You're watching TBPN, and today is Monday, September 28th, 2026. We are live from the TBPN Ultradome, the temple of technology, the fortress of finance, the capital of capital. Let me tell you about Ramp.com. Time is money, save both. Easy-to-use corporate cards, bill pay, accounting, and a whole lot more. Let's check in on the road to Christmas. How many days we got? I think it's 87 days until Christmas. 87 days until Christmas.
And you can feel it in your bones. You can. You can.
You can. You can. Well, thank you for tuning in to TBPN today. We have a great show. We have a bunch of great guests coming on. It's been too long.
Get down. It has been too long. Just a couple more sleeps, but good to have everyone here with us in the chat.
Anyway, the most viral video I saw over the weekend was the parody of Dario Amodei on SNL. That felt like it really took over the timeline. Martin Casado said it honestly downplayed how ridiculous the rhetoric and broader situation is. It was a funny clip. There were a bunch of interesting nuances. I mean, first, it's like, okay, clearly the AI rhetoric and the public discourse is getting to a new height to the point where a lab leader of a private company can even be parodied on SNL because the audience understands who that is. So it's interesting to see that, like, okay, yes, Dario's done enough on Oprah and CNN and NBC and ABC and all the major news networks to be relevant to the broader audience. Yeah. And because it only really works, like, you can't, like, they could never parody... There's a lot of people in tech that we know that wouldn't work as, you know, SNL punching bags, but he's made it. He's made it.
It says a lot that the, like, the sweater vest thing is iconic enough to be, like, a part of a character.
I wonder if he'll stick with that or switch it up. Because it's not his leather jacket yet. It's not his Hawaiian T-shirt yet, but there is such value in leaning into, like, a consistent thing, but I don't know. We'll see. The other interesting wrinkle here is that while Dario is at the point where he is enough of a name that they can do this bit and make jokes and it's relevant to the SNL audience. Obviously, it went way more viral on X than I think it probably did on YouTube, but when Michael Che is introducing the segment, he sort of stumbles over Dario's name a little bit. Like he's reading it from the cue cards and he says, like, 'And this is Dario.' And then he pauses and it seems like maybe Dario Amodei is not fully a household name in the SNL writers' room yet in the sense that they were like, 'Okay, who is this guy? Let's look him up. Okay, get him on the cue cards.' I don't know. Maybe I'm reading into it too much, but it felt like this was certainly like a crossing the chasm moment, like the first time everyone gets to know Dario. But very, very funny and it really illuminates a little bit of the complexity of the discourse, the irony of saying, you know, somebody's got to save me from me, basically. But let's play a little bit of it. See what you think.
You know, thanks for having me, Michael. I may have been alarming in my recent interviews, but I want to assure you that if we can pressure lawmakers to create guardrails, we will be able to stop me. Dario, can you just explain how your company isn't going to kill us? So, so, so, so... Not the most flattering depiction. That's a really hard question. It shouldn't be. So, you know, come on, Dario. Come on, Dario. You can do this.
Is that an explicit reference to, like, the ring? Like Gollum, like where the voices in your head? AI, the devil, it maker. What? It seems like it, right? No. Well, look. Well, AI is the devil and I major AI players about this extensively and we are all on the same page here. We do not condone what we are doing. Shouldn't... it's very, very tough to thread the needle on this messaging. Cancer. Yeah. So, put it this way. In 10 years, there is about a 10% chance that cancer won't be a problem for anyone. For anyone? I don't like the way you're saying that, buddy. I mean, how is this a good thing? Look, so Jay, it's simple. If you could drink a milk that made your life easier, but there was a 10% chance it killed you, would you drink the milk? No. Really? Yes. Well, what if there was a button that when you push it, everyone dies? Would you push the button? No. Really? Dario, are you okay? I'm sort of blacking out every time I talk. I'd love to start this interview over or, better yet, you know, my whole life. All right, Dario, just tell us what are the pros of AI. All right, land the plane, Dario. I can't. You can. We can hear you, Dario. No one can hear you, Dario. Say something reassuring, like what? Try the milk thing. I really like that.
Then is milk a reference to something or is that just random? AI is not a weapon. It's a tool for building weapons. And I urge you to urge me to stop. Bar everybody. Ridiculous. Let me tell you about the New York Stock Exchange. Just want to change the world, raise capital at the New York Stock Exchange. No, but yeah, budget downstream stuff from this. I mean, obviously, getting pilloried by SNL, not the first time it's happened to a tech.
Yeah. Right. Rite of passage.
Yeah. Yeah, for sure. They've done this for all sorts of people. Also, there's a number of tech partners... Yeah. Yeah, going pretty hard on him in some ways. That being said, he's always, like, incredibly fascinating to listen to and he has incredible clarity of thought and he does, at times, he'll, like, pause, you know, in a way that would maybe scare normal people a little bit. But the thing that I think he can do that will have the biggest impact is to stop.
