It’s the Context, Stupid

Speaker A: Speaking personally, you have to find interesting ways to occupy your time as you age. You know, clubs, you know, cultural activities, all that.

Speaker B: That's why I'm in Miami. The clubs, 11.

Speaker A: Oh, I'm talking about tennis clubs, golf clubs.

Speaker B: Oh, oh, uh, different

Speaker A: stuff like that. Yeah,

Speaker B: well, yeah, you could go that route.

Speaker A: Yeah. I'm, um, in the club.

Speaker B: The guy who, who like, developed 11, he's advocating against all through Florida these like, Islamic cent enters because he says, like, he's very biblical and it runs against his biblical beliefs. The guy runs like a massive strip club.

Speaker A: That's a different kind of church.

Speaker B: We went to a show last night. Theater.

Speaker A: That's a good thing to do in London. I love doing that.

Speaker B: Absolutely. They have great theater. And I went and I think theater is really well positioned because when we got in there, first of all, it was really expensive. It was like a couple hundred pounds, like a ticket. And we were like, at the back, it was the show Hay Fever. I'm like, look, this is new for me as a man of the people. And we were the youngest people in that theater by like, quite some. Um, I mean, there were a few younger people, but like, holy shit. Like the theater, you know, the audience is very old and I think, you know, because the only people who can afford the theater probably are retirees and they need something to do. And it was filled with old people. They were having a great time. And I think it's really well positioned because we're going to have a lot of old people with disposable income. And let's, let's be real, like the game is going to be longer, so there's more theater time for, for older people. So I'm bullish on theater because it's not an AI mediated experience.

Speaker A: Uh, just a couple updates from being out and about, as they say in Canada. First, now is a long time ago now. I was at Organic and we had a project with Sony, I believe, and we couldn't get it done. And it occurred to me that this was 25 years ago because I went to my friend Dan Gardner and Brandon, who I used to know, haven't seen in a long time, were the founders of a digital agency called Code and Theory that went on to do much of the. Much of the work in the media space for. Well, I used them extensively at Hearst. They did work for Conde, they did work for Time Inc. They did work for everybody in media. They were always excellent, always curious, always trying to kind of figure out what we could do next, you know, in the Confines of, you know, uh, ad driven media business. You know, why did you guys have to

Speaker B: use. Why did you. Can I ask why? Why would you have to use them organic built websites. You couldn't build a freaking like publishing website.

Speaker A: Oh, so, so why. The first chapter was the guys at Code and Theory were really good in something at the time which will maybe bring back memories called Flash Video.

Speaker B: Oh God.

Speaker A: And we were doing this really complicated flash microsite for Sony Ericsson and really it was kind of arbitrage. I went to them and I said guys got, is like 100 grand just fit do the work because I can't get it done for 100 grand here. And they worked, you know, crazy A uh, crazy amount to, to. And then you took credit

Speaker B: for it, right?

Speaker A: Probably, yeah. Uh, anyway, I learned, I learned to. I developed a great relationship with those guys, really love them and they invited me to their 25th anniversary party. That was at Chinese Tuxedo. Lots of nice folks. Really. Yeah. Uh, the food was terrific and it was great to see everyone and was reminded of, of you know, the old days apparently Code and Theory and its little uh, constellation of agencies owned by Staiwell. You know, like over 2,000 people growing, looking for people. You know, really nice group of accounts distributed like low cost production in South America and you know they do great work for lots of companies including Apple. Anyway, there's a little ad for, for you, Dan.

Speaker B: They just bought Freestar by the way.

Speaker A: That's right. And super fan and great guy. Mark Howard was there. Mark Howard is the COO of Time and he, he likes our podcast. Brian, you know him well, I really like Mark, but Mark, you know, and, and I, I saw

Speaker B: he's always leading from the front. He's always, he's always going for the, he's

Speaker A: not ashamed to lay it out on LinkedIn. He's, you know, he's going to do that.

Speaker B: That doesn't sound like compliment.

Speaker A: Yeah,

Speaker B: I've been locked down by LinkedIn. What's going on there? I don't know,

Speaker A: I don't, I don't go there. But the. I was thinking maybe LinkedIn for retirees. That's a good idea that something I could maybe spend more time on other than going to the theater. Having said that, Mark was very proud of making lots of jokes about their next foray at time and by the way, I did have to and I, and I, and I would publicly congratulate those guys because I remember when Mark and Jeff started a time and they, you know, brought the Playbook from various previous chapters in their career, including

Speaker B: Forbes. Forbes playbook. It is the

Speaker A: Forbes.

Speaker B: There's like, you know, like NFL has, like, coaching trees. There's like a Forbes tree.

Speaker A: There's totally a Forbes tree. And Meredith is the queen of that. Right. She's the queen of the four. The Forbes alumni.

Speaker B: That's the New York Times CEO Meredith Levine.

Speaker A: Yeah. And, you know, now the. The, you know, interim CEO of Conde Nast, Mike Perlis, also, uh, Forbes alumni.

Speaker B: The Forbes of vacation. I would have assigned that story in digiday, and it would have taken a very dark angle, but go on.

Speaker A: Okay, well, let me. Let me get to the. The story then. So, you know, I. I think that the time cover still is extremely useful in the media ecosystem. I think they made some decisions that, you know, maybe they ought to be free and go for reach sort of the, the semaphore play as opposed to. To try to build a critical mass in their newsroom m. To. To justify a subscription thing. You know, they. They pimped the lists. They've done nice things with. With newsletters and video and all the things. So good on them. Mark was saying, and. And we had to have a little laugh about it that. That they're now intercepting agents on their websites and sending them back home with advertising. So, uh, you know, this isn't unlike something we discussed on Field Notes, which is a distinct part of a website that's there to greet. You know, in this case, it was a book review that was there to greet the agent and give them information to take back to pops in. I hear a lot of

Speaker B: talk about this.

Speaker A: Yeah. So I want to get into it because I think it's a good topic, but a couple of little funny things about it. So we're going to develop a layer that isn't something we've ever done before, which is what happens when agents meet agents and how to manage those relationships and those transactions. But what he said was funny because Noah Goodheart was there.

Speaker B: Is that Jonah Goodhart's brother?

Speaker A: Yeah, yeah. They're. They're a team.

Speaker B: Uh, there are a lot of these like, like brother duos.

Speaker A: Well, these guys are pretty brother sisters, too. Yeah, These are like, you know. Oh,

Speaker B: they're winners. I mean, come on. They. They fly

Speaker A: brothers. Yeah. They plan with multiple helicopters. So I call them the Viewability Brothers. And they're brilliant guys. They made hundreds of millions of dollars selling a viewability company to Oracle called Moat. Correct. And I remember when Noah first came in and it was good that they got out.

Speaker B: Moats got eliminated. You know,

Speaker A: uh, well I remember thinking wow, this is interesting. He's going to go back and sell the viewability metric to our advertisers and that's going to be a pain in the butt and he'll be, you know, occupy a nice little place in the ecosystem. And sure enough he did enough to justify I think 850 to Oracle. And those guys were. And

Speaker B: then they killed it, they killed it probably within like a couple of years. I don't even know.

Speaker A: Yeah, that's a lot of money. And they were the big boys on the cap table so they made lots of money. So anyway their job over at Moat was to fight the bots. Right. You don't want bots pretending they're human beings stealing your valuable media money. Now they're creating a money a business to be friendly and manage the bots. So you know, 180 degree turn, change of times used to fight the bots now are welcome well welcoming them and their new effort, I think it's called Mobian M M O B I A m greets m incoming agents with friendly ad offers and they are doing work with Mark over at Time and Mark that this was. Can you, can you connect them with me?

Speaker B: I think we should do business together.

Speaker A: You should, you should. And now and you. Yeah. And you're very, you know, conversant in the topic of agent management.

Speaker B: Yeah,

Speaker A: so, so, so it's,

Speaker B: I, I hear this all the time, you know it's going to be advertising to agents and trying to manipulate these agents. I, I wonder it, what I don't know is if this is yet another short term arbitrage opportunity or if it's long term or if it even matters at this point. Like maybe you don't even think about what's long term. You're just like look, this is going to be an opportunity. Is it three years? Um, is it ten years? Who cares? Like let's go.

Speaker A: I would, I would bet on the viewability brothers. So in other news just to give continue in my you know, news field notes. So two other important points that are topical. One is an executive from a uh, very senior executive from a public company made the point to me yesterday. He's like I don't know what to think about this S1 from anthropic but I do know that I listened to a podcast and it was two agentic hosts and it was I guess spun out of NotebookLM that will create podcasts for you if you dump a bunch of information in it'll create A podcast is now called, I think Gemini or Notebook. Gemini. Notebook. Anyway, he's like, I have worked at this public company for a very long time. I know everything about it. And someone dumped a bunch of stuff into NotebookLM and, and the podcast that came out was better than almost anything in terms of its accuracy and ability to grasp nuance than anything that I had heard reported on a quarterly print from our company. You know, he thought it was staggering the level of accuracy and nuance that had come out of this automated media creator. And I thought that was an interesting observation.

Speaker B: Yeah, I think a lot of people slept on AI, ah, when it was hallucinating a lot and when it was really just producing extreme slop. And I've been very critical on both fronts. And I have to say, if you have that, if you have that, hang up still, I think you need to revisit it. You need to get a max plan. And it really has improved. Do not try to get it to do long form writing and whatnot. It's not going to become. It's

Speaker A: getting much better at that though, brother. Uh, and, and here's the thing. You'll recall that many of the financial news outlets were early to automation in terms of writing summaries of quarterly reports. That was like pre AI, that they would like, process the quarterly announcement and turning it in, turn it into a news release.

Speaker B: I didn't get my first job in journalism because I screwed up a number with, uh, that like a Dow Jones newswires. I wasn't made for the pressure.

