Google's Huge Antitrust Win, AI Agents, and Stablecoins | Sep 3, 2025

0:13Welcome, everyone, to the Informations TITB. My name is Akash Pasricha. It is Wednesday, September 3rd, and we are diving deep into the analysis of the Google antitrust ruling from yesterday. We've got a great discussion planned with everything you need to know with our editor-in-chief, Jessica Lesson, and our co-executive editor, Amir Afradi. We're also talking to antitrust lawyers about the ruling to get their take on what this could mean for other antitrust cases in motion right now. We're then going to bring on some more friends of the show, the co-founders of AI agents company, Utori, and we're also talking to stablecoin company, Rain.

0:48There is a lot going on. Let's get right on into it. Yesterday, an antitrust judge dealt Google a big win in ruling that the company can keep its Chrome browser. It can keep making deals with Apple to be the default search engine on Safari, even though courts found last year that Google was operating an illegal monopoly. It is hard to overstate the significance of this deal for the tech sector. And last night, we put out a special edition of TI TV, part of which we want to play for you today. It is a conversation our editor-in-chief, Jessica Lesson, had with co-executive editor, Amir Afradi. And where I want to start this clip is where the two of them just start getting into how heavily the proliferation of generative AI impacted this decision.

1:34Here is that conversation. Well, OpenAI won this case for Google. And it's pretty incredible that most of the firms in Silicon Valley, including these big tech firms, they are run by a certain brand of free market capitalists. And this decision, you know, could not be more aligned with that view, which is that, you know, when you think about these kind of antitrust cases and issues, they're so backward looking that by the time they're done, the world has kind of changed. and Judge Meta agreed with that essentially and almost raised the whole question of like, why did we go through this entire process to begin with?

2:24You know, the initial trial where he determined that Google was an illegal monopolist ended in November, 2023. ChatGPT was a thing, but it was still in its relative infancy. Since then, it is obviously, you know, jumped by an order of magnitude or more. in terms of the people who are using it. But going back to what you were saying before about kind of big tech and kind of compounding advantages, it's not as if this new landscape is going to be the complete and utter decimation of Google. Its advantages persist. And I think part of this case definitely got into that, which is why Judge Meta is saying, yes, you have to give OpenAI your search data and give them a better chance to compete.

3:10because Google is essentially taking its gigantic, greatest business ever created, you know, canon and training it, you know, toward AI in terms of bringing AI results and making Google search more like ChatGPT. It's able to do that, you know, in ways that no other company really can. Now, there are limits to it. Obviously, ChatGPT is growing like a weed. And, but so far, you know, it has not crushed Google. So far, Google is taking these baby steps toward having a better product that gives you the kind of results you want or you have come to expect, especially if you use ChatGPT. So they were going to be okay anyway.

3:55This decision just absolutely reinforces this idea that Google is going to be fine. And now they don't have all these distractions. If you think about if they would have had to divest things or kill the Apple deal, I mean, putting aside the insanity it would have been for Apple, it just means that they don't have to be distracted anymore. I don't even know if they're going to appeal this decision. It was so good for them. That's a good question. There are some other cases. Not as important, right? But talk through those. They're not totally out of the wood. Yeah, there is an ad tech case, which is really all about Google's advantages in advertising technology and how it has essentially kind of tried to rig advertising auctions and find a way to get more and more margin from the way that it helps companies buy and sell inventory across the web.

4:50There is a serious thought of having to kind of spin out part of that business, but it is a part of the business that Google has been kind of deprecating anyway and is not as critical. This was really the case. This was really going to be the moment. And I think this is a huge, huge crushing blow to the people who brought this case to begin with. I think companies like Microsoft are probably quite disappointed given the decade and a half work that they put into this. But yeah, that's where we are. We've got to get Luther Lowe on the line. Maybe dial him up. And you know, you're right. I mean, not that, I mean, obviously the tech regulatory wheel will turn and, you know, massive companies, we were just talking in our edit meeting this morning about how do you regulate AI and how do you think about responsible AI and all these questions.