No, I actually like the, I like the whatever kind of shawl sweater thing that is. Isn't it? Is it Brunello? Probably. Yeah. I think someone, I think someone clocked it, right? It's Brunello. So, you know, it's nice. It's quite a luxury. And I feel like a power tie, some shoulder padding would go a long way. Anyway, the biggest thing he can do is to stop smiling when he's talking about really bad, devastating things. Yeah. And so I'm sure that that's something... Yeah. There's something to be learned from, like, the way, like, Fed chairmen speak or the way, you know, the head of the FDA or the head of the military speaks. Like, we've talked to some of these folks and, like, they're very austere, very serious and our show is very lighthearted. We kind of have to, like, click into a different mode when you're talking to somebody like a Michael Kratsios or someone who's, like, dealing with the weight of something big and puts up a wall that is a, it is performance, but it serves a very valuable purpose, which is, like, I take this very seriously. Or you can go the opposite way, the Mark Zuckerberg way, and rip it and just be like, 'Yeah, I'm having fun with AI.' Which is honestly, like, it feels like people are, like, 'Yeah, I kind of want that energy a little bit. Just give me the white pill.' But I don't know.
Yeah. And again, I don't think that Zuck's approach is at all appealing to the people actually pushing the frontier. No. Right. No. Like, no, no one's going to, no one's going to take, no, no researchers, like, take... I mean, there are, there are some reasons, dating back to basically Demis being, like, you know, that moment where he realizes Zuck is just as excited about VR as he is AI, right? Um, anyways, um, I hope that, um, maybe VR will be bigger than AI, Jordi. Maybe one. Wouldn't you like that? I would love that. Wouldn't you like that, John? I would love that. But yeah.
Anyway, so from SNL headed straight into dinner with the president. Sunday. I mean, he wasn't actually at SNL. But yeah, it's like you wake up to that on Sunday morning and you have to imagine that Trump sees it. But Trump has been both, he's hosted, he's hosted SNL multiple times and he's also been made fun of on SNL for years. Like when he gets elected, they basically hire an actor to come and be him for four years, for the entire tenure because they're going to be doing the cold open with him, like, basically every other week. And so maybe there's a little bit of camaraderie. They're like, 'Hey, we've both been through this, Dario. It's not so bad. It's actually kind of funny.' Or maybe they're both like, 'Yeah, the news really is fake.' We got, we got... But yeah, as critical as it is, at the same time I think it's something that all founders should aspire to, right? When you're starting your company and investors say, 'Well, where do you see yourself, you know, where do you see yourself in the company?' and getting on the biggest stage, perhaps, perhaps... Andrew Curran says President Trump has personally invited Dario to a private dinner tonight. That was yesterday, so that happened on Sunday night. This will be the first time they've actually met one-on-one. Big chance to patch things up or crash the plane into the mountain. They're actually running it back tomorrow night. Tuesday night there's a dinner in D.C. for all the AI leaders with the White House again. So, I mean, the pace of this stuff is really, really, really picking up. There's meetings every week now, basically. And yeah, I mean, there's lots to work through and no one is going to get to put their head in the sand. Not even the comedians in SNL will have to chime in as well as the politicians and the lab leaders. And so, yeah, despite Dario being excluded from, it seemed like he was excluded from that White House dinner with Xi Jinping and that basically every top tech leader was invited to, he was noticeably absent. President Trump, you know, followed up with a one-on-one dinner with Dario to hear his side of the story. And I wonder what type of leaks or reporting will come out of that. You would think that that would be going on today. Maybe there will be some news or maybe they'll put out a joint statement. Maybe everything went really well. But I mean, both.
Or a joint. You could see Dario sign up for Truth Social.
Sure. Or UFC courtside or ringside.
Ringside would be pretty electric.
Doing the Nitro Circus thing before UFC. That... I want to see Dana White, Dario, Trump. Yeah. Yeah. Just squatted up. You think he'd be doing slap boxing before, too? Oh, yeah, for sure. For sure. You got to do the full suite if you go all in.
Anyway, so there's also interesting timing that I don't know if this, how timed up this was, but every day there's another proposal for how to manage AI safety. This one feels like the most formal consensus exercise from a group of top AI researchers. It's published through the University of Cambridge and it's a good list that pulls from all over the industry. It doesn't feel like there's any lab that has, like, a finger on the scale. Not that that really matters. I feel like if Demis comes out with some great thing and Anthropic and OpenAI sign on and, you know, there's a lot of energy, like, people will just be like, 'This is great. We should take it seriously.' But this one was, this one was interesting because it was notably, like, diverse in the researchers. I didn't even know that there were this many researchers that didn't look at, that didn't work at labs by now. But there, in fact, are a lot that signed on. So Geoffrey Hinton, also a Nobel laureate but Turing Award winner alongside Yoshua Bengio, Andrew Barto, industry leaders from OpenAI, Jakub, the chief scientist from Microsoft, Eric Horvitz, the chief scientific officer, and Jack Clark from Anthropic. So, this, you know, Sam was talking to... was it, wasn't Axios, it was, he was interviewed and he was saying that, like, maybe we're going to sit down and get beers. Basically, you remember this whole thing? 'Why don't you guys just sit down and have some beers?' That was the proposal. And he was like, 'I don't want to leak anything that's going to happen, but we are thinking about, thinking about it.' And, you know, he's like, 'You can't just...'
Personal agents are talking about getting together for a beer.
