familiar with this? See their post. You know the story of Sheal, Monot, guests of the show multiple time? Yeah, he got married. He got married in the Metaverse or something. Like, it was some crazy thing where he posted, like, if somebody pays for it, I'll get married in the Metaverse, like, as a joke.

And then, like, Taco Bell was like, yeah. Yeah, let's do it or something. It was like very crazy, but it's like, like, there aren't a lot of companies at that scale that are like as online and like in the culture as Taco Bell. And beyond just like, oh, they have somebody who's pithy on the Twitter. It's like, no, they actually can go and execute. I've seen multiple people are like very niche internet microcelebrities that have been like given the full dog and pony show and like tour of the Taco Bell headquarters and like impressed upon them the greatness of the Taco Bell brand. It's pretty, pretty awesome. almost as awesome as getting a luxury watch on getbezzles.com.

Shop over 26,500 luxury watches, fully authenticated in-house by Bezell's team of experts. This post from Neer, did you want to skip this one? Neer says, guy who doesn't want to be old. It seems like we'll get age-reversing tech right when I'll be old. How favorable. Guy who thinks it is different this time, but this time it's different.

This is insane, bass. Read the next one. We got some massive news from none other than Bill Gurley. We got to get the gong out for this. what you you read it uh amy girly huge congratulations to bill girly for receiving the texas distinguished alumnus award a remarkable honor for a remarkable longhorn congratulations bill well well well deserved congratulations this this made my day this made my day everyone was everyone was wondering if he was going to make it yeah it was kind of the elephant in the room it was a hugely

hotly debated there was there a lot what 10 million dollars in liquidity betting on this whether or not he was Probably. Millions. Yeah, it was sort of, it was sort of the longhorn in the room. Yeah, yeah, exactly. It was a huge deal, but he did it. He pulled it off and we're very excited for Bill.

Everyone was wondering what he would do if he didn't win. I guess we'll never know. I guess we'll never know. Riley Walls is truly the greatest internet rascal of his time. Incredible. He did it again. He cloned Gmail except you're logged in as Epstein and can see all of his emails. It's such a great workflow because people are workflow.

I mean, it's like it's a way, because I've seen other people do, here's a fine tune on the emails, here's a, here's a, you know, a searchable database. None of them have been particularly usable, but this is fascinating because it just really substantiates the emails in a, you know, in a way that, you know, everyone knows what a Gmail inbox looks like. And you can just. They add to this random page button. It's also funny because he's getting this, one of the last emails that he got was a flipboard week in the week in review. And the subject line says Alex Acosta resigns, Jeffrey Epstein arrested. So imagine, I guess he wouldn't have been really able to read this email where he learns that he's arrested. He's arrested in there.

Yeah, this is one of the thing, chat a while ago was like, oh, I'd love the first. the TPPN to cover the Epstein thing, but it's not really their shtick. Well, we found an angle. We found a tech angle. And so this is when we get to talk about it. I know everyone wants us to talk about the hottest news story in a decade. But you have to get it through the lens of Riley Walls and fun little hack projects. Before we move on, let me tell you about Aethsleep.com.

Exceptional sleep, without exception. Fall asleep faster. Sleep deeper. Wake up, energized. Uh, the reason you feel insane, according to Wilmanitis is because you're, you're talking to chat chibati when you should be talking to God near, I felt back and says, the reason you feel insane is because you're talking to God when you should be speaking with Claude. So funny. Claude.

Oh, uh, Keller is coming on the show tomorrow from Zipline. Keller Clifton, one of the greatest. Uh, get that goat sound effect ready for tomorrow. Um, but. We'll do it early. Keller said that he's launched zipping points. He can pick up packages and deliver them autonomously with the zip line autonomous drones. This is the private plane for your burrito, folks.

It's arrived. We're here. It's here. The future plane. The flying car is here. And it will deliver you, Chipotle, in 15 minutes, in four minutes while it's still warm. So here's a zip grabbing a package from one of our restaurant partners. It'll take so many cars off the road over the coming years.

That's great news for environmentalists, for congestion, for anyone who wants to be able to really let it loose on the roads. If we're getting less congestion, maybe the speed limit goes up to 80 miles an hour, maybe 120, maybe 160. Maybe we get up to 200 and you can really let it loose. I would like that. I wonder if they'll need to make, like, will you need to prove that you're at a certain address in order to get stuff delivered there? because it's such a funny dimension to mess with people and just be like, hey, look on your lawn. And there's just like a burrito.

A burrito just chilling there. Well, people do that with pizzas, right? They prank call. Like a dozen pizzas delivered to this address. I'll pay in cash. This is like a famous prank. And then you show up and it's like, I don't need all these pizzas. I'm being pranked.