Speaker A: Well, now we have fake podcasters interacting and having a thoughtful conversation around a quarterly announcement. So that's big change in media. And then the next one, and this is a little outside the media world, but hits home for me. And I wonder if it doesn't, if there's not a bigger story here. But I have a friend, really good friend, who's a senior partner at Weil, and he does fund creation and he does that stuff.

Speaker B: Weil, Isn't that like a law firm?

Speaker A: Weil Gottschel is a top 10 law firm, uh, prestigious 100-year-old legendary firm that navigated, you know, bankruptcies at, uh, Bear Stearns and Enron. And now they are the victim.

Speaker B: What do they bill at? What are they, like two, two thousand an hour?

Speaker A: There's a lot of 2,000ers in there. But here's the point. They had a couple of senior people, rainmaker types leave. They didn't kind of bounce back from that, manage it. Well, more left seven people on the management Committee found the door. It was like a run on the bank. A potential incoming new CEO turned down the job and the, the whole thing lost momentum and it's over. They tried to do a last minute merger or, you know, kind of cut the firm back, make it smaller. That didn't work. They have a third thousand, over a thousand partners, 2 billion in revenue and a thousand support staff, and it's gone, it's over, it's done.

Speaker B: Is that, is that unique to this category or this firm or is there some bigger. Well, it's,

Speaker A: it's partly the category because it's hard to lock an attorney in like you. You can't prevent them from walking out the door legally. And also

Speaker B: clients like, I'm sorry if your, your like, relationship with your lawyer is like, you know, your

Speaker A: therapist. It's intimate. Yeah. Right. Yeah. Yeah. So I think it's a category that's susceptible and I think this is going to be something we talk about in the episode. So this is a way of setting it up and you can get to the intro. But we are moving into extremely high leverage times. And if there is a story inside of the Weil situation, it's that the high leverage people left. They didn't, you know, stop the bleeding and the whole thing imploded. There are increasingly a handful of people in companies that are keepers of tacit knowledge and the stuff that makes the company go round, and people that are also the ones that know how to use modern tools. AI in particular, there's going to be a, I think, more sort of liquidity in the, in the business market that isn't unlike a law firm. So I think that one of the big stories and something we should be talking about this week is what's the next? How should we be thinking about AI in organizations next? Like I get it, slop cannons, people summarizing shit, automating your communication. All the basic stuff that like Jimmy, the guy down the hall that's good at AI does. But what happens when, when AI really permeates a company such that it is the owner of the workflow, it's the manager of the people, it's the keeper of organizational folklore and knowledge. I think that's a very interesting thing to talk about

Speaker B: on that.

Speaker A: Where

Speaker B: are we? Welcome to People vs Algorithms show about detecting patterns in media technology culture. Alex is missing, joined by Troy Young. Um, we're going to be joined by CJ Gustafson to break down the anthropic S1 because

Speaker A: I think CJ's becoming a character, a Mascot. He wears hats now that say, like, EBITDA on them and stuff like that.

Speaker B: Uh, yeah,

Speaker A: yeah,

Speaker B: he's got a merch line going. Yeah. No, I mean, I think, look, when you get out in the market, it pushes you in a lot of different directions. We've seen this with Taylor Lorenz. I mean, Taylor Lorenz was like, oh, I was, um. I'm mad. Masking out. I'm masking up outside and doing all this stuff. And then she's like, there's no money in this. And so all of a sudden, she's like, super pro AI. You know,

Speaker A: she's almost right wing now, which is bizarre. I actually like her a lot now. So I've come around. Yeah, but I mean, I don't want to see, you know, C.J. turning into Grimace from McDonald's, you know.

Speaker B: Well, we can talk to him about how he avoids that. But this is, you know, this is. Can be audience capture, but it can be like you can be captured by your audience. You're like, who are these people? They have no money. What am I doing here? I'm going to go hang out with those other people. It can be hard. You

Speaker A: mean like, like rich, you know, technologists or woke journalists? You pick, you know.

Speaker B: Well, I mean, look, I think when you look, you always go to a lot of journalist type, content type, whatever, creative. They don't, you know, you got to be realistic about these things. And you have to fill a need. You have to find a white space. You have to find your leverage. And I think the balance is always. How do you do that in a way that is authentic or is at least organic?

Speaker A: It, uh, doesn't

Speaker B: come across as inorganic.

Speaker A: You're bred to be skeptical, right? So you don't like anything. You're good to drink with. You just, you know, you're, you're, you're, you're sort of a. A classy narc.

Speaker B: Yeah. No, journalists order the third drink. That's the good thing about them.

Speaker A: Yeah, I love those guys.

Speaker B: I thought you made a great point about the law firm, because I think the question ends up being whether that leverage, and it's going to be a mix, comes from relational capital or whether that leverage really resides in the context of the organization. And I think if it resides in the context of the organization, then the humans are going to be working for the AI. The AI is absolutely where the leverage is, and then you're going to have the relational capital. But I don't see how having specialized knowledge within an organization, particularly a complex organization, is Much leverage because the AI can figure that shit out like, very, very quickly. So before we get to that, give us an update on the PVA Media Lab.

Speaker A: We talked about it a few weeks ago. We're all in this together. How do we learn together and share what we're learning? And that was really the genesis of field notes, if you recall. Field notes came about in a bunch of back and forth conversations between us. I had tried to build the Eye of Sauron media machine for myself. And I thought, what else can I do to, you know, make this, you know, new shift, this shift interesting and personal and a learning moment. So I wanted to build something you said I want to do. Takes all the time. And we start. And I thought our, our WhatsApp group chat was a, uh, cool little media product. And so we turned it into something and that. And that was field notes. And what's happened is it's kind of like totally non. Not promoted in any way other than the conversations we have here and the people probably getting sick of us talking about it. But there, there is a small group of mighty, mighty people in there, and they're all really smart. And I think it's because we, we didn't really promote it or made an effort in any way.

Speaker B: Well, it's organic. So

Speaker A: you get these great, great people in here.

Speaker B: Brad Robertson, Vincent Murphy, Paul Isaacson, Jason Scoggins.

Speaker A: Yeah, you know,

Speaker B: Antarosa.

Speaker A: You knew it. Hasn't been around. Jwl hasn't been around in a while. Maybe he's.

Speaker B: That's true. He got sick of this.

Speaker A: He got sick. So to get to the Media Lab point, Brian, I had aspired to make this, uh, an experimental environment that we could all learn from together. And Brian posted the architecture of the rebooting. And yes, you, Brian, Brian, you. You posted a sort of summary of how you're using your agent Terrence. I'm only ever going to call him Terrence. No, no, no,

Speaker B: it's tj.

Speaker A: You sort of showed how the whole thing worked and what. Here's the thing. The reason it resonated with me, Brian, is because it came up in a couple calls I had this week where people were talking about how do we build AI into our organization in more fundamental way and what's the consequence of that? And what you had laid out in your document was basically a workflow. And you said, we get an opportunity with a client, it turns into, you know, an opportunity document. It triggers all of these actions that people have to do things. And Terence orchestrates it all. It occurred to me that the, the most important and fundamental thing that an AI can support in an organization is understanding and coordinating workflows and capturing as much context around that. That's the word you'll hear a lot today, context to make it useful. And context are things like conversations with clients, histories of your transactions, all of your communication preferences, any of your deliverables, all of that context. And I thought, this is cool. Let's turn this into our first lab entry. We made it a lab. I built a little feature to separate those types of posts in the feedback and I followed up with how a uh, post on how PVA notes works, including all the APIs and how we use LLMs, how much money it costs to do it and all that kind of stuff. So that's the lab and I, you know, nobody's posted besides me and you. Hopefully someone will. There are examples in the, you know, several months that we've had field notes running of people sharing like how they hacked their TV or you know, an app they built to help them, you know, work more effectively, whatever, but I think it's kind of cool and if anybody has an interesting case study as to how they, you know, tailor Lorenz themselves or build something cool, we'd love to know about it.

Speaker B: Well put. I think. Can I talk a little bit about tj, formerly known as Terence? Because I've just decided to immerse myself in trying to use. I want to put Claude at the center of my business now. Maybe I end up creating my own harness and all this stuff. Honestly, I'm not that advanced.

Speaker A: What is a harness, Brian? Is that, is that uh,

Speaker B: I'm unclear to be honest with you. It's just the front end. It's the front end to an LLM. Am I wrong here?

Speaker A: That's how I understand it. So let's go. But

Speaker B: the bottom line is I'm not going to do it. I'm just, Claude is going to basically I've just decided I'm committing to Claude and I got access to Claude Duet, which is a great app. I'm very bullish on, um, Claude Duet and I want it to basically handle all of the coordination work, as much of the coordination work as possible. In my business I have a lot of non standard things that I end up doing that are at best customized and often, let's be real, are custom. And it carries a tremendous coordination cost. It involves a lot of humans moving data back and forth. It involves a lot of follow ups. It involves a lot of me as a disorganized Person saying, where is this? What is where? How many people do we have coming to this dinner? Or whatever? And what I realized is not a knock on the humans, but humans basically have a 10 to 15% error rate when they're, like, moving around data and stuff. This is the atoll Gawande. You know, handoff of care. That's when errors happen. TJ works 24 hours a day.

Speaker A: You sound like Lucky Palmer. We have an error rate.

Speaker B: Yeah. Humans have too high of an error rate. They're.

Speaker A: Humans are

Speaker B: lossy.

Speaker A: Is it humans or humans?

Speaker B: That's. That's a

Speaker A: Philly thing. That's a Philly thing.

Speaker B: Philly accent thing. Yeah. Humans, they're lossy. And that's. That, that's a reality. And, and organizations, I think, have this lossiness across them. And if you commit, I feel like, to these, these agents right now, like, you're going to end up doing better. You have to, like, work through all of the messiness. I, I have. I had TJ and then someone else firing off like, multiple, like, calendar invitations to people. Like, it happens. It's okay. We're gonna go through a weird

Speaker A: period. What's the J in tj, by the way? Terrence James something. What's the J?