Yeah. But it was very funny framing, but this is like the start of that, a start of more collaboration between the labs, between the academic community. And what's interesting is, like, they go a little bit further. You know, Tyler over there, his criticism has been, like, everyone comes out with an AI safety proposal and it's basically always, like, we got to talk about it. And this is a little bit more concrete. So the first one is they want to set in place the ability to monitor the level of AI-enabled R&D at labs. So what percentage of the code, what percentage of the dollars, what, like, give us, give us metrics on how much of the AI R&D is being done by artificial intelligence. Anthropic, OpenAI, they've all sort of, like, mentioned this in blog posts that it's happening, but there's also this, like, vague jockeying because it's better to be further along on the RSI curve for financial reasons. So there's not really an apples, like, no one really feels like they have a strong handle on, like, whether there's more RSI going on at OpenAI or Anthropic. Like, it feels like they both have very different metrics. Like, OpenAI says, like, we have an intern level, and then Anthropic will say, like, we had Claude spin up a thousand ideas, and, like, you can't really apples-to-apples those. The proposal is, like, let's get some metrics in place so we can actually monitor this thing. Then they also have, again, it's sort of, like, we need to talk about it, but they want the government to create firm plans for responses to potential theoretical bad scenarios. So, like, if there's a big cyber attack and an agent goes rogue and brings down the internet and hacks into a bunch of stuff, like, what did, what will the government do? What's the equivalent, like FEMA? Like, we have a response for, like, there's a big hurricane hits town, you bring in people, there's food, there's shelter, there's, like, a plan. And we don't really have that for, like, the internet went down because, like, an agent took it down. Same thing for biosecurity. They have a couple others, even, like, economic dislocation, that type of thing. So they want, they want the government, they're not proposing the solutions to those yet, but they're saying, like, this is where the government should be prepared. So, I don't know. It's a good read. They have an executive summary you can read and then the full paper is, is pretty digestible. It's not...
Did you see that China has places you can go to to simulate being in a disaster, like high, you know, super high winds and, and, um...
Flash floods, things like that? Should we have, like, AI disaster simulations where, you know, you can be in a data center that's going rogue and kind of work to shut it down with friends? Yeah. Just a kind of, like, live-action role-play through it. I like that. I like that. I'm kind of treating Modern Warfare 4 like that. I feel like that's getting me enough cycles of, like, the simulation of, you know, you're already playing against bots. You're already shooting, shooting AI, basically. So getting some drills in.
Yeah, I know it's a joke, but I wouldn't actually be surprised that a future first-person shooter or, like, one of these open-world games is centered around the Terminator scenario.
There was, there was a game, there was a game, I forget what it's called, but yeah, like, you drop in with three people and you're on this, like, planet. There's one that's Starship Troopers themed, so it's, like, the bug planet, and then there's another planet that you can drop into that's, like, the Terminator planet, and you're going up against, like, they basically, like, ripped Terminator's intellectual property. Like, they look exactly like T-800s.
Why is Tyler not back in school doing the best internship in history? He turned it down. I turned it down. He actually turned it down just to grind for this. Yeah. Well, the things will never be chill again. The doomers who shaped the AI safety freakout. The Wall Street Journal has an article here about some early Anthropic employees who are considering buying remote land in the United States to relocate to if AI goes seriously wrong. They're going bunkies. They're going bunker mode. Yeah. Employees also trained doomsday scenarios and preparation plans in a private Slack. While Anthropic workers have their own office space inside Berkeley's Constellation AI safety community, the wider safety scene has discussed iodine pills, remote islands, and electromagnetically shielded desert bunkers. Interesting that no one's trying to go to space. I would have thought that would have been, that was George Hotz's original pitch. He was like, 'If AI goes crazy, I want to be on a spaceship leaving the light cone as fast so AI can't catch up to you. Blast shield behind you and then you'll be safe from the intelligence.'
Could just say he's not very AGI-pilled because obviously the superintelligence would just figure out a way to just go faster than the speed of light and catch up.
Yeah. I don't know. I don't know. I mean, if you go the maximum speed, they can't catch up ever, right? So then you are safe, but you got to accelerate to exactly one c.
This is, like, an interesting example of, like, an actual answer to the, to the Teller Count question, which is, like, if you're, if you really care about doom, aren't you making financial decisions? Cover up, get a lot of debt and buy a bunker, buy remote land with a massive mortgage.
Yeah, because usually the answer to the doomer thing is, like, well, I'm not going to be able to get to my bunker because the, you know, bioterrorism will already have killed me or something.
This is an interesting instantiation of this.
The thing that always, the thing that always just, like, breaks down with these scenarios is, like... I, you know, every guy has sort of played out the scenario in their head, right? And where are you going to be, where are you going to be best off in some, like, remote area where you don't know any people that actually live there? Right? Let's pick a, let's pick Wyoming, right? You really think that the world is, like, falling apart and you're just going to head over to Wyoming and set up shop in a community that knows you've been building your bunkie there? Yeah. Right. And they're just going to be like, 'Oh yeah, like, there's a bunker over there. It has everything you need in it to survive.' And you know what? But, you know, they paid us to set it up years ago and let's not, we shouldn't, we shouldn't just go occupy said bunker for ourselves. So the prepper community has a term for this. They say don't be a loot drop.