I think that that's been mostly resolved by modern payment solutions. But I'm sure there will be odd oddities around these drones. Cheesh, Sheesh, Theo Vaughn, Flamen Hot Tweets, is joking around and says, imagine how cool it will be to shoot one of these out of the sky to get a free meal, go and hunting for your Chipotle Burrito. That, of course, is extremely cyberpunk and hilarious, but it will be massively illegal, and Keller breaks it down. He says, we're regulated by the FAA, so the consequences are similar to shooting at a 737

as it's taking off from the airport. Not a good idea. Also, communities love the service. And I imagine he's not saying the details, but if you shoot at a 737 as it's taking out from an airport, I think you're going to jail for a long time. And I think you will not just be able to shoot one of these out of the sky and pick up a free burrito with a 22. What about like a really big net? If you own the land, how high can you build a net? I think that would be the same as throwing a really big net at a 737 on takeoff, Tyler.

I don't know. I don't think I think there's no way you can think the consequences are similar. Like, yes, it's illegal to shoot something out of the sky, but you're not, I think somebody could shoot one of these things down and get a misdemeanor. I think with a good lawyer, you're getting a misdemeanor. You're not going to, whereas a 737, you know. If you get John quit on your side, you're eating burritos for free. Yeah, so David, David, David, David,

Chang was saying, like, it's just there's, like, he, he, he, it wasn't that he, was like bearish on the attack, but he didn't think it solved like delivering it in the right form factor. Yeah. We got to ask Keller about that tomorrow because right now, if you order food, depending on where you are between the time that it's cooked and picked up and delivered, it's like there's a big gap. Well, speaking of cooked, it has to be shorter. Speaking of cooked, there was a wild job description that hit the timeline. Rachel Mayer says, coolest PM rolling crypto. Cook. Real impact.

Real impact. Needs to be clinically. Clinically online. Not even chronically online. Apply here at Circle. We love Circle, of course, but. Wait, was this like intentionally? Yeah, this must be a troll.

This is range made done perfectly. And Fekundo says, cooked. The idea, I was cracking up with this, like, the idea of showing up to your job and being like, oh man, like, yeah, the team is here is cooked. They're all exact.

all like washed up. It's just the, the worst possible team. I think it's a perfect, I think it's a perfect post from Rachel. There's no way this job job just like, like posts would have gotten nearly the reach. I think, I think this was successful. Did you see this in the replies to her original post? Did you mean to say cooked, cooking or cracked? And Rachel says, I don't even know anymore. They all sound fun. Rachel, we ain't cooked. We are cooking. Yeah, one of the developers, like, at Circle is like, you know, like, let's correct this. This is so funny. Let me tell you about Wander.com.

Book of Wander with Inspiring Views, Hotel Grated Mannings, Dreamy Beds, Top Tier Claiming, 24-7 concierge service. What is this? Oh, the IP blocks are going out. You saw that, right? There's, uh, when you, when you identify someone on X, where they're from, if you, if you signed up a originally with an IP, it can be transferred to another, like the IP block can be sold internationally, and then you will show up as an international participant. So there is some, there's, there are some false positives there a little bit. Um, uh, Sheel, who I mentioned for the Taco Bell

wedding, uh, I hope I didn't get that story wildly wrong, uh, is mentioning, uh, a post from Alexis Mulliner who says, FY, DHH, I'm noting that learn. Umacomacom.io is inaccessible from Spain, while La Liga matches are being broadcast. And DHA says Spain blocks all of Cloudflare during those matches in an insane medieval attempt to counter piracy. That's... Wow.

And so O'Marchy is hosted with Cloudflare and they just bring down the internet when they just stop piracy. That is insane. And so Shill breaks it down. He says, oh, my God. Spain blocks Cloudflare IP ranges during La Liga matches,

La Liga, admits lists of pirate IPs and the blocks take out lots of legitimate services too so you can't use LinkedIn in Spain during La Liga games. I mean, that's kind of, that's pretty on brand. That's pretty on brand. I'm not, I'm not going to look up the GDP growth figures. I'm not going to do it. I'm sure it's going great. I'm sure it's going great over there. Very, very on brand for Spain. Did you see this? This is like the most Nathan, potentially it was like a Nathan Fielder idea. Yeah. There's a bunch of accounts saying that J.P. Morgan appears to have an existentially

threatening short micro-strategy position that can potentially bankrupt JPM if master, sorry, micro-strategy trades 50% higher above Friday's clothes. Okay. And Wall Street Betts comes in and says, you have no idea how understated this is. So a systemically critical bank is going to fail because there's a, short micro strategy. I'm not sure about this, but it's working. Yeah.

Because micro strategy is up 5% today. So could have played meanwhile, Bitcoin is up two and a half percent today. So could have played a part. Who knows? But it's a fun idea. It's fun to...

Yeah, the thing I think... I don't think Jamie Diamond is risking at all for... Yeah. On a on a micro strategy short. Yeah. wonder how they even got that information. Maybe it's from a 13-a-f or something. Who knows?

Tyler has become extremely black-pilled on the 13-F. He's building a thesis against it. We might be talking about that later this week. Also, I mean, the whole thing with the, oh, we're going to blow them out, like, it completely, it just negates the idea that, like, they can trade a lot faster than you can trade them. Like, you know, you can just, like, they can get out of that trade when it's up 10%. And it's like, yeah, it would have been disastrous at 50, but once it went up 10%, they were like, okay, yeah, we're out actually.