5:41You know, the scrutiny is not going But let's shift gears a little bit and just talk about this moment of regulation in the Valley. I mean, we've seen... we're waiting a meta breakup case decision from the FTC, I think. I've lost track. We need our regulatory tracker back on here. You know, under the Biden administration, there was this steady drumbeat of it was looking like everyone was headed for their big breakup case. Again, many of these obviously way predate the Trump administration, to your point. In general, the Trump administration seems like it's backed off on a lot of areas like privacy, but I kind of feel like there's a little bit of a collective sigh across all the big tech companies this evening.

6:25Because really also, I mean, when you talk to the CEOs of these companies, you know, yes, this product launch and that product launch, but it's really the breakups that they know are at the end of the road, right? it's hard to imagine you're going to have 10, 50 trillion dollar companies eventually, right? So like this breakup notion does hang sort of existential. And it was kind of feeling like this, you know, was going to be a little chip off the old Google. But I, but, or am I, am I making too much of it, right? Is each case individual here? Well, you know, even in the worst case scenario, you know, it's a sort of, it's a sort of thing where even if you do have to break off an Android or a Chrome, the existing Google shareholders get a piece of that, and they're going to do just fine.

7:16These are inherently valuable. No CEO wants to be broken up. Yes, I mean, that's the shareholders can make a case, but we're not, but it's a different landscape, right? Where these companies to be cleaved off, right? Yeah, yeah, for sure. And look, the meta situation, I mean, these are deals that that happened like you know 15 years ago uh or longer it just seems really hard to believe especially given what we saw in the google case that that suddenly a judge is going to require uh meta to spin off whatsapp or or instagram or instagram gram yeah tired um yeah and and that has always felt quite confident about that case but we'll see i mean I think these things tend to happen in waves a little bit.

8:03And so it's certainly one of many messages. Acquisitions are back, you know, a little bit of collective relief. Before we go, because we've got stories to edit and publish shortly, but let's talk about Google. You said earlier that, you know, it's a sign sort of not to count them out. Maybe people have been too hard. certainly, you know, the stock has been at a relative discount to other big tech companies, maybe because of regulatory kind of overhang, Gemini inching along. What's your view? I mean, do you think that the public, public being both Wall Street and the sort of insider tech, had turned too negative about Google and this changes that?

8:48Or do you still see reasons, you know, to be concerned? Well, there's the general concern around ChatGPT, And once it starts monetizing through advertising, what kind of a bite, if any, is that going to take out of Google? But no, I mean, Google is going to be a, you know, top two, top three company in whatever this kind of combined AI and search paradigm is going to be. So they're here and they're not going anywhere. So from that standpoint, it seems like they're going to be fine. And, you know, they're - But will they be a monopoly, right? I mean, I think that that's, and it's interesting because that brings us back.

9:33I remember, I don't know, maybe a year ago, six months ago, Brad Gerstner, the investor who was one of the first to sort of turn more negative on Google, right? And basically said, chat GPT's here, Google's monopoly days are numbered. and therefore, you know, you just can't value the company the same way. Yeah, I mean, that's fair. If you're looking at it from like a, you know, price to earnings ratio kind of thing, yes, there are implications here. At the same time, you know, Google Cloud is growing like gangbusters. It is not a massive, massive profit center, although it is getting better in that sense.

10:11You've got kind of Waymo in the background that's looming. If it could scale, you know, that's - Oh, not in the background. I mean, buy our office 20 times a day. About buy our office. 20 times an hour, actually. So, you know, they do have all these other potential tailwinds, as analysts like to say. But you're right. I mean, the era of Google, you know, owning, completely owning the most lucrative, profitable, you know, business or product that's ever been created. that is starting to change for sure. So from a valuation perspective, I think there is a bit of an adjustment, but man, all these like compounding effects of being who they are and acquiring what they've acquired and being able to subsidize these newer businesses for as long as they have.

11:04The fact that Google as a consumer company has been able to subsidize and slog it out on the cloud infrastructure side for as long as it has to now becoming a player that is able to have open AI as a customer and meta platforms as a customer. That's crazy. That's incredible. It's an incredible thing. I remember years ago when I think you broke the story, and I should add for our listeners, Amir has been at the information 12 years we've been doing this. So we've written a lot of Google stories, but you had a story saying that Google had set some sort of goal. If it wasn't hit, they were going to shut down Google Cloud, right?