You're a loot drop if you do this. If you go and set up a really luxurious bunker with a bunch of supplies, but you don't actually know the community. You're not integrated into, like, everyone else.
Yeah. The fabric. So ideally, it's, like, you want your team and everyone else to all be in it together. And so that, you know, there's no, there's no squabbling. But yeah, you can just set up a loot drop and then everyone's like, 'Oh yeah, I installed a crazy safe over there. Let's go get that as soon as the crisis happens.' Yeah. I mean, there's a whole bunch of fun game theory you can play through there. I would, I would personally, I think we need to regulate because I would personally, I would hate to be in a self-defense situation with a bunch of rogue agents. And imagine if all the Waymos went rogue and we were just forced because of self-defense reasons.
You and all your best friends all just drive around and just have to shoot them out of the windows of our cars, like, all day long. Like, that would be, that'd be a nightmare.
That is a nightmare. You would never... You've never even...
If I was in a self-defense situation, I would do it, but it would, it would, it would pain me.
Totally. Totally. Yeah. Totally.
Let me tell you about Console.com. Console builds AI agents that automate 70% of IT, HR, and finance support, giving employees instant resolution for access requests and password resets. Also a Journal piece on... what else happened... on Jaan Tallinn. Skype founder. Skype founder. Big investor, early investor in the labs. Take a wild guess where he's from. Estonia. Yeah, Tallinn. He's from Tallinn. Anyways, also the primary funder of all, I would say most of the sort of AI doom... the EA movement. He also dances hip-hop. I wonder what that, how they get that scoop. It's interesting. Interesting. He warns about racing to build a technology that could end humanity. Yeah, there's a funny, there's a funny anecdote in here about champagne. I like that part of this. I didn't like the doom part, but I did like this. Said they debated machine consciousness at 40,000 feet over flutes of champagne. See, white pill. Anyway, that's enough of that article. Let me tell you about Public.com. Investing for those who take it seriously. We got stocks, options, bonds, crypto, treasuries, and more with great customer service. Let's talk about some personal agents.
Noah Shinn, founder of Instinct, broke his silence finally with none other than our friend and legend Patrick O'Shaughnessy. I was trying to listen on the way in to the gym this morning. I got maybe 30 minutes through. Patrick pulled out the highlights. He summarized his own podcast before you could summarize it. He beat you to the punch. Wow. He said, 'A list of surprising and mind-boggling stats from this conversation. 1 billion a year in transaction volume, still invite-only.' I can't remember if that's total. I think that's actually total volume to date. I don't think that's, like, run-rate volume.
It says a year. So that would be...
Well, that's what it says in here, but I just remember from the episode, I heard it as...
Well, yeah. I mean, if it's, if it's today, then it's higher.
That's what I'm saying. Could be, it could be, like, higher.
I read that as $1 billion of annualized... that they are at $1 billion of annualized transaction volume. Yeah. But I guess it's maybe even higher. I don't know.
Growing 10% a day. Spent zero dollars on marketing. They have this, you know, pretty amazing flywheel where people post screenshots all the time. That's doing a lot of marketing and...
Wait, a lot of the, but a lot of the screenshots are, like, a horrible accident, but maybe those actually drive curiosity and, like, people being, like, well, I'm not going to make that same stupid mistake and the service is probably getting better. You think those actually help?
I think those help. People are curious and they're, like, I want to see it and I'll be safe. I won't give it my login details, but I'll still...
No. And it can't be said enough. Yeah. A lot of people within tech have experienced AI doing things for them, like making a product. Yep. Most people outside have not experienced that yet. And it's just a magical thing. Yeah. Totally.
40% of users shared a personal credit card with Instinct within 3 weeks. Compute demand is doubling roughly every week. At 5 to 8% daily growth over the months it takes to bring new compute online, Instinct would reach about 100 million users on Instinct's own A/B test. It matches Opus 5 performance at a fraction of the cost. Noah spends about 40% of his time on compute. Buying compute at the last minute costs 3 to 4x more. Makes sense. Background work runs on setups that are 3x, 5x, and 8x more efficient on the same compute. 50% of Instinct's transaction volume is travel.
Instinct plans to take a cut of what users buy. Take rates. Shopify is 2.5 to 30%. Amazon is 15%. Apple is 30%. Noah's personal goal is to keep Instinct free for everyone for life. Some small businesses run their entire back office on Instinct. That's interesting. Users who connect at least one piece of sensitive information retain at about 80%. Just give me one piece. Just one piece.
Just give me one piece of sensitive... Tell me your darkest secret. Skeletons you have in your closet.
Some users send more than 90% of their messages to Instinct by voice. Say an app makes 70% of its money from ads and 30% from actually selling you things. Agents could wipe out the 70%.
But Noah thinks the 30% gets much bigger because people buy more when buying takes zero effort.