Speaker B: I don't know. As I said, It's TJ McConnell. Whatever, whatever. TJ McConnell. I don't know what TJ stands for.

Speaker A: No doubt. Uh, he sounds like a football player, but no, no doubt inspired

Speaker B: by

Speaker A: a basketball player. Inspired by cj the intrepid cfo.

Speaker B: Now, it's coincidence, to be honest with you. He's timothy John is T.J. mcConnell. He's a junior, too. But you know what I like about it is. And what I think that the thing that's opened my, my eyes a little bit to this is that when we talk about humans in the loop, I always thought, like, okay, the human is going to be the orchestrator, and the tool is going to be there. And it's. And it's usually like Odie in Garfield. It's got its tongue hanging open, it's super eager. And it's just. And it gets annoying. Just like Garfield would get annoyed by Odie. I think it is going to be when they're moving to this proactivity thing, basically, humans are not good at using these tools. And so the tools are not going to give up. And they're going to say, we're going to use you. And I think that is the transition we're in, particularly within organizations that, to get the full productivity benefits of this you have to basically do as the coders did. There's a reason that the coders are 100x more productive. Now. We don't see that productivity is because the tools run the coders. The tools are doing all of the work. And that to me has been the unlock is you've got to push these things as far as possible and deal with, I guess, what Ethan Malik calls the jagged frontier. You'll find it. But it's pretty remarkable how much better I think Claude in particular is to how I was using it a year ago. Just like shockingly better.

Speaker A: Right, so why don't you break down maybe in a Peter Thiel esque, you know, way with references to religious philosophers, how, how the humans stay relevant when they're being orchestrated by T.J. maxx?

Speaker B: Well, I mean, well, look, there's still a role for, for humans, like where I see this. So I believe that this agentic client delivery system will eventually replace tasks that were done by humans. It already is, and I think it will replace over time, probably in my business, about $100,000 worth of labor costs that go into what I consider broadly like coordination. And I think it will do it at only. I would only do it if it's, um, improved effectiveness.

Speaker A: Will it hit the top line at all, Brian?

Speaker B: Yes. The way it hits the top line is because I don't want to save that money and just have it like put. I want to grow the business. And the thing that has inhibited growing the business, other than the fact that the media industry is completely compressed, is too much organizational energy going into what I consider the coordination function. Things that are not growing the business. Now, if you can redeploy that those assets, that energy, that human capital into building closer relationships with the clients, not finding where documents are or giving them updates and whatnot, then you're in a much better position. And I want to collapse sales and client delivery so that sales is not like just talking to the client when they want more money. I want them to be in the trenches with it. It's just. So

Speaker A: have you considered putting it towards like a used Lambo?

Speaker B: A used Lambo? Yeah.

Speaker A: You're not going to get a new one.

Speaker B: I could do fractional. I mean most, I think most like Lambos, at least in Miami. I'm actually 100 sure are, are rented usually by the day or by the week. So maybe I'll just have like a Lambo day, like once a quarter where I just go

Speaker A: around in

Speaker B: a Lambo. But until then, I need to expand the Top line.

Speaker A: You remember when, uh, you and Anna and I and Jillian M. Met at the Four Seasons? If you had pulled up in a Lambo, God, that would have been great. When now Anna gets out.

Speaker B: No, I drove in the Altima. And my trick is I, I, I park like in the neighborhood. I don't use valet. The valet is like 30 bucks.

Speaker A: I know where

Speaker B: to park. I know where to park. I don't mind crossing the street, but it's been a big unlock. I'm very into it. Have you, have you used duet yet?

Speaker A: I, um, have not.

Speaker B: So that is Claude Anthropic's version of Muse. OpenAI ChatGPT has dots. There's a few others. It's these personal assistants. I think what's great about duet, I've only had it for a day and a half, is one. It is very proactive. It is always suggesting it's trying to get downhill and get things done for you. And it's not waiting for you to come to the tool. It's saying, let's run this. If anything, I want it to be, I want it to be a little bit more off the chain. You know, it's always asking for permissions.

Speaker A: Brian, I don't know if you, you've ever been here, but like what I wonder about, uh, it seems like you're doing all the right things. So congratulations. I wonder and I don't know how this transformation happened because you went from like, you know, kind of jaded journalist to editor to salesman to like it guy. And it's been really great to watch.

Speaker B: Yeah.

Speaker A: And, and ah, and like. Or it's been even

Speaker B: better to live

Speaker A: the organizational strategist. But just

Speaker B: all my strong suits sales, it, it's great.

Speaker A: Yeah. And so what, what it makes me wonder is, is what happens. Like I had like 2500 people in the org and inevitably, like, let's say 50% of them, you know, don't care about any of this. Maybe more, maybe 80%. And let's say that the path to, to the promised land is completely re engineering workflows and roles and talent around these AI driven workflows that you're describing. Right. Where it's, it's the big boss. TJ's the boss. And that in so doing it, you know, it's emancipating for the, you know, more human side of your business. Whether that's assessing the landscape, using your insight and taste, managing relationships, making connections, all the nice human things that Peter Thiel talked about this week. The question I have is how hard is that going to be at a multi divisional, you know, multi thousand person company? And how many people have to die so the others can live? And I think that's going to be the sort of corporate story of our time. I don't think that

Speaker B: 20%,

Speaker A: maybe more. More. And I, I don't think that any kind of, you know, nice HR driven change management programs are, are going to solve it for most companies. And I, I think it's going to be a bloodbath is really what it's.

Speaker B: What do you think is behind? So Microsoft this week came out with a voluntary buyout, which I always thought voluntary buyouts were like something. Do

Speaker A: you see how I set you up there? Perfect.

Speaker B: Yeah. Next, the next step in our evolution is you not like actually mentioning it. But anyway, we'll get there. It's all an evolution, it's all a learning process. What I wonder is, is that cut? So the voluntary buyouts, first of all, I always thought that was something that like newspapers did with like unionized work workforces. And I always thought it was the dumbest thing because I said, I always thought, wait, the people who would take a buyout are the people who have the most market like leverage. Like someone who like has limited market opportunities is not going to take a buyout. But whatever they're doing, the voluntary buyouts, they went out to what, 27% of their workforce. I wasn't able to find like the rule of thumb about like what percentage you want to take it. I mean I'm sure they have it there. I haven't seen it reported about what, let's assume, I don't know, maybe about 50%. So like, you know, they take out 15% of the org. What can we take out of that? Is that like a sign of anything more than these tech companies got over their skis? And we don't even, we can't even fathom how, how bloated they are.

Speaker A: I don't think it has anything to do with that. I, I think that the tech companies getting over their skis to me is just part of a kind of expansion cycle where revenues were growing and they were investing in new parts of the organization. Maybe some of it was irresponsible, but these companies were awashed with cash and rather than distributed to shareholders or by companies which in many cases under the Biden administration they couldn't do, they expanded payroll. But I think it's much broader than that. I think that the waves of layoffs that you've seen in media are coming for every sector. I just think that there is going to be great opportunity for service companies that are in the business of helping organizations transition to this new world. The McKinseys and of the world included, and

Speaker B: Bain,

Speaker A: there's. Yeah, there's a lot of business there. And you know, I, I don't think. I used to sort of think like, I don't know why

Speaker B: people aren't excited about AI. I mean, uh, McKinsey set up. Well, Bain.

Speaker A: Yeah, I didn't. So to answer your question, I think we're like 5% into this change process. And though there's going to be. And, and I, you know, there was a, There was a post that I think we shared on field notes. It was a bit sobering about people not unlike me. It was. And the analogy was to turkeys, and turkeys, you know, have a good life. And they keep getting fed more and more until a couple days before Thanksgiving when the graph turns the wrong direction. And I think that that's coming for a lot of people that are in sort of manager roles in companies where those. That coordination function doesn't need to exist the same way. And the tragic part of it is that in their 40s and 50s, they're probably not financially ready to call it a day or. Yeah,

Speaker B: there's, there's massive human costs to this and there's, you know, there was a viral video going around on X of this woman. I think she was, she was probably millennial or something, but she was like weeping over what AI was doing to all the chiefs. She's like, it is putting people out of jobs, creative work. It's being stolen from us. And like, it was held up as, as is its want on access as an object of ridicule and to ridicule her for at least 24 to 48 hours and then move on to someone else. But it did kind of strike a chord of. You know, look, we've all seen like Reddit posts from computer scientists or, you know, whether it's the mathematicians or whether it's, you know, my designer friend. A lot of

Speaker A: only fans is the new mathematician, Brian.

Speaker B: Even the only fans. People mostly women, I think. Right. Are maybe not. You tell me, Troy. No, look, there is a human dimension to this that I don't, I, I don't think is good to lose it. You know, we did, ah. I did a dinner in Chicago last week for a group of publishing product leaders and I do a lot of these, these things. And you know, of course it was about AI Right. And you know, in previous ones just six months ago it was a lot of like tiptoeing into this stuff. There was a real tenor change. Now it could have been the people around that table. It's a small set, it's 10 people, right. But they, they are reporting being incredibly more productive. They're getting right to prototypes. They're talking about being very quick into market with things and being able to do way more and produce way more. If anything they're being held back. Not by the technology, they're being held back. They feel at least what was reported by organization, organizational design for the most part they've sort of written off a lot of the hold offs, holdouts. It's like look, if you're not going to adapt this tough luck like you're, we're going to give your clawed credits to someone else. You get co pilot. There's also like this, the haves and have nots. The haves get more tokens and the have nots get co pilot. One guy, this is a paraphrase but this is what he said. He said I'm sympathetic to the holdouts but this is going to sound like I'm an asshole. No one has a right to be a designer. No one has a right to any job they want. You have to earn it, keep practicing it, keep learning and growing. And if you want to earn it, you have to use the things that give you leverage. And this thing leverages intellectual power. If I'm a farmer and you're a farmer, work in the fields and you want to work the fields with your hands and I've got a tractor, I'm going to out farm you every day. And that's the point. And I think you know, look, some people are not going to adapt either because their skill sets are you know, mostly in uh, a layer that is no longer, you know, needed. But some are going to change up the game. And I think about that like with managerial class, I think we go back and forth. On one hand the managerial class is doomed and dead. On the other hand we're all going to be managing. The people who have leverage are going to be able to manage tons of different agents. So which is it?