11:42There was that much skepticism and pressure. And I think to your point, its growth has been a major story over the last three months. And it's sort of strategic wrangling with some of its competitors. So, yeah. Well, I've always believed that you can't count Google out. But it is, which seems silly to say because it's Google. But it is interesting to see such a decisive win. I mean, I'm also interested if we, you briefly indulge me, my media criticism hat too, you know, how outlets have been covering it this evening because many outlets like the New York Times were quick to just point to the search licensing or search data sharing as, you know, a blow delivered by the Justice Department.

12:32And I mean, to me, as soon as I got a text from Martin Pierce in New York saying decisions out, I just said Apple Chrome. Like, those are the two things that matters, right? Does it keep Chrome? Does it keep the Apple deal? And such a clear victory. So it's interesting. I think you have a range that we're in a moment of a wide range of media coverage. And some people will be quick to point out, again, and we should, those areas where it's not just business as usual for Google. But as with most things, text, really a question of what's needle moving and what the company is very happy to change to make the case go away.

13:10So as you said, I think Sundar should buy Sam a beer tonight or 20. Tim Cook should also. Tim Cook, too. Yeah. Big, big thanks. Big, big emoji prayer, you know, going back and forth on the iMessage. But it will be very interesting to see if we get some more reaction from Google. Given there's continued litigation, they probably aren't going to be beating their chest. But, you know, it was a significant deal, one that was a long time coming, both in the verdict and now in the decision and the remedies. So, Amir, thanks so much for joining. We should keep the Jessica and Amir show going on TITV.

13:55There's actually been a request for us just to debate each other since we have some healthy disagreements on these topics. We'll bring Martin Piers in here and it'll be an explosion. It will just be boisterous. That was Jessica Lesson and Amir Afradi. Well, we are continuing our coverage of the Google ruling, and I want to bring on someone with a background in antitrust law. Joel Thayer is the president of the Digital Progress Institute in Washington, D.C. Joel, it is great to have you. Welcome to TITV. Thank you for having me. Excited to be here. So let's get to the ruling. Did it surprise you?

14:31Not really. I think we got exactly what we expected. We got a very strong liability case and a somewhat measured remedy case. I mean, I was a bit surprised on the default side, but overall, it made perfect sense. It looks like the judge was definitely trying to take a narrow focus on the specific facts of the case and try to address the remedy the best way you saw it. So no surprises here. What did you make of the idea that this ruling was very forward-looking? You know, I think we think about antitrust cases as looking in the past, and they take a long time to get through the courts eventually.

15:06The mention of AI here was explicitly a mention of, hey, we're taking the future into consideration. Did that part surprise you? Not really. Judges typically do a bit of forecasting. You kind of have to when looking at antitrust cases. It is based on economic principles at its base. So you do have to look at the market, and you also have to look at when the claims were lodged, when the alleged offenses happened, and whether or not those alleged offenses still exist in the current market, and whether or not your remedy is going to impact future growth of a market. So that wasn't all that surprising to me.

15:45So the AI piece, I mean, there was no sort of reflection that you have there. I'll tell you what. So rather than talking about the AI, because we just had Jessica Lesson and Amira Faradhi talking on the show about that, one of the things that I wanted to talk to you about actually a little bit more is given your antitrust background, we have these other cases too against Apple, against Meta, against Amazon. And what we're trying to figure out here in the newsroom is what this ruling could mean for those cases. Did this ruling tell you anything about which way the judge might rule on those cases coming up?

16:17Yeah, I think it's going to be a tough road to hoe for the tech company. especially when it comes to a specific remedy that the court did grant, which was the data sharing remedy, which will allow smaller competitors to basically leverage some of the infrastructure that exists from the big tech players to grow and actually scale and be able to compete. So what it would signal to me was that these judges are starting to recognize how these markets operate and what is the actual widget being sold here and what's actually being harbored. So what my read of this and that particular remedy being granted demonstrates clearly to me that we are no longer living in a world where big tech cannot be touched.

17:02We are now looking at, judges are looking at them far more critically. So if I were a betting man, I would say that I would probably bet on the government on those cases over big tech. But say more about the data sharing piece. Why was it that the data sharing remedy indicated to you what will happen in those cases? I want to unpack that. Sure. So for a very long time, without boring your audience, we operated on what's called the consumer welfare standard for antitrust. And that basically suggests that there are certain aspects of the market that have to benefit the consumer. So we're actually okay with, you know, three companies ruling them all, provided that the consumer gets some sort of benefit out of it.