This is a big question for me, right? Because already buying something on a Shopify store takes close to zero effort. Like, the checkout is actually down to if I land on the website and I know what I want. Yeah. I can check out within... And again, Shopify doesn't have an ad business. Are not fun to shop for, though. There's a lot of stuff, especially if it's, like, I was thinking about, like, cable management. I wanted to get some cable management stuff for my desk. I'm like, I don't really care about the brand. I don't care about the flow. I just want, like, you know, a sweet... I needed some, like, command strips, maybe, like, a little tray. Like, whatever, like, the modern solution is to, like, a couple cables, a power brick. Like, just go and get something that's well-reviewed and, like, a suite that will figure that. What kind of voice are you doing? Instinct launched to about 200 friends and family. Invites have sold on eBay for around $300. More than 50% of Instinct's traffic doesn't run on iMessage. Instinct has proactively called Noah three times. That's, like, when it's doing something and it has, like, a deadline. It's, like, Noah gave the example of, like, hey, you really have to sign this, like, within the next 5 minutes. Are you available? It'll just call you, which is cool. Raised on 10. Yeah. Yeah, less than a month after raising... two and a half. So Noah, Noah is 23. So very, very good summary.
Also same colleges that I went to. He went to Northeastern.
There you go. Yeah. So I guess initial reactions... I, he, they talk about, like, remaking, like, the entire internet, right? And the example that they talk about is restaurant reservations. And Noah made an interesting point about how restaurant reservations to date are all effectively, like, first-come, first-served. So, like, if you just happen to be the first person to get there, you get the prime time slot even if later someone else is making plans and it's, like, you know, he gives the example of, like, an anniversary or, like, a special birthday. And he's making the point that, like, the restaurant should be optimizing for the person with the special event happening and there should be more of, like, basically a negotiation in some ways or, like, the user should have an opportunity to sort of make their case. Okay. For why they should get a table during a time where there's more demand than supply. Okay. Yeah. And I think that, I think that that makes sense. That being said, I think when everyone has a personal agent that's calling and, like, hammering restaurants and, like, trying, like, I think, I think it's very possible, like, his argument is that the restaurant wants, like, the whole space to just be filled with people that are, like, it is the most special moment of their life.
25 proposals, everyone clapping for five seconds at the restaurant.
And it's been a while. I mean, it's been a very long... I mean, I worked in a restaurant in high school. And it's been a while, but my read on that was that it would be insane if every single table was, like, it was, like, the anniversary or the birthday. Like, restaurants are not generally, like, set up to, like, give that. So maybe it's, like, if, maybe it's, like, a feature, not a bug that the reservation, some, at least some of the tables are, like, hey, this person's a regular, like, they're just, sure, they're just, like, they just plan ahead. Like, maybe that's, like, actually, like, a higher-quality customer on the average night than somebody that's, like, coming in and this is, like, their special moment. So, okay, I got a question for you. Yeah. If you're at a restaurant and it's someone's birthday and the host brings out a birthday cake and they're singing, how close do you have to be to join in on the singing? If you're at a table over, will you start singing Happy Birthday?
I could be at the other end of the restaurant.
Oh, you're a birthday singer no matter what.
Okay, I like that. I might be belting over four or five tables.
Yeah. Trying to get them involved. You don't necessarily need to be at the same table.
Yeah. So potentially this opens the, this creates the opportunity for a restaurant that's just constantly singing Happy Birthday on repeat the whole time you're there because everyone's having a birthday.
Yeah. I mean, there are some restaurants that sort of lean into, like, they're romantic. So they're the place where anniversaries and proposals happen. And then there's other ones where it's, like, yeah, birthdays happen there and there's, like, kids' birthdays. Like, you go to a CPK, I'm sure there's, like, many more, like, kids' birthdays than, like, proposals all...
Yeah, it is interesting, though, because a lot of our, like, so much of, like, so much of being a consumer, like, the benefits, the highest-quality experiences typically go to the people that plan the farthest in advance, right? So, like, if you're booking a hotel right now for, already for, like, the Christmas holiday, you're at a disadvantage, right? Because there's some people that were booking, like... hence why we tell you about the road to Christmas, start planning ahead.
And so, anyways, it will be interesting if personal agents kind of rework that. I'm not, I don't think it's sustainable for, like, every single basically reservation or booking to be, like, a, um, open to basically, like, negotiation, right? Completely disagree. I completely disagree.
Because I think that the reservation system will be an agent as well. And there might be a country of geniuses in a data center negotiating against another country of geniuses in a data center. It sounds crazy, but I'm not joking. Like, I actually think this is what will happen every time. If you say, if you say, 'Get me, I want to have dinner right now. I want to have dinner tonight with Jordi.' Like, there will be, it will spawn a bunch of agents, call a bunch of restaurants. It will be annoying for the restaurants that aren't on board, but very quickly all the restaurant, you know, phone systems will be equipped to handle it. On the other side, they will all share all the relevant information, negotiate, find the answer to any questions, oh, food allergies, oh, birthdays, how many birthdays are there? You want to go for your birthday, but you don't want that many birthdays to be there. Whatever your preferences are, all of that will get penciled out and it will be just more seamless matching. And I don't think that's a problem at all. The thing that I'm uncertain about is, like, are we going to move into, like, a SeatGeek regime where it truly is, like, surge pricing and the best restaurants might wind up being, like, oh, yeah, you're sitting down, but you had to pay, like, two grand for that reservation. And then there are some that are, like, it's, we're selling the food, like, 1% above cost because, like, there's no, we want to fill seats right now. And this...