Yeah. You know, we're hedging, we're doing something. We're buying the other side of the trade or whatever. Especially like, if you're, if you're a short seller now, like it's not like you don't have Google alerts for what's going on on Waltry Bats. It's not going to like take you by surprise. You're going to be like, oh yeah,

I know that retail armies exist. If they get marshaled, it can be really bad. So you do have to be a little bit careful because even though they might have like crazy memes and they might be like somewhat silly they are powerful but you can quantify the power you can look at what has what what capital has actually been marshaled for game stop is it being sucked out of the game stop army because if the entire game stop army like leaves and you see game stop selling off by 80 percent and now they're buying into micro strategy or something like you probably have a problem yeah and wall street has always been

pvp yes and in this case you're just it's like p it's like p it's like you know, one group versus 30,000 individual. Well, we might want to close out with this post from semi-analysis about PVS person versus shrimp. They're breaking it down. He says, it has become extremely trendy among some San Francisco artificial intelligence researchers to donate to shrimp welfare.

They estimate that they help improve the welfare of 1,500 shrimps per year for every $1.5. are donated. Why do they donate to shrimps? They claim that it is the most cost-effective way of reducing suffering of sentient beings. Note that shrimp welfare, that the shrimp welfare non-profit actually does not prevent shrimps from being killed, but instead promotes the use of electrical stunning as a more humane slaughtering method that aligns with the goal of reducing shrimp suffering. And so I had an idea for a new semi-analysis product, shrimp max, and what they'll do is, every night they will take thousands of shrimp and execute them using hundreds of different methodologies, and then they will quantify the amount of suffering on a nightly basis, just like

what they do with ClusterMax. Yeah. And so we could know for sure, because, sure, today, the SFAI researchers, they might be, yes, electrical stunning is better than the previous method. But what if there's a shift tomorrow? And tomorrow, electrical stunning is not the best method. They need to have real-time data. They need charts.

They need graphs. They need a dashboard. They need cluster max for shrimp suffering. I didn't believe this was real, but shrimp welfare project. org seems to be, they say you can track their shrimp act. Shrimpact. Well, I like a good. There's 440 billion shrimps are farmed each year.

Yes. That's more than five times the number of all farmed land animals combined. I didn't know shrimp was so popular. This makes shrimp welfare an area of high impact where your involvement can lead to great change. Yeah. Are you pro-shrim, anti-shrim, Tyler? Do you mean like shrimp welfare or shrimp in general?

Do you like shrimp? Do you like eating shrimp? I enjoy shrimp. Do you think the shrimp welfare thing holds any water? I think it holds some water. At some point, it's like... Should we have a global shrimp welfare tax? It's like an extra dollar per every time you buy,

every shrimp you buy is like a dollar on top just for shrimp welfare. So if you have a nice meal, you might be paying like a $20 premium, but, you know, thousands of shrimp are going to live better lives because of it. Well, it's not that they're not going to live better lives. They're going to die better deaths. It's a noble cause. It's a noble cause. It is funny to go through all of this.

Yeah. Like you clearly care more about shrimp than probably anyone on earth. And you're still like, yeah, they must die, but we have to do it. You've got to be nice about it. You've got to be nice with it. It is crazy that you just don't save them. But I mean, I'm sure the spreadsheet's mathed out in a certain way and the rest is history. Should we close out with this iconic mid-century home?

It is apparently for sale. Look at this photo. Beautiful. And I think someone might have already bought it. Adam Mayer said, I'm going to buy it. This is in the SF gate. Have you seen this home before? I think I've seen some photos.

It says after 65 years, LA's most famous mid-century house hits the market. It's the first time it's ever been sold. Somebody's owned it for 65 years. The pinnacle of mid-century modern architecture with floor to seal ceiling, ceiling glass windows proffering a panoramic view of Los Angeles below. The home has been carefully preserved for decades. I just like that, oh, it's for sale for $25 million. I just like that for once we get a real estate listing that kind of just becomes the current thing.

And I saw a lot of people posting about this. We obviously love the mansion section. Love posting about real estate, talking about real estate. We find real estate very fascinating. And I was happy to see this breakthrough in such a meaningful way. Anyway, lots more news, but we will get to that tomorrow. Thank you for tuning in. Thank you for supporting us, listening to this show.

Wherever you listen, leave us five stars on Apple Podcast and Spotify. and we will see you tomorrow, a little update on the schedule. We will be off on Wednesday and Thursday, no show Wednesday, no show Thursday, but we will be back on Friday for Black Friday, and we will be taking you on a whirlwind tour of the e-commerce world. We have some very exciting stuff playing for that. So we'll see you tomorrow, and then we won't see you Wednesday, Thursday. Have a wonderful afternoon and evening.

We'll see you at 11. Goodbye. Cheers.