17:45And typically how it's been reviewed is through price regulation or at least price moderation. So if, you know, this is the pricing competition angle, essentially. That's right. So does the monopolist able to control the price of a product? So the question with the tech companies, like, well, what happens when the price is free? So if the price is set at zero, I mean, doesn't the consumer always benefit? It looks like there is going to be a significant turn on that because now it's starting to look at the other aspects of the consumer welfare standard, which the consumer welfare standard also stands for making sure that monopolists can't harm innovation and that the outputs themselves cannot be controlled by particular firms.

18:28So other aspects like? Like promotion of innovation versus inhibiting innovation or quality of the product, for instance. So there was a lot of things that were getting discussed in this case that I thought were very pertinent to a lot of these other cases. One is that it looked at the quality of the search results based off of Google's monopolization. It turns out that the court was not at all sold on the idea that Google's monopoly was important to maintain the integrity of the quality of search. In fact, the data sharing arrangement basically says that you actually have to allow other companies that may do this.

19:06better than you to scale upwards and actually compete with you so the consumer can get better benefits out of this. And he even cites the consumer welfare standard as a justification. So that is a monumental shift. And I think that is something that's going to reverberate into these other cases because it's the same analyses that you're seeing in the meta case. It's the same analyses that you're seeing in the App Store case with respect to Apple. So again, this is, I think this is going to have a much bigger impact than most people are probably giving it credit for. Great. Well, Joel, thank you so much for coming on the show.

19:40I think this story is far from over. We're going to see all these other cases play out. And so as we do, we're going to bring you back on the show to walk us through some more of your analysis. That is Joel Thayer, the president at the Digital Progress Institute. Okay. Well, it is always exciting to see when teams of big tech talent leave their companies to start their own new ventures. That is exactly what our next guest did last year. Devi Park and Abhishek Das left Meta in 2024 to found AI agents company Utori, which raised$15 million a few months ago. The company aims to build what it calls an AI chief of staff, and it is backed by all sorts of big names, including Conviction and Felicis.

20:20Devi and Abhishek Das, I'm to call you, you said, welcome to you both. It's great to have you. Thank you. Thank you for having us. Yeah, thanks for having us. So look, I'll decide who gets to explain the most fundamental question, which is what is it that Yatori does? Someone take it away. Yeah, maybe starting with the name itself. So Yatori is a Japanese word. It means a sense of mental spaciousness. So as an example, if I leave from home early to get somewhere early and I have time to take a breath, look around, that feeling is the feeling of Yatori. and the reason we adopted it as our company name is because we are looking to build AI agents, AI assistants that invoke the feeling of Yatori in our users.

21:03That's the product experience that we want to deliver. Okay, and how is it that you're going to do that? Yeah, so specifically what we're going after are AI agents that can reliably do everyday digital tasks on the web towards an AI chief of staff for everyone. So anything that we currently open a browser for in our day-to-day lives that can broadly be categorized as digital labor. So anything from like reservations, scheduling appointments, coordinating with people, having to constantly monitor websites for tracking news, or campground reservations, and so on. All of that is within scope of what we're going after.

21:39And keep going. I was just going to add that we're taking a fairly opinionated stance on how we get there. So the kinds of agents that we're building, they don't take over your device or your browser. They work in the background on the web. and unlike how a lot of the AI assistants today work, for example, ChatGPT, that's reactive and only respond when you ask it a question, we're trying to rethink that and ask the question of, okay, what if these agents were proactive and you didn't even have to ask? Okay. And so, David, that was sort of my next question for you, was that as it relates to AI assistants and attachments to browsers, for example, we know that OpenAI is working on a browser.

22:20Comet from Perplexity is a tool that has been getting a lot of traction. So Das is saying that you're not going with the browser. You're going with something in the background. Why is that better than a browser, in your opinion? I think for a few different reasons. One is the browser is in your device. And so anything that it's doing is happening on your device, sort of in the foreground, in your face, so to speak. whereas a lot of what we are going after is if these agents can just be in the background making progress on this task while you are living your life that's that's how sort of we get to delivering utility utility there's also some technical challenges that come up if you're in the browser for example if you want multiple agents to be doing multiple things in parallel for you that's just much harder to do if they're all sort of in your tab in in your browser if it's a long task that's going to take a while or if they're sort of always on persistent agents, if you shut your laptop and walk away, what happens to these agents if they're in your browser?