Yeah, I'm just thinking about, like, I'll use a... there's a restaurant near where I am that people will plan months and months and months and months out. Yeah. For, like, special occasion. Yes. So, like, if you go there... Yeah. There will be a lot of people that are, like, celebrating an anniversary or a birthday. Yep. And then there's maybe four or five tables every day that the manager kind of just has on lock. Yeah. And if you have a personal relationship with them, they're going to be offered up.
Those tables are never in a restaurant that has way more demand for tables than supply. The restaurant is not going to ever take those tables and be, like, 'Hey, we're just, like, handing these over to our AI agent.'
But they were never going in the booking system anyway.
Yeah. Yeah, like, they were never going on OpenTable or whatever.
But I'm just saying, like, in that case for that restaurant in particular, like, the system is working as intended, which is, like, if you want to, if you want to be with us for your special day, plan ahead.
And we also want to take care of, like, friends. Yeah. This is a good, this is a good, that's a good point. That's a good point. And it's just, like, that extra inventory is not going to shift over just because you have an agent that can make your case.
Yeah, I think some of the friction in hospitality is designed to reward people that plan really far in advance and people that develop a real relationship with the business and the people that work there. And so I'm just wondering how this sort of new dynamic fits in where everyone... In the first case, I have my agent book my birthday dinner, like, six months in advance very easily because it knows when my birthday is and knows that I want to go out to dinner and get me... And that, but that's, but that's using the existing... And then for the other one, it's tracking the owner of the restaurant on Instagram, sending him cigars and champagne regularly and potentially hacking and blackmailing him if he needs to go. No, I'm very, I'm very bullish on telling your agent, hey, I know that this special moment is coming up. Make sure that the second bookings open up.
But that, again, that fits into the sort of long-term planning category.
And it's like my, but it's not reworking the fabric of hospitality.
It's just sort of like reducing friction.
Yeah. I'm just saying, like, Resy and OpenTable and Toast and all these things, they're going to, they're going to build the sort of counter infrastructure for the onslaught of agent inbound that is coming to every business.
That, yeah, it will, it will quickly be resolved as soon as these things scale and they are, and they are scaling. And I'm sure that, yeah, I mean, I guess a good prediction is, like, there will probably be several breathless op-eds in the various newspapers about, like, AI agents are super annoying for my small business, you know, and it's, like, now I have to get some tool to solve this.
Which is kind of funny because we've all been through this phone. How much sympathy are people going to have that business owners being, like, I'm being flooded with demand for my service?
It'll be, it'll be, like, fake demand because the agents will be, like, checking and they'll be, like, 'Oh, well, there was nothing available at 7 p.m. on Resy via this API, but, like, maybe if I call I can get an extra thing,' so they're going to call everyone all the time and it's going to be very annoying. There will be, like, some frustration that is justified from this. Like, anytime that there's a transition, there will be frustration. Yeah. I think, I mean, a good preview of this is, like, what we all went through with our phones where, like, you used to be able to have a cell phone and basically just pick up the phone when it rang. And now it's, like, I have my phone in do not disturb 24/7. You can't call me unless you're already in my contacts. Like, I can't just pick up the phone randomly because I get so many spam calls. And even Apple, like, clearly admitted that, like, the phone lines have fallen, like, because they have a separate tab in iMessage for, like, unknown now and, like, it gets filtered off because they know that, like, the rails have truly broken, like, the spammers have won unless you wrap it in a filter. And so there will be significant spam filtering that goes on at the phone level. And that's probably already happening with, you know, just there's so many services that are calling restaurants and saying, like, hey, do you want to, do you need more supplies? Do you want to switch to this point-of-sale system? Right? And they don't want those to go through their, they don't want cold calls and sales calls going through the reservation system. So they probably filter those already. Anyway, Josh Constine. Oh, yeah, please. This is the whole Instinct pitch deck, by the way. And it's Christian Kyle saying, 'My three-month-old son is now twice as big as when he was born. He's on track to weigh 7 and 12 trillion pounds by age 10.' Very funny.
Such a viral post. This went super viral on Reddit. It goes viral, like, every couple months on Reddit and stuff. It's a wild post.
But I did want to set some comps around the Instinct growth numbers that are interesting. So, a billion dollars a year in transaction volume. We're not exactly sure if that's annualized or to date, but it's kind of the same. For reference, it took Stripe around 18 months to hit 1 billion of annualized transaction volume. Although Stripe also spent two years building in beta before launch, Instinct got to 1 billion transaction volume in closer to six months. But it's important to note that with Instinct, the transactions flow through third-party cards. So they're not making any money off of these transactions yet, but it's very clear to imagine where monetization might happen.
Yeah, at the very least they could just launch a credit, like, their own card.
Yeah, of course. And sign up, then they're getting that interchange fee and they don't need to get, they don't even, they don't even need to get anyone to opt in.
Yeah. And you might be able to... and it's an easier pitch, too, because you're onboarding this AI agent. You're, like, I've heard some pretty crazy stuff about this thing, like, yeah, I'd actually like it if you used your card as opposed to, like... it already has, basically, a computer, a phone number, email. Yep. Give it a credit card. Give it a credit card.