Speaker A: Well there was a case study in that this week and it's wondering what it made me wonder when I'm going to see the rebooting revenue skyrocket. But this, you know,

Speaker B: 2027, we're, we're, we're forecasting 35% growth.

Speaker A: There was some SAM almanaphorism about one person companies and you know, there that I think it was, Anthropic just acquired a one person company called Mithril Medicine. And this enterprising gentleman was using Claude to take a, uh, muscular dystrophy antibody treatment from concept to proof of principle, uh, in mice in 18 months. And it was only him. It sounds

Speaker B: way more impressive, honestly, than what I'm doing with tj. I gotta be

Speaker A: very. Yeah, it was him, Claude, a couple consultants and a bunch of outsourced lab work. And I don't know what, what Anthropic bought Cody's company for. He will continue to do research and such inside of Anthropic and now that they have a wet lab. But this is an example of a high leveraged individual cashing out.

Speaker B: Okay, but this seems like a very niche example. I mean, look, biotech has always, I believe, had small teams that are super smart and are able to leverage technology and come up with compounds that they can't possibly actually get through trials and distribute, et cetera, and then they sell out usually to larger biotech firms that have that all set up.

Speaker A: Right, so you're saying that we need to, this is the lawyer case and that in the case of a small B2B media company, there's no way to ladder up quickly.

Speaker B: I'm saying like there is a long history of very small teams in biotech selling for tons of money to larger entities that like, it's almost part, I believe, of their innovation cycle is instead of innovating internally, we're going to have an ecosystem that's going to do these innovations, whether it's through universities or whatever. And then they're not going to have any possibility of going through trials, bringing this to market. Moderna was, has been, I believe, an outlier in this, in that they've done like a more tech oriented full stack.

Speaker A: Right. But Brian, we're trying to put you in the therapist seat here, if you don't mind. And I'm saying what would have to happen at a, at a small media company to get any leverage such that your, your, your terminal value was high or more than just you being around or that you were, you know, pumping up revenue and Ebitda and you know, hypothetically, you would have to have, you would have to be creating more coveted data sets, selling those either to people that need data like at frontier companies or whatever, or, or to the people that, that want your information, or you'd have to be a more robust marketplace for buyers and sellers to, to get new business. Right. Like something, you know, there's no, I'm going to scale up and get more ads from like Jonah Goodhart's new company. That's not a thing. Right?

Speaker B: It's

Speaker A: going to be doing more.

Speaker B: You do,

Speaker A: right.

Speaker B: When I get off, Claude is going to track those guys down.

Speaker A: You doing more, you know, 20 person connection dinners in, in Minneapolis isn't going to help make your

Speaker B: Chicago

Speaker A: okay. So what gives you some leverage? What gives any media group or don't you need it? Maybe you're just happy with your couple million dollars and your weekend rental of the Lambo.

Speaker B: It's fractional. I can have it on uh, a Wednesday. It doesn't really matter. I mean I think everyone is doing all of the above and seeing what sticks. Right. Like I mean to me the most valuable part of everything that you mentioned there was can you produce some kind of non public data asset that is not visible to the world? Everything on the Internet is visible to the world. I appreciate publishers crying over their recipes being sucked up by, by ChatGPT and by Aunt Claude and regurgitated out to, to the rest of the world, but that's over. Like that's gone. But there are still, you know, non public data assets that are very valuable that either because they're very current or because they just never existed in the open, open web or the open Internet and allowed people to get at it. So I think a lot of publishers are going to be locking down any and all data assets and figuring out how do we produce unique data assets and then I guess be agnostic about how that is monetized and distributed. You know, like uh, did you see Jim Vande Hei who I think is a, is, you know, he's smart. I feel, you know when, when talking about these things. I'm a sucker for journalist CEOs of media companies. But he came out with Axios 2030.

Speaker A: I think he's the media's Lebron to be honest. But yeah, uh,

Speaker B: you do,

Speaker A: yeah, he's good man, he's smart. And the other thing is he doesn't, he doesn't mince words and he gets in front of change and that's the hardest part. Try to go to like look, look at Conde Nast's uh, results that what's his face snuck out of there somehow and great, you

Speaker B: know, great to bail on that. He's missing Q4 ad revenue by 21% and that was already like dialed down

Speaker A: company. Like I'm just telling you, I used to run one. 21% is really bad news unless you have incredibly Rich, generous owners. Because that money, like there's no cost savings that's going to insulate you. It's terrible.

Speaker B: Yeah. And I think, I don't know, Mark had lot of. Mark Stenberg did a nice job reporting off that story because I know that like I texted him Rick before. I didn't get a response about like I was like, oh my God, what

Speaker A: kind of probably hungover.

Speaker B: What kind. What did Conde's PR team put you through? They probably threatened all kinds of different things to Mark. But yeah, their events business, even like Vogue World, you know, came in like really short and you know, that is a big question. When these media companies diversify into these services categories and they all go to events. Look, there are a ton of freaking events out there. You're gonna have to like there is, there is a limit to the number of events. Like you become, you know, like Crusty the Clown when he was quitting smoking and he put those patches everywhere. Like, you know, you run out.

Speaker A: I've done that before. I've done that, you know, three. Right on time. Wow,

Speaker B: Crusty was covered.

Speaker A: Yeah. But I was also eating nicotine mints at the same time.

Speaker B: This isn't a publisher's dinner, but Jim came out with a really good. And I love to ask this of, uh, what

Speaker A: did Jim say, Brian? What did Jim say, Brian?

Speaker B: Well, he said a lot of things, but what I like to get at is what is your hypothesis? Right. Because you have to start with a hypothesis. It's okay. I get a lot of people in these publishing companies that throw up their hands when I ask, what's your business going to look like in five years? And they're like, oh, you can't possibly know that. I'm like, really? I mean, you might have an idea. Surely you have some idea. Like, I'm not asking for exact. Anyway, his hypothesis comes down to this. We believe the value of verified reported distinctive truth and curated human connection will spike. Okay, that makes a lot of sense. And he also thinks that AI is going to be the interface, basically. I mean, I think it, uh. Basically

Speaker A: he's talking about a lot like kind of a head. A headless media company. Right? Yeah. Which I. He wants to give your. Your hungry TJ data. He wants to give the agents his information. Right to the vein.

Speaker B: Yes. So basically their thesis is that the open web is, is kind of a goner. Right. Search is dying, bots are everywhere. And so they created a new business line, Axios Direct, which basically is selling feeds of their data to client run LLMs and the idea is like, everyone is going to be running their own information LLM, some kind of knowledge center on the other side. And we're just going to feed that data to it. Now, they used to feed that data. I mean, they have, they have a couple of different businesses, but like through email newsletters primarily. And I think that in using Duet, I'm convinced that email itself is going to be like our desktop. Like, I do not curate my desktop. I never was that person with all the complex folder systems and subfolders. And once we got desktop search, uh, I was like, I don't care. I'll find it somewhere. I'll drag it into every.

Speaker A: Mars and Venus, baby. Mars and Venus. Alex is like that. He does the folder organization. Does

Speaker B: he do the folders? No, I

Speaker A: got. No. He's also an Inbox Zero guy.

Speaker B: Oh, yeah. Inbox Zero is dead too. There's no need to do it. The email inbox is just going to be some dumb data repository. And you're going to use an AI interface on it. Now, whether that's supplied by Google or whether you want to use Claude or whether you want to use Muse, I have no doubt that you're going to be using an AI interface for that message. Well, that's a new mediation. And establishing direct connections was always the idea of email. Well, guess what? That direct connection ain't that direct anymore, you know, So I think that's a big shift. And they're also doing a lot more with like. It's funny that everyone, everyone's so into B2B. They're doing, they're. They're doing, you, uh, know, new verticals for specific job titles. This is,

Speaker A: I told you, I told you, the oldest B2B. I don't mean to be this guy, Brian. Another CMO

Speaker B: publication, by the way. Holy shit,

Speaker A: Brian. I gave you the vertical to go into. I begged you to go into the vertical. You disregarded it and all your. In your scramble to do your media product dinner in Chicago. Data. Data,

Speaker B: uh, data's

Speaker A: the

Speaker B: vertical.

Speaker A: Data on data, media on data.

Speaker B: Well, you know, I was thinking about that on walking back here. By the way, London's, like, wonderful with these little, like, urban parks and stuff. It's so, like, calm here compared to New York

Speaker A: if it wasn't for the hysterical immigrants and fires and protests. But other than that,

Speaker B: you know, not in Bloomsbury. I'm going to Paris on Friday. I'm doing. I'm going to the.

Speaker A: I'm Joe. I want to make it clear that I'm I was only joking about.

Speaker B: I'm going to the front lines. I'm going, I'm going to Paris on Friday and I suspect that I will escape unscathed. I have a feeling that civilization is not collapsing in Paris, but I will report hopefully next week if I'm still here. Let's talk about the holdouts a little bit more because there's this idea that has, that has popped up in field notes about cognitive mercantilism. And I think this is kind of the latest offshoot of the D cell degrowth opposition to tech. I mean I just think we love binaries. I mean it's always going to be

Speaker A: kind. I don't understand what it is. Can you explain it to someone dumb?

Speaker B: I was hoping you would explain cognitive mercantilism.