23:23Okay, so Das mentioned that reservations was one thing. What are the other types of things that you hope Yutori can do for us? So things like monitoring information that whenever something happens in the world, let me know that if there's news about X topic or if the price of something drops below threshold. If cam count reservations open up, let me know. And that's the first product that we launched a couple of months ago. In the future, not just monitoring information, but also taking action on it. So if the price drops below threshold, go ahead and buy it for me. Or if this becomes available, make that reservation.

23:59Taking actions would be a natural follow-up from there. And then getting closer and closer to an assistant, a chief of staff, that is just taking care of more and more of your digital chores. And Das, what is it that makes you Tori different? Because there are a ton of different companies working on exactly that use case in the background. I hear this example a lot on the show is price drops, you buy, you sell, right? Reservation opens, you book. Flight opens, it comes down to this threshold, you book. I mean, what makes you Tori different? Yeah, I think it absolutely is spot on that. There's basically a ton of agentic companies out there that are promising the world.

24:36And then when you try it the first time, at least the experience that I've had is many of them don't quite work. And it is very much top of mind for us to not be the 101st company that's making the same mistake. One thing that's different in how our team is organized is that we are sort of innovating in a full stack manner, like across the board from training our own models to do some of these tasks, as well as innovating on what these interfaces look like that make interacting. So you have your own models? We post-train our own models. We don't train them from scratch, but we post-train our own browser-used models.

25:10And Devi, I mean, we should say, both of you are former employees from Meta. You were both researchers there, from what I understand. So what was it that made you sort of believe that you couldn't build something like this within Meta that you wanted to do it on your own? So I think a couple of reasons. One is we are rethinking what the interface with the web looks like. And that requires a certain sort of just thinking from first principles, thinking from scratch, thinking from a clean slate, which is something that's just much harder to do in a larger company that has all of the baggage associated with it from its prior success.

25:54So that was one. The second was also just from personal motivations. All three founders, we've worked together for a decade. We've done some of our best work together. We've talked about starting a company together, and we've been all starting to feel saturated in our roles at Meta. And so that was sort of another motivation to go do something of our own. Abhishek, I keep wanting to call you by your first name. Abhishek, before the show, Abhishek told me he wanted to go by Das. So we're calling you Das. Das. One of the questions I had for you is Meta has gone through a lot of restructuring. We've written that they restructured their AI team for the fourth time in six months now.

26:33We obviously have Alexander Wang coming in. They're bringing in a lot of talent. What do you make of the restructuring that Meta is going through? And do you think you could do what you're doing now within the new structure at all? Yeah, good question. I will probably deflect this to Devi. Okay. Devi, we're coming to you. Yeah, yeah. So I think, yeah, meta has been, like, this is one of everyone's favorite topics right now, right? Everything that's going on in MSL, in the TBD lab. But meta has been investing in AI going back all the way at least to 2013. Meta has done several experiments of how it structures and organizes its AI work, both from a research standpoint, sort of exploratory bottom-up, curiosity-driven research, more focused research, in terms of how to bring this research to products, having a centralized AI org, these are decentralizing their talent to broader groups and so on.

27:28So this is another experiment. Good things have come out of the previous experiments. I'm excited to see what comes out of this one. It is not clear to me that we would be able to do this in the new structure. I think the new structure is still meant to sort of push on the foundational capabilities and sort of having a small group of people going after the whole stack of training our models, thinking through the product experiences, designing those and shipping it out every few months, I think my guess would be it's still hard to do something like that in that structure. And that's probably not what Meta is optimizing for, right?

28:01It's going after other things. Devi, what did you make of Meta's partnership with MidJourney? Yeah, to be honest, it was a little bit of mixed feelings because while I was at Meta, I had worked on a lot of image generation technology. I left last year. And so, yeah, it would have been nice for sort of in-house meta models to have continued living even after I left. That was interesting to see from the outside that it is now betting on external models to bring that in. But, you know, thinking about the work that you did at Meta and now thinking about sort of this new era of talent that has been ushered in, I mean, what do you think was the mindset there?