And so I'm sure they're working on that. And when they do, they're not going to make 3% because you do have to pay Visa and Mastercard, like, the underlying rails. But earning 1%, half a percent, quarter percent, even 0.1% on, you know, they grew to a billion. You know, you're talking about getting into, you know, tens of millions of dollars of revenue pretty quickly. And the team's just 14 people. And so you can sort of see why they raised a billion at 10 billion from Benchmark, Sequoia, Coatue. There's, you know, yes, you have to believe that they won't be, you know, killed by Muse and that they will continue to scale, but if any of the growth rates sustained for even a limited amount of time, it's going to be a lot of transaction volume, great business, and you can underwrite it pretty safely. And then there's still the offshoot into Amazon world, I think, because Amazon doesn't really have a strong play there, but could fit in very well with...
Yeah. They have a lot of potential dance partners and ultimately things grow really fast because they don't just, like, grow really fast for no reason. No. Right. So, no. Good.
And people have, and, like, people have been trying to do different personal agents for specific things. There's definitely been other pitches for, you know, doing commerce, doing flight booking. I'm shocked by that 50% travel. That number seems really, really high. What was the exact stat? 15. He said, Patrick summarized it as 50% of the transaction volume is travel.
Okay, so volume, because when you book a flight, you might be spending a grand, but when you book a restaurant, you're maybe putting down 10 bucks, 20 bucks, maybe 100 bucks. Like, you might not put down anything, right? And so travel, it's not that 50% of, like, economic activity, but that's where they're making money. And so, yeah, that's another way where they could build, like, an Expedia-type business where they're earning fees on top of booking.
We got to give a round of applause to Tate Hacker who says, 'We're so early. Everyone will be using a full-time assistant of compute like this within a year.' And the reason I'm clapping is because it's a parking ticket for not just any car that Instinct took care of. It's for an R, a blue R8.
And I just thought this was a wonderful...
Way to sort of share a wonderful car, potentially flex.
Hopefully, yeah, hopefully it's a V10 with the gated manual. But a very underrated, underrated car.
What is this? This is a very detailed parking ticket, but let's get taken to this. Paid 50% on your personal visa. Pretty cool.
Like, we get it, dude. We get it.
You got a parking ticket in your R8. We get it.
Yeah. Anyway, let me tell you about MongoDB. What's the only thing faster than the AI market? Your business on MongoDB. Don't just build AI. Own the data platform that powers it. There's drama in MongoDB world because poaching.
You want to get to that later?
Let's get to that later. Let's just close this out. So, apparently Adidas is running anti-bot software on their store. You ran into this like pretty recently, not even with Muse or Instinct, but just researching with Codex.
And you were... thought it was pretty funny because you were actively trying to give a company business. I think there's a bunch of historical reasons you would want anti-bot software. And then there's now a bunch of very real new reasons why you need to update that and figure out a way to prevent sort of like bad bots.
But support like actual consumer bots.
So, Gary Tan says, 'Anti-bot is for losers.' TBH.
Really got him with it. He really owned him right there.
Just slow. It's ridiculous.
Other people, the chief economist at Apollo is warning that mass adoption of AI agents could trigger a bank run as they optimize user investments. Gary Gensler spent his last few years as SEC chair warning that the exact same problem will hit the markets. The pursuit of algorithmic perfection at scale destroys system stability. Personal finance agents making simultaneous optimal decisions could potentially cause a massive flash crash. Agent-to-agent causal coordination with no messages passed between them will show up in a lot of places. Markets will get the big headlines, but the huge number of clever advisers making very similar choices will impact all of human society. Blah blah blah blah blah blah blah blah blah. Human laws were not written for this. They were not written for a lot of what is coming.
It is about to start happening very fast. Credit to Catrini for basically being the first.
That was a different thing, but yeah.
I don't... no, it's the exact... it's similar, but... No, it's just this idea of like there's a lot of aspects of our economy that currently like businesses that work because of friction.
Sure. That friction in some ways will be removed when you have an always-on agent that says, 'Hey, you haven't touched this product in a very long time. Do you still want to be paying for this?' etc., etc.
I don't think this is... I don't know, like my banker will just hit me up and say, 'Hey, I'm just clearing this with you. I'm going to sweep funds into a money market account or T-bills.' Like it's not like... a good banking experience is already doing this proactively for you.
Now there's of course people that maybe they just are set up with like a simple checking account and they could be moving those into a higher yield account of some sort that aren't doing it and maybe the agent starts doing that. But like, fundamentally a great banking experience, they're like doing this on your behalf, like they are aligned to you.
Yeah. So I don't think it's as like...
Yeah, I would be very hesitant about how fast this adoption can happen because when someone says, 'Dang it, if ordinary people get fair deals and aren't ripped off, it's the end of...'