Speaker A: I can't. I can't. I guess it's when it hearkens back to mercantilism, which was, I don't know when was that 1500-1700 where you hoarded gold and hope that you, your success would be at the expense of others, other success. And uh, well, I think

Speaker B: to me the big thing is like what do you like, you've got these competing and I think this goes to where you find leverage within your career and whatnot. I think this is a big part of like what we talk about every week honestly is the question is how you maintain your sort of cognitive powers, right, which are, which provide unique leverage to you while you use all of the tools for these other sort of to unload like the cognitive tasks that are getting in the way of the very valuable cognitive tasks or. But the thing is, humans are not very good at delineating between the two. I

Speaker A: think the analogy is that it is treating intelligence and expertise as treasure to control and protect rather than something uh, who's expanding availability could create more wealth and opportunity. It's like tariff thinking, export restrictions, industrial subsidies. That's the connection to mercantilism of old. And it's that you have to somehow, uh, it's really about protecting your data, isn't it? What is unique and valuable to you.

Speaker B: But just like mercantilism generally it uh, loses to free trade and having a, uh, zero sum approaches generally do not do as well as positive sum approaches. And I think AI is a big test of that because on the one hand you have a lot of the venture capital set and you know, the AI boosters saying, hey look, we're going to have 15% growth. When we have 15% growth, this pie is going to be so Bit why are you guys arguing about your slice? You know, meanwhile, they've got, like a massive slice that's getting bigger, so

Speaker A: you got to protect what you got. Brian, there's a lot of superficial responses to what role do humans play when, you know, knowledge is abundant and machines are smarter than you and all of that. It just strikes me that, you know, to remember the hierarchy. We are the things that the machines should sustain that, you know, I find it hard to believe that we are going to be, you know, the minions in this equation. I think that. That there are fundamental human skills around networks and taste and emotion and point of view and connectivity that will become far more important and critically, like asking the right questions, being discerning, thinking about systems, all of this. You know, it's just the thing is, I have very little faith in most people's ability to navigate something so fundamental. And I think that just to tie it back to the previous discussion, I think that the managerial construct of old, which was, I guess born of, like, you know, how to create, uh, a structure to accommodate the immense companies that came out of the Industrial revolution, was. There were a lot of people in coordination roles, and I'm one of them, dude. I don't know how to do anything other than just try to kind of like, get people to focus on something, find good talent and, you know, manage a process. And increasingly, I just think that's lower value. We certainly don't need as many of those people. You got to learn new skills and, and, uh. And I think when it happens later in your career, it's totally devastating. I don't know what to say other than. Than, you know, store your nuts.

Speaker B: Store your nuts. I think that's a good point. And also, a lot more industries are going to undergo. I've been too flippant in previous episodes, but a lot more industries are going to undergo the deconstruction that happened to the field that I know best because I entered the journalism field in 2000. I mean, right as it fell off a cliff. So when I entered journalism, um, to now, they were training us to go work in newspapers, which was honestly crazy at the time. And newspaper employment now last year was 82% lower than when I graduated from journalism school. 82%. There's going to be other professions that go through very similar compression. And look, uh, I went to school with a lot of, A lot of smart people, a lot of good people. Most of them are not in this profession anymore. And they've gone on and they're, they're, they're doing other things. They've like, you know, some went to law school even, you know, some are in real estate. There's pr. I got a friend who's a diplomat. One got into real estate. That was

Speaker A: smart. Did you ever. Yeah, that's a common one. My mother, my mother did that. Did you ever get close to being sort of destitute and alcoholic or anything bad? I mean

Speaker B: I, I, I lost my first job. I got, I got, you know, Jason Calacanis like laid me off, Fired me.

Speaker A: Yeah, I

Speaker B: don't know. Is there a difference between getting fired and being laid off? I feel like we sort of like just say laid off or something. Like there's some moral thing of getting fired. That's basically, I think it's just one category.

Speaker A: You're gone.

Speaker B: You know, it's, it's an unpleasant feeling. It's not pleasant. But that's going to come and that's going to be a uh, uh, a uh, very difficult. And so we're going to see a lot of these rear gu. You know, Kathy Hockiel just has a, has a bill that bans AI chatbots from its unsafe features. And you know, look, a lot of things are going to be done to slow this down in the name of child safety. I've seen it a million times. Doesn't mean that there should not be

Speaker A: by the way, public service message. If your AI is telling you to do things that maybe are life threatening, you should maybe either try another model or ask your physician.

Speaker B: Are, ah, you talking.

Speaker A: This

Speaker B: woman is suing. The

Speaker A: woman died. Her family is suing.

Speaker B: Okay. Her family is suing because Chat Shibi told her to stop taking her medicine,

Speaker A: her epilepsy medicine. Yeah.

Speaker B: Which, yeah, that's a big mess. Let's be real. It's not good. But you know, you update the software and we'll just keep plowing forward. One guy in Kentucky got a religious exemption from using AI. Uh, that's again, these are not going to be long term strategies. I don't think

Speaker A: I, if I could take the podcast in just one more direction. I was dying to ask you about something.

Speaker B: Yeah. And

Speaker A: I had asked you about the inherent conflicts of running a free media, journalistic media business while catering to the, you know, the executive clients that come to your events. And I noticed that some, a couple of people took a swing at Semaphore on X this week about them. Apparently Justin had proudly said that they were training the folks in UAE or something to create content at a ridiculous margin and that the king that that they were appearing. I guess it was it uae. Brian, I. Do I have this right? Were appearing more frequently in positive framing inside of Semaphore's reporting. And this person had suggested that, that this kind of fundamental business model creates conflicts. And I had asked you about it. I thought you, you'd probably be, you know, circumspect about it because of your relationship with Ben Smith. But I thought, I thought, I thought you would, would have a point. What

Speaker B: does that mean? So what you're saying I pull, I, I pull punches because I don't want to offend Ben out of some sort of,

Speaker A: uh, journal. Well put. Yeah.

Speaker B: Journalistic. The fixes in or something. Look, you

Speaker A: were

Speaker B: the one, you educated me when I was being naive that media is just about managing conflicts. Choose a model and then you'll find the conflict. Oh, okay, you have a billionaire benefactor. Well, guess what? Billionaires don't do things just because they feel like doing them. They also have interests in all kinds of different things. We have seen time and again the billionaire owners suck. Oh, you got an ad model. Oh, well, you know, you're not aligned with your, with your audience. You're going to be carrying water for your advertisers. You're going to be bought and paid for. Do you see this new media chart? You know, they went from bought and paid for to independent. You know, they put Ben Thompson before his Stripe deal as independent and then after he did a deal with Stripe to be like the presenting sponsor, all of a sudden he was bought and paid for. Okay, so you're going to have a subscription model. You're going to get paid directly from the audience. That's clean, that's neat. All of a sudden you got audience capture. You're Taylor Lorenz. You're like, what are you talking about? I'm masked up. Get my, uh, what do you people want? I've got my mask on outside. And then she's not getting, she's not, she's not making enough money. So she's got to find a new. She's got to find a new audience segment. I don't know, it's just managing, it's managing conflicts. They've done a tremendous job with building a events business that has a journalistic orientation. Right. I don't think this business works as a direct on a direct to consumer journalism model. If it was just Ben and team are going to create amazing journalists that are amazing journalism, um, that the most powerful people in the world need to access and they're going to pay us Directly, I don't know, anywhere from a thousand. What is Bloomberg? Like $500 a year? Up to 50,000. That sounds like a good model, doesn't it? Business doesn't work with that. Do you know how long it takes to build that? Do you know how long like Bloomberg, I mean Bloomberg had a, had, first of all they have the terminal business, but they had like a paywall. Then they took it down and whatnot. But then they went back to it. It's really hard to, to build these businesses. Uh, you're the one who always tells me this. Are you playing the naive. This is like a Freaky Friday episode.

Speaker A: Yeah. All right, so that really good. That was. Can

Speaker B: I be Dudley Moore in this? You can be Kirk Cameron,

Speaker A: you can be Andy Rooney.

Speaker B: I don't know why, I don't. This is the problem of the uh, this is maybe this is cognitive mercury. Like I have, I know who, I know like on Demand, who was in the Freaky Friday movie of like 1983. I can't remember a lot of other things. What was your take on it? I, I'm like, am I, am I off base here? I, I, I don't know all the details honestly of the UAE thing. They've been through this before with their China operation. You know, they struck a deal in China that was controversial to do, you know, an events business stuff. Anything you're doing in China is, is, is going to be fraught and the government's going to be involved. They got, they got beat up a little bit for that. They didn't want to go to VCS because freaking VC is going to invest in, in a publishing business in the year of our Lord. Well now it's 2026, but it was like what 2023 back when they were doing. So they went to rich people and guess what? One of the rich people was Sam Bankman. Fried. Oops. Hey, get back the money. Find some other people. I

Speaker A: don't know. At the risk of being ponderous, I think, you know, Forbes Media was a couple hundred million dollar media business with very diversified. It's actually a good case study because you know, I think History of innovation, very diversified now has a separate arm, um, called Marketplace, which is the most successful affiliate marketing company in the world. And the media business was, you know, a couple hundred million dollars. And that came from lots of sources. It came from accolades, it came from programmatic, came from direct sales, it came from affiliate and it came from events. And I think that the pressure's on all of those areas right now. You know, while some like activations and stuff present growth opportunities. And you look at the Atlantic and I think they're very different case studies. And Atlantic is probably sub $100 million business with a uh, gigantic news and editorial operation that serves a subscriber customer. And they were able to do it I believe because they had the cushion of philanthropy underneath. They could navigate the losses on the path to that model. And now on the other side of it, seemingly have built it a reasonable mid sized media business. And so you look at, you know, it's just there's no sort of elixir for anybody in the media business. And you know, other than being a platform and, or you're like a baby, you know, you're like a tiny focus, super niche, high value, you know, organization that serves a customer that's willing to pay a bunch of money for a very kind of data driven exclusive product. And, and so that's kind of where we are. And I think that in the case of, of you know, Ben and Justin, they're, they've scaled a convening model around uh, you know, high quality journalism product. And it does all of them, as you point out, come with trade offs and potential conflicts. And so someone wants to, you know, there's always going to be someone coming after you when you build a media company. And so someone took a swing. I just thought I would get your take at it, take on it. And, and did I carry

Speaker B: water for them? I think I was fairly.

Speaker A: I think, I think you did a good job. Yeah. Okay. No, I, I'm convinced.

Speaker B: All right,

Speaker A: so you, what do you want to talk about now? The Cheesecake Factory?

Speaker B: Yeah, I want to talk about Cheesecake Factory. Obama took a shot at Trump. He's really good at needling Trump and he took a shot at Trump. He said that he decorated the White House like he was Saddam Hussein at a Cheesecake Factory. I'm not as familiar with, I mean I know the Cheesecake Factory mostly from its voluminous menu. I've used it as a, as a metaphor as I'm want to do because I always, I hate when sales kits become like a Cheesecake Factory menu. You know what I mean? Like any like sales kid I've ever seen, I'm like, this has never been sold in like five years. Why is this still in here? And why do you want to give a client? Like it just shows a lack of conviction if you put like 230 options in front of people. And I mean I find that all the time with you know, These menus, I'm like, can you really make all these things very unlikely? I don't know if this is a smart move or not. Like, and I don't know, Cheesecake Factory, some people, you know, the. The anti Obama, I guess, pro maga crowd was, you know, trying to. To make this a bit of a tan suit moment where of all the outrages that are going on right now, we should be most outraged at, you know, Barack Obama besmirching the reputation in the cheesecake factor.

Speaker A: Well, yeah. So what's your question, sir? What do you want to talk about? I

Speaker B: just. I guess my point is, one, is, is this an unfair shot at an American icon, or is this a fair.

Speaker A: Is the icon Trump or the Cheesecake Factory?

Speaker B: The icon is the Cheesecake Factory.

Speaker A: Well, I think what it represented is, you know, a very American ideal of luxury and suburban abundance and choice. And the thing that struck me, I did go in Marin County. I went to the Cheesecake Factory maybe twice. Uh, the menu, as you know, is a novel. Why is there

Speaker B: one in Marin County? Why would there be one in Marin?

Speaker A: I don't know. But they had ads in the menu for liposuction, which I thought was in really poor taste. Like, maybe, maybe it's just good targeting. You eat a lot of cheesecake and you, you know, get it sucked out. I don't know. The food is reasonably good. You can get a nice carbonara. You can get, like, you know, chicken piccata, or you can get spring rolls. And so it's like all ethnic. Ethnic categories. They serve it on giant plates, I think. So the food looks maybe smaller. You know, it is decorated lavishly to give you the sense that you've arrived. It is part of the promise of America. And so I think that Obama needs to be careful in alienating, you know, people, middle class people that want affordable luxuries. But it's fucking hilarious. I love it. You know, Trump has the worst. I love

Speaker B: that you went to. I love that you went to. Marin is, by the way, the eighth richest county in the United States. So the fact that they. You should have gone to the Greenwich cheesecake factor. Well,

Speaker A: the one I really like, if you want me to.

Speaker B: Anna had a British friend who was very upper class, and I talked to her one time and she said she's like, oh, I love American suburbs. I was like, really? I'm like, that. That surprises me. She's like, oh, yeah, my brother lives in Greenwich.

Speaker A: Yeah, well, you know, my favorite of all of them is PF Chang. I really like them. I love those lettuce wraps. I like their soups. I think that's a solid signature chicken. It's really good. I really like it a lot.

Speaker B: They're very big in the Midwest. And I think if you're going to be like a, uh, classic American eatery, you probably want to be in the Midwest. Maybe you can be in the south, but you can't be, like a Northeast or. Well,

Speaker A: being a friend of mine for a while, we're doing this thing where on the way back from Shelter island, we would always stop at, like, exit 56 on the LIE and we wanted to try all of the chain restaurants, like Cheesecake, P F. And what's that? Steak One Outback. We're even gonna go to, like, TGI Fridays, which I've never been to. I'd like to try that one.

Speaker B: You never been to a Friday's? You've been to a Chili's? I like Chili's. CJ's gonna show up very briefly. I just wanna do a couple minutes on, you know. Cause we like to touch on world global affairs. It was very exciting for me because all week Serbian politics have been. I feel like they've been in the news now. Maybe it's my feed. Are, uh, like, regular people getting a lot of content or engaging in content around the earthquake in Serbian politics by the success of this student movement called the Students List. Did you see this at all, or is this my.

Speaker A: Uh, just briefly. I think that the message was, you know, the students can win and be careful of the students.

Speaker B: But what was interesting is why the students. Because I've, like, m. My wife is. Is Serbian. I've. I've spent a fair bit of time there. For an American, I guess I'm probably in the top.

Speaker A: I thought she was French.

Speaker B: I'm in the top. Like, definitely, uh, in the top 1% of Americans with their familiarity with Serbian affairs. And it was very exciting for me as, uh, someone who got married in Serbia and has friends and family in Serbia that now a lot of people are discussing. You know, this thing that I've seen kind of up close because I. I was in Serbia two years ago when this movement started after a tragedy, uh, in. In Novi Sad, where a railway basically, like, overhang like, collapsed and killed, like, 17, I believe, people. And it was all because of corruption in the government and in their procurement and stuff. And this is typical, unfortunately, in that. In that part of the world and maybe even in our part of the world, who knows? I've been. I've read Some stuff. But this student movement sprang up and they had everything going against them. In particular, because Alexander Vucic was the long serving kind of quasi strongman, kind of the Orban kind of Trump mold, you know, competitive authoritarianism. M. I would, I would call it. He controls all the levers of power, just, you know, the judiciary, the police. Absolutely. The intelligence services, which are much more powerful in these, in these societies, but then also the media. He took over the media and, you know, the students were able to break through because they did not compete in. They basically decentralized themselves. They had no visible leaders and they held all of these constituent assemblies all over the country. It was very like communist coded. You know, they went back to the playbooks from, from Yugoslavia, I think, in many ways. And they were able to outmaneuver a centralized authority by taking a decentralized approach. They didn't choose anyone who had experience in politics for their list of candidates with the way their parliamentary system works. And I think it could become a playbook for translating a lot of this energy, particularly among younger people who rightfully feel like they've been screwed over to how to get power. It's one thing to expend energy and to shut down campuses and all of that stuff. It's another thing to get power. And they got power. So congrats. It also brings to the fore, because the way they did it was not by having a set political, you know, agenda, but by having broad principles because they want to tell the story and it's everyone's against corruption. But like the tent that they built because this is Serbia is filled with all kinds of characters. There's mystics, there's healers, there's nationalists, there's ultra nationalists, there's liberals, there's like, everyone under the sun is in this like, coalition. So, you know, it's great parties now. I'm sure there's going to be bloodletting later. And I just find when people look into a really complicated thing like Serbia and it's, it's, it's politics, they come to the wrong conclusions because they're always just looking at it from how they think of their societies being run. These places are totally different. So anyway, why

Speaker A: are Serbians the way they are? Why are they that way?

Speaker B: History? I mean, they were, you know, the, the, they were part of the Ottoman Empire and you know, they are, in their national narrative, they, you know, they're. I always find a nation whose, like, sort of touchstone is, is a loss on a battlefield is a little bit different and Theirs is like a loss in the battlefield in. In the 14th century to the Ottoman Turks. But in their telling, they sacrifice themselves for Europe in holding back, like, you know, the Ottomans and they sacrifice. So Poland has. Has some of that with Yalta and some other things. But, yeah, I mean, it just leads to their national characters is just being obstinate and.

Speaker A: Do you like to party with Serbians?

Speaker B: Oh, yeah. Serbians are tons of fun. They're very hospitable people. You know, they have a, uh, they have a little bit of a gruff exterior, but they're very hospitable and they're very fun. That's why I think the scenes out of Belgrade have, you know, they've traveled well as marketing. They've done. They've done really well with this. And the Serbs are terrible at pr, But I think they're. They're better at marketing because they love. They love drones. There's always pyrotechnics involved in any kind of, like, Serbian, like, celebration. Like, pyrotechnics were at, like, our wedding. They really are into pyrotechnics.

Speaker A: You guys have fireworks?

Speaker B: Yeah, they had some, like, they, you know, there's. It's anywhere from, um, sparklers up to mortars, you know, and there aren't a lot of regulations, so, like, you know, there's a lot of gradations. You start with the sparklers, but you could end up with mortars.

Speaker A: Well, should we do a quick hit on good product?

Speaker B: Yeah, let's do it.

Speaker A: You know, I actually love my sort of retro tech stuff. I, uh, like, you know, old amps and cars and stuff. And I like. I got this. This friend of mine sent me. Did I already use this? This isn't what I'm going to focus on, but a friend of mine sent me a really nice Marzocco espresso machine. It's called the GS3. He did it because it was in his office and he had no use for it. And it is really the Cadillac of, uh, of espresso machines made in Florence and, you know, beautiful hardcore equipment. This one's a semi automatic, which means that you press, you know, your selection and you get a certain amount of water that comes out. But it's a beautiful thing. It truly is a good product. But m. Mine is a little bit more. My. My real example is a little more pedestrian. Uh, and again, I can't remember week to week what we've done, but a lot of times when we're on the road, we stop at Starbucks. Do you stop At Starbucks, like, drive through?

Speaker B: No, I've never been to a Starbucks drive through. I,

Speaker A: I go to a

Speaker B: Starbucks. I go to a Starbucks every now and again, like in an airport or if I'm rushing.

Speaker A: I, uh, they've, they've up, uh, they've, if you go in, you know, they've, they're shutting down stores and improving, you know, the ones they have and making them nicer. They have nice furniture, and they've, they've improved service and the quality of the food and all that. But they make a egg sausage bacon scramble wrap that. I'm a huge Starbucks. Yeah, I think it's a good product. I, I really do. I, you know, I love the breakfast offerings from, from fast food restaurants. We've done the,

Speaker B: we've done the egg McMuffin.