Yeah. I wonder, like the personal AI agents are taking off, but like we're still in like single digit millions of people doing them. It's going to take time for people to adopt them and then they need to actually authenticate and add these things, link their bank accounts, and then presumably the first version of this will be the agent coming to you and saying like, 'Hey, is it okay if I do this?' Second order will be like, 'I just did this. I know it's okay. You trust me.' And that relationship, that trust, I feel like that's going to take like years to build up and diffuse through. There's still plenty of friction in the economy. I think it's going to take a while for people to onboard to this fully. I mean, people were... it's the same idea of like when ChatGPT came out, people were like, 'Google search is dead. No one will Google search. Google's search revenue is going to go to zero like next year.' And it was like, yeah, it was a reasonable replacement, but it took a little bit longer even though it was growing so fast, faster than any consumer product ever, incredible revenues, blah blah blah blah blah. It still wasn't that fast. And so it's the same thing where it's like, okay, if this happens over like a couple of years, like banks adjust, money market funds adjust, like it's not the end of the world potentially, but we'll keep monitoring it. Anyway, let me tell you about Railway. Railway is the all-in-one intelligent cloud provider. Use your favorite agent to deploy web apps, servers, databases, and more. Railway automatically takes care of scaling, monitoring, and security. Zuck, what happened?
We believe superintelligence will create significant new opportunities for all people and businesses. Meta already serves billions of people at scale and helps hundreds of millions of businesses reach customers. Today, we are starting the next major pillar of our business, Meta Enterprise Platform, to help businesses use AI to grow and transform in new ways as well. Meta Enterprise will use our strengths that few other companies have: advanced models, leading agents, large-scale infrastructure, and years of working closely with many businesses. Initially, we will focus on bringing our full technology stack, including the Muse Agent, Meta Business Agent, Muse API, Muse Code, and more, to businesses and developers to help them grow. To lead this effort, I'm excited that CJ Desai, former guest of the show, will join Meta as Chief Enterprise Platform Officer, reporting directly to Zuck. CJ is an experienced enterprise leader with a track record of building full-stack software and delivering results in AI infra, business applications, and security. Blah blah blah blah blah blah blah blah. Anyway, so here's how I think this is going to go. I think that they created this new business unit.
They still have a lot to prove on if they can... Sorry, if you scroll down there's some wild comments. Anyway, continue.
Anyway, they created this new business unit.
They're talking about apps and agents and all this stuff. I think what's really going to happen is they're going to do a big compute deal.
And then it's going to look like this sort of unit is doing it. It'll quickly go from let's say zero to ten billion a year of revenue. They can sort of use that. They don't have to necessarily break out initially all the where that revenue is coming from, but they can say like our enterprise platform has gone from zero to ten billion in...
Yeah, this was like the joke SemiAnalysis was sort of trolling with around like, 'Oh, all the growth in Google Cloud Platform is selling TPUs.' It was like a little bit more nuanced than that. But there's a bunch of ways to wrap up revenue into an enterprise platform that could be tokens, could be doing inference, could be selling compute raw bare metal deals, could be actually selling services. It'll be interesting to see where it goes. It's still an interesting offloading valve because there's some people, and we're going to have a niche from Andre and Horwitz on the show at 12:35 to talk about personal agents because there's been a lot of debate over like how compute intensive is the average Instinct user, is the average Muse user. Is it like $10 a month? Is it $100 a month? Is it more? Obviously, everyone knows it's going to get cheaper, but then people will be in an arms race for, 'Oh, well, this one gives you the best possible intelligence, so it's more expensive.' And so there's a lot of business model considerations there, but it's interesting that if Muse doesn't fully take off, then you do have the enterprise offtake for that compute. So shareholders are maybe more reassured that, okay, we're really optimistic about Muse, things seem to be going really well, but if it doesn't... I think it's important...
Because he finds it important to not... when he sort of capitulates and has to sell compute, he wants to be able to position it as like, yeah, we have this massive enterprise AI business.
Versus when SpaceX did it, it was like, oh wow, they don't have demand for their models. They have to do these compute deals. And I think that this positioning is smart.
Because even if he is selling compute, it's not like he doesn't have the same broader long-term ambitions of like actually building a durable, you know, AI business here with their own models, their own agents, etc.
Yeah. So I just think it's a better, you know, in... SpaceX needed to make a lot of moves really quickly to be in a good spot for the IPO. Had they not needed to move so quickly, they could have positioned it more like this, and I think it will sit better with people.
Yes. Yeah, I agree with you on that. I just think there could be a tension. Like imagine if they were like, 'Hey, we're really good at video hosting from Instagram. We're going to start like a video hosting platform.' Like wouldn't the Instagram team be like, 'Wait, we need all the resources we can for building out our...'
Yeah. But that was the same exact tension that SpaceX and Cursor have had with their own teams that are building models.
Yeah. But this is like injecting that tension right unnecessarily potentially.
Not unnecessarily. I think...
Why is it necessary? Why is doing this necessary at all? Yeah. Why not pure play it? Why not say, 'Yeah, we want to have the whole company aligned behind Muse, behind the products, and yeah, we have a bunch of compute. We're not selling that at all because we're the best to monetize it. Why would we ever give it away?'
Yeah. But of course, if they're deep in the enterprise and they have a product that flows through, there the margin will be there and it makes a lot of sense. But you are kicking off tension, right? Like we've seen this with other companies. I don't know. We'll see.
Oh, it popped. It was down 26% earlier. So yeah, it recovered a little bit. Yeah, CJ hadn't been there long.
That's a very great sign of respect, right?