Speaker A: Well, and I, and I, you know, I do love the sausage egg McMuffin. Uh, a huge amount. Just to remind people. I think that's, it's, it's a great product, and this is my replacement. I think it's, it's probably a little fewer calories, it's a little more healthy. It is made in an airline oven, but it's, it's, it's, you know, it's kind of softish, scrambled, little piece of bacon, couple little sausage balls. And I usually pair it with a iced matcha latte light on the ice, and that's my go to order. I think it's about 12 bucks, and it's, it's kind of perfect order. Well, it's like maybe 12.99. If they ask you for the dollar tip, then it's 13. It's great. I love it. And it's my road food. It's always my road food.

Speaker B: Wow. Yeah. I don't think I've ever, I don't think I've ever had hot food. I remember when Starbucks got the hot food. It was a little controversial because it comes with odors.

Speaker A: Really?

Speaker B: Yeah. No hot food, like. Yeah. I mean, that's why I don't think that you should. Uh, yeah, the flight over here, I'm a. It left at 11:30. And they're serving meals for like.

Speaker A: Yeah, I highly recommend you don't eat that because you won't feel good the next day.

Speaker B: C.J. uh, Gustafson is here from Mostly Metrics. C.J. when you're flying, you might be doing private now. I don't even know. But when you're flying scheduled, do you. And, and it's like you were just in London. Did you eat the dinner, like, on the airplane?

Speaker C: Yeah. It was surprisingly amazing from my middle seat on Spirit. Spirit Airlines?

Speaker A: Yeah.

Speaker B: You can't take Spirit abroad.

Speaker C: No, I'm kidding. Um, I mean, not private. Across the pond. Brian, that would be. That would be egregious. What

Speaker B: does that cost? The

Speaker C: dinners you think I'm throwing?

Speaker A: How come you're not wearing your get up, man? You're, like, adjusted. Even though wear.

Speaker C: I got the hat behind me. If you want me to put it on. I've been. I've gone deep down the merch rabbit hole. It's, uh. It's a lot of fun. It's addicting.

Speaker B: Is this. Is this for revenue? Is this for brand? Because, like, I feel like, you know, you're doing great. I'm very. I think we're all, like, happy for you, C.J. but, like, what it. Like, why. Why. Why are you doing merch? Why are you doing ebitda?

Speaker C: I'm doing it just for fun. Honestly, I think I. I think I do a lot of things for money. This. This. This is for me.

Speaker B: I like that. You gotta have a portfolio, and you gotta.

Speaker A: You know, I. Brian, I have. I don't know if I've ever given you any. I have my own hobbyist merch line. It's called the. It's called Rila. Ram Island Liberation Army.

Speaker B: Oh, I

Speaker A: know. And. Yeah, no, you haven't

Speaker B: given it to me, but.

Speaker A: Well, it's. We have an agenda. We have a whole. You know, we have a philosophy. We have a aggressive idea about what we should be doing, me and another friend on Ram island, which is connected to Shelter island, which. Where we live. Yeah, we made a lot of merch hats, hoodies, shirts. Reload. This year. Next year, we're actually doing a drop, C.J. we're doing the Ram Island Liberation Space Force, which is a nice, um, brand extension. And it's going to be an osprey in a space outfit.

Speaker C: Troy, I was really sad to hear about your car getting broken into, man. Sorry. Didn't

Speaker A: get broken into, cj. It got stolen. Stolen?

Speaker B: Yeah,

Speaker C: I

Speaker A: got it back.

Speaker C: You know what it made me think of? Have you seen the other guys with Mark Wahlberg and Will Ferrell?

Speaker A: Yeah, you've seen it. You're,

Speaker C: uh, Dirty Mike and the boys, they keep breaking into his car.

Speaker A: My wife didn't want the car back, and I had to talk her down. She thought that there would be, like, methane stuff in there. Bodily fluids.

Speaker C: Yeah,

Speaker B: it smells.

Speaker A: So anyway, cj, I got questions, man.

Speaker B: Yeah,

Speaker A: cj, this is

Speaker B: your Super Bowl. This is your super bowl right, like as uh, and everything is coming your way. I gotta say everything because like I think there's so much in these tokenomics. Like the CFO area is going to be the locus point for where this goes with AI because this is my theory. Like the bear case. Dario said it best. He was like if we get the timing wrong on this build out, we could go to zero. Like if we're off by a year, like we're screwed. And it's all premised on whether the, and I think it's going to happen in, in, in the CFO suite is whether or not this is going to drive the kind of productivity within non tech businesses that at least that anthropic is claiming with its within its own business. It went from 1.6 million, $1.4 million per employee sounded good to me. I was like okay, but it's not outstanding to like over $6 million employee in like nine months. So break down the uh, S1 for us. But don't get too nerdy.

Speaker C: I'll try not to go too deep. I'll uh, give you the high level cuts here. So just for context, so people know how big this company is, they did just under 20 billion over the last 12 months. So if you like just had a rolling calendar year, they, they generated almost $20 billion and they did 11.6 billion of that just in the second quarter alone. So in three months. So that was up 1,400% from a year ago, which it blows all records for, for growth of a company this size. And it ipo, there's a lot of back and forth of like people always saying oh well, what's their ARR, their annual recurring revenue? I like to look at this method because it's, it's just more honest of like how much revenue they've actually done. They're growing so fast that they'll probably end this year at something close to close to $100 billion in run rate revenue. And they're doing it with efficiency that we haven't seen before, at least on the revenue side. So they have, they have over 5,500 people working there, but they're still generating 6.6 million per person. And so for context, the median public SaaS company generates about half a million dollars per person. Okay. Top decile, the best of the best out there, which you alluded to Brian. They do 1.5 million per employee. So by my math

Speaker A: that's like the rebooting Inc.

Speaker C: Yeah. The only company per employee out there. The only, the only Company out there with better revenue per employee is OnlyFans.

Speaker B: Really.

Speaker A: And they're also upset over at OnlyFans because of anthropic. So that's a nice connection.

Speaker B: By the way, do you know who got screwed over? Is these the only fans chatters? Do you know the chatters?

Speaker C: What's that?

Speaker B: They're being. It's $6.5 billion industry. It's people usually in the Philippines, but in uh, other outsourced that take over from the only fans models and will chat with the whales because that's, it's all whales. And so you get the whales to keep like spending and keep spending, but they're just chatting with like someone who's hired like outsourced,

Speaker C: can't believe anything anymore. Wow.

Speaker B: Yeah, I know.

Speaker A: Okay, okay. So it was

Speaker B: impressive.

Speaker A: Don't, don't. It was impressive. What else?

Speaker C: It was impressive. I think the number one question in my mind is if the use cases for AI can keep growing at the same pace that the cost of the tokens are falling. So you can totally mistime m this to your point, Brian, where they can actually work themselves into a situation where it's not as economically viable to keep growing with the models they have if everybody keeps transferring over to them, but also coming up with use cases on the stuff that drops by 40%. So Opus 5.5 cut the cost per task of the previous model by 40%. So what's going to be the trade off there with the cheaper models?

Speaker A: This is where, you know, Jevons Paradox breaks down. That's a term that they like to use in Silicon Valley, which is that it works at certain parts of the demand curve. But if your cost comes down too much, even if people voraciously suck up more tokens, you can still be pooched, as they say in, in, in Canada,

Speaker C: in, in what it's like, ah, at what point is it smart enough, right? Like for a lot of the stuff that I'm doing, I'm doing this report on the state of the finance tech stack, right? It is legitimately 10 times better at helping me with this report than last year. And it's better than I was as a human putting these together two years ago in PowerPoint. So I don't know if there's another 10x jump that I can even do for this task to be done. And that's just me in my world. I'm thinking about, like, even in finance, you probably get to a point with forecasting where I don't know how much better you can get okay, we're within 99.5% of the budget. Can we forecast to 99.7? I don't know. And so I think we're going to get to a point, and this was in the S1 of like, we're going to have to invent use cases. And they made a big push at the beginning to. And this is where like, the storytelling comes into play. They had, in my opinion, the best shareholder letter. So the founders always start an S1 statement, which is saying, this is the first time you see our company's financials. We're coming out of, you can buy this, you can buy stock in us for the first time. They write this letter instead. And they led with, these are all the things in science that we're trying to do to help cure cancer, to cure heart disease, etc, and so they were showing all these things that impacted people on a human level, but also increased what you would call, like, the productive capacity of what these models could do for the economy.

Speaker A: That's a great point. Hey, can I lay something out for you and get your take on it?

Speaker C: Sure.

Speaker A: C.J. okay, so it's just to state where we are, it's pretty clear that models can be changed out, right? Like, you know, you might use a good model, you know, Gemini one week and another later, and that tokens fall in value. It's also seemingly clear that the way that the world will answer your competition using AI well is to use more of it on your side. So there's always competitive dynamics that support a case where we use more and more of this stuff. You know, Jevons is great, right? Prices fall, you use more, but if, if it falls too much, it's not great, right? As you just pointed out that the harness and memory and applications are sticky, it seems to me. And I just know that personally, like, I like Claude, I use Claude and I. It's now, you know, deeper and deeper and deeper with all the institutional memory and that that's really sticky. And as you get into, like, you know, there was a case today where they've got this legal document platform. There's going to be a lot of cases where having the, you know, the behavior with the consumer combined with a vertical use case allow you to jettison competitors, I think, I think it's going to be very, very powerful, right? So I think that compute and brand and applications and memories solidify a leader's position, right? But their pressure is going to be to go deep, right? Open the wet lab, cure cancer, do all that stuff. And I Think the next bit here is they're not going to release the stuff that's super cool,

Speaker B: they're

Speaker A: going to keep it back. They're going there, there's going to be shit in there that will do amazing things. There's

Speaker B: just too much risk. There's too much risk beyond that there's, there's so much risk and they, they've, they're just going to use it themselves to run

Speaker A: this idea uh, of them having models know nothing about. They, you know, they're going to use it to their advantage in ways that nobody appreciates yet. I guess my point is that people like the more or less people, what's his name more or less the guy Brian. That uh,

Speaker B: Sam lesson

Speaker A: is like the, this is sort of like utility level stuff. It's going to fall in value. The comp, the frontier companies are going to go bust and I think that it's more of a Google situation where we are going to get comfortable with these tools. They're going to get better at trapping us in their ecosystems and the use cases are endless. And when I look inside of companies that lean into AI and I see what they're doing, people like Brian or bigger companies, it's becoming the entire kind of operating infrastructure of these companies. That's where all the money I want to

Speaker B: get to the 40 trillion. Like

Speaker A: uh, I want to know where CJ stands personally. I want to know how he is he, is he long or short.

Speaker C: So what's the most mind bending part of this is you can be selling something that's equivalent to alchemy but it's also commoditized. It's also commoditized product. Right? Because there are other large language models out there that you can pay for. There are also open source ones. And so in a lot of ways it reminds me of the airline industry which is incredibly capex dependent. They have to buy all this physical in the real world and m you still price shop even though you're flying through literally time and space in this metal tube. It's like if we went back 100 years and we were talking about flying or whatever. I don't know when they came up with it in Kitty Hawk but like it was the equivalent of alchemy back then but now it's a completely commoditized service. And so that's the part on the valuation piece that is hard to triangulate that this is the most technically superior product we've ever had. It's completely changing how we do work but the cost is Also going to be driven to a unit of value that's similar to buying like oil or electricity for your house.

Speaker B: But uh, isn't the counter to that? Flying is not a general purpose technology. Like they can point this thing at like a hedge fund, they can point it at the wet lab, they can point it at like law firm. Like it can do so many different things. It's just a, like that they can. And they get so much data on how people use it that they're just going to pick the most valuable things, keep back their, their, their most powerful models and just go after the like uh, that's the only way they get to these like $40 trillion of TAMs. Well that's

Speaker C: why they've started to go into specific areas of domain expertise like design or personal finance and why they're releasing those plugins because I think they're realizing that the value will accrue at the application layer. And if you're vertically integrated and own the large language model and also the application and then they're also building out their own data centers now so that they alluded to in the S1 that we partner with all these different companies but we're trying to be as vertically integrated as possible. And so they're, they're making sure that they have their bets spread out in terms of where the value will occur.

Speaker B: But what, so when you look at the. Give me the, give me the bears case. When you look at the, the numbers and with the adjusted EBITDA and whatnot, like what is the, what's the bear case in the numbers?

Speaker C: The, the bear case is that we're not able to come up with enough shit to do to substantiate the investment that we're making basically off. Okay.

Speaker B: So that it's that mismatch thing like. Okay, we overshot on how much compute.

Speaker C: Yeah. And if you think about like the US economy is financial statements, we have invested a ton off the balance sheet but that hasn't amortized over to the P L yet. What I mean by that for all the non financial nerds out there is we haven't come up with other uh, enough like revenue generating capabilities to substantiate like Meta announced the other day, like I think it was a hundred billion dollars in bonds or something. So we were going hard into this and eventually it has to show up in revenue. Whether that's through the LLMs, making it directly through the tokens, through the applications, et cetera. So that's, that's the bear case. But I think the the bull case is that it's fundamentally changing not only the ceiling of what people can do, but it's also replacing some of the labor budget that was at a lot of these companies before. So it's the pie is getting bigger but they're also absorbing a piece of that budget envelope at all these companies. That was for humans before.

Speaker A: So when you read the S1, was there anything in it that you were like holy shit, other than the 10x revenue increase?

Speaker C: The biggest holy. What was that? Honestly, it was the fact that like they are, if you squint your eyes and take out all the stock based comp, they have had a profitable quarter. I know that they're investing a lot but like I don't think a lot of people expected them to be at least this close to profitability.

Speaker A: Isn't that an adjusted EBITDA story because of, you know, the way training's accounting for and all of the other capex?

Speaker C: So the bigger like bugaboo for me was on the piano that a lot of the training stuff was kind of in the wrong place on the P and L. So their gross margin looked a lot better. So for people out there who aren't as familiar with the P and L, you do your revenue minus your cost to serve or your cost of goods sold to get to your gross profit margin. And gross profit margin is important because it's a theoretical ceiling on how profitable you can be if you didn't have to pay for all the other stuff like within the business. And it went away. Right? That's like your starting point. And they made the growth, the gross margin look a lot better by uh, they, their biggest growth lever is giving this shit away for free. Right? They let people try it without paying for it. They put all the free trial inference down in sales and marketing. It's typically up, up above in the P and L and hits your, your gross margin. They also took all the cost of paying their channel partners 50% of revenue in the second half, in the last half went through the Googles, the Amazons of the world. Right?

Speaker B: Are they, are they penalizing OpenAI? Because OpenAI is not going to be able to do that because they're monetizing it through ads. So they're going to, they can't put it as sales and marketing.

Speaker C: It's kind of apples to oranges because, because of the ads like that, that shows up in a different way. But OpenAI was the one who raised the alarm and was like your revenue is actually 3 billion, 2.7 billion lower than what you're saying, because that was used to pay other people to sell for you. So that again, Troy was down in sales and marketing, not in gross profit. So,

Speaker B: uh, this is all like Google all over again with tech and all that.

Speaker C: Yeah, it's like, where do you, where do you put these things to look better? And I think that's why a lot of people like me spent a lot of time looking at like the definitions. Because it sounds like a small, sounds like minutia, but it actually reflects like what the, what, what like the profile of your company is in the durability of it.

Speaker B: I mean, because that uh, was back in the day, it was traffic acquisition costs that like was. It's like, what do you pay to get it to you? And like they were like, oh, I don't know that that was why. Anyway, go on.

Speaker A: I like your dog.

Speaker C: Thanks. He's a big fan of your show.

Speaker A: So that's, that's what you got. And now are, uh, when are they going to go public, CJ Is it, when's it going to happen?

Speaker B: Soon?

Speaker C: It'll happen before the end of the year. I think they want to go before Thanksgiving. We'll see if they get it done.

Speaker B: Okay, so first earnings call. I know you don't know what the earnings are going to be. And, and then these, these analysts get on and they're like, great quarter guys. And like Dario, uh, are, uh, they going to let Dario be on the earnings call? He can't be.

Speaker C: Great quarter Dario. Great

Speaker B: quarter

Speaker C: diary. Try

Speaker B: to, try to, try to keep eye contact. We'll see if he's on the earnings call. But what, what's the first, what do you think is going to be the thing that financial analysts are gonna zero in on in their like first question after great quarter guys,

Speaker C: it's going to be all about the revenue growth rate for at uh, least I think the first two years. I, I think the profitability is something people were hoping to check the box on, but it's, I mean what people are under underwriting here, Ryan, is the ability to grow non linearly for periods to come. And if you look at what you get in terms of valuation Multiple, you get 3x the amount of uplift from a point in revenue growth than you do from a point in profit. So even though we say profits matter, it doesn't show up that way in the regressions.

Speaker B: So is it too simplistic? Probably. But like for me to say they have to do what they did for coding with like particularly VC funded AI native Companies, they need to do that with not just the Fortune 500 1000, but like all the businesses out there, in order to reach their quote unquote potential.

Speaker C: Well, their potential work. It's not cooking

Speaker B: your friends Ramp. I think they're your friends of yours. They say that 1% of spenders make up 80% of spend. Ramp is like very heavy with startups. Like, I don't think that Mondelez is using Ramp. I could be wrong. But you know, so I'm just saying the data is, you know, a little bit geared towards early adopters. But you're saying like it's. But they have to prove, and it seems like companies have not yet been able to find the productivity gains in other knowledge work, broadly speaking, that they have been able to find with coding.

Speaker C: Completely true. I look at it like engineering had their quote unquote chatgpt moment with like Codex and with cursor then support had it with the likes of a Decagon or a Sierra. And now finance is starting to have it with the ability to use MCP to get your data from various different sources. But they're going to have to do this, I think, department by department and industry by industry because what they quote is their TAM is all of knowledge work. They don't say our TAM is coding. They don't say our.

Speaker B: No, I know

Speaker C: it's healthcare.

Speaker B: So how is it, Is uh, it going to like wipe out a lot of FP and A people and stuff? Like, how is this going to change the, the finance department? It seems like a lot of this stuff can be, can be done by Claude.

Speaker C: A lot of it can be done by Claude, I think.

Speaker B: But I don't want to be sued and I don't want to end up in an orange jumpsuit. That's my problem.

Speaker C: Exactly. So I don't really know. I don't, I don't know how, how that part shakes out. But what I do know is it's the best time in the history of humanity, I think to be a small business owner with all this stuff, it's crazy.

Speaker B: Woohoo.

Speaker C: Yeah.

Speaker B: All right. Cj, do you have any, any London tips for me? You were just here. I'm going to Fat Badger tonight after, after the book party. Great bar.

Speaker C: You're more worldly than I am. I just followed my wife as, uh, she shopped in different stores and it was.

Speaker A: Well, I have one tip. You should go to Tamarind in Mayfair. It's my favorite Indian restaurant, so

Speaker B: I've been in that basement. Yeah,

Speaker A: uh, they have an upstairs, too. It's good. Also, cj, do you like Cheesecake Factory? Do you have one in your neighborhood?

Speaker C: I haven't been there since, like, high school, but that used to be. That used to be great. They got quite the menu.

Speaker A: Better.

Speaker C: Oh, yeah. PF Chang is great.

Speaker B: All right, let's wrap it up.

Speaker C: All right. See you guys. See

Speaker B: you guys.

Speaker A: That's it for this episode of People versus Algorithms, where each week we uncover patterns shaping media, culture and technology. Big thanks as always to our producer, Vanya Arsenov. She always makes us a little clearer and more understandable, and we appreciate her very, very much. If you're enjoying these conversations, we'd love for you to leave us a review. It helps us get the word out and keeps our community growing. Remember, you can find People vs Algorithms on Apple Podcasts, on Spotify, and now on YouTube. Thanks for listening, and we'll see you again next.