AI for Wall Street, Gemini IPO Signal, Paramount’s Bet on Warner Bros. Discovery | Sep 12, 2025

0:00Welcome, everyone, to the Informations TI TV. My name is Akash Pasricha. It is Friday, September 12th. And what an end to the week for news we have, folks. We have a lot to get to today. We're breaking down everything you need to know about the Microsoft Open AI peace deal that appears to be coming together. We're also going to get into the paramount bid for Warner Brothers Discovery and what that could mean for both of those companies. We're then going to bring on Jay Doss at Sapphire Ventures to help us understand what this crazy week in IPOs and crypto IPOs in particular means for the market more broadly.

0:34We're going to end with two conversations about how one startup is selling AI to Wall Street and why tech companies aren't buying social media platforms anymore. It is going to be a busy show, but a great one. Let's get right on into our first segment. Last night, Microsoft and OpenAI said that they have reached a tentative deal that will see OpenAI restructure its very complicated governance. This is part of a long-running saga that we have been covering for many months and years here at The Information, and yet it somehow never gets less complicated. I want to bring on our Microsoft reporter, Aaron Holmes, to help us make sense of all this.

1:12Aaron, welcome back to the show. It's great to have you. Hi, Akash. Happy to be here. Okay, so you must be exhausted. I mean, this is very hard to understand, and the incremental details themselves are very complicated. So help us understand what the news was, and then we'll sort of get into the backstory about how we got to this point? Yeah, so for now, there's still a lot of unanswered questions, but what we do know is that Microsoft and OpenAI said they've reached a tentative non-binding agreement about the next phase of their partnership. And reading between the lines, OpenAI is trying to restructure as a for-profit entity still owned and controlled by its nonprofit, which would allow it to raise more money and go public.

1:53But to do that, it needs Microsoft's permission. And as a result, the two companies have been basically locked in negotiations over the past year over what that's going to mean in terms of Microsoft's stake in the for-profit entity, as well as a whole host of other questions like how Microsoft will keep getting access to OpenAI's technology, especially the most advanced technology, which would be AGI, which is on par with a human intelligence when OpenAI eventually reaches that. Right. Okay, so we're going to get into some of those details. But the thing that really showed out to me is they said, tentative non-binding.

2:29I mean, that sounds about as walking on eggshells as anything. Why do you think they actually decided to put out this announcement if things are still getting hashed out? I mean, I think it's clear that the two companies want to signal that they're getting closer to a deal. It is in both of their interests to ultimately reach some sort of agreement. You know, OpenAI needs to keep raising money in order to keep powering its extremely cash-hungry business. And Microsoft needs OpenAI to be able to keep raising money so that it can keep spending so much on Microsoft's cloud servers, as well as, you know, keep developing cutting-edge AI that Microsoft has been using to power the bulk of its AI features over the last few years.

3:08So they might not see eye-to-eye on every tenant of their agreement, and I'm sure, you know, that there's still some last negotiations that they're working out, but this is them essentially saying to the world that they are getting close to a deal. Okay. And with this news that came out last night, I mean, is there a winner at all in this? I mean, it still remains to be seen. You know, we know that one of the top things that Microsoft has wanted is continued access to OpenAI's technology, even when the startup achieves AGI. Currently, their previous contract stipulates that Microsoft would lose access to OpenAI's technology at that point.

3:48So the exact kind of way that that has been hammered out has yet to be seen. There have been reports that the AGI clause is still in their new contract, but has been modified somewhat, but we don't know exactly what that means yet. And at the same time, of course, you know, Microsoft wants a sizable stake in the for-profit entity. Currently, it only has profit units that would entitle it to future OpenAI profits rather than a traditional stake. And one thing I was hoping you could explain to us is they talk about the exclusive access to OpenAI's technology. I mean, as far as I know, OpenAI's models are used very widely.

4:23So what exactly is exclusive about this and what are they negotiating there? It's a really good question. So right now, Microsoft has the exclusive rights to resell OpenAI's models in the cloud, meaning if you're a company that wants to use OpenAI's models in your products, you basically have to go to Microsoft Azure in order to get those models. and that has become a huge business for Microsoft. We've reported that that, you know, surpassed a billion dollars in revenue last year and has grown a lot since then. So that's one big piece of the exclusivity and Microsoft also gets the rights to reuse OpenAI's models without paying the startup, you know, per token like other companies do, which it has the exclusive rights to do.

5:06So it remains to be seen whether those things are going to continue beyond the current deal term, which is set to expire in 2030. Right. And last question for you, big picture here. I mean, why should we care about this deal at all? I mean, from my view, it doesn't feel like as a consumer, ChatGPT is going to change that much. But talk to me about the bigger picture here. I mean, is this weighing on other parts of the AI ecosystem? Yeah, absolutely. I mean, the individual consumer might not see a big difference when OpenAI is able to raise money and go public, but OpenAI's ability to keep spending heavily is arguably kind of holding up a big corner of the economy.

5:43I mean, it just signed a$300 billion deal with Oracle. It has committed to spend... it doesn't have. Right, that it needs to keep raising. And, you know, building out data centers and continuing to sort of fuel that demand is hugely important both to the tech world and, you know, the broader U.S. economy at this point. Great. Okay. Well, if those stakes aren't high enough for you, well, then I don't know what it is. Aaron Holmes, thank you for coming on. Aaron covers Microsoft here at The Information. We're going to keep the show going with our second segment here. It is stunning that this next story was the second biggest news that happened yesterday.

6:20Paramount Global is putting together a cash bid for Warner Brothers Discovery, according to the Wall Street Journal. Warner Brothers Discovery shares when shooting up on the news, it would cost Paramount something like$71 billion, which would include all the debt that Warner Brothers Discovery has on its balance sheet. This is a monumental deal in media, and given that the Ellison family is behind it, there is also a tech angle. And so I want to bring on our co-executive editor, Martin Peers, to talk all about this. Martin, happy Friday to you. I know that you are very excited to talk about this one because it is tech, it is media, it is billionaires.

6:55I mean, gosh, it's everything. Okay. So the thing that I want to ask you is, you know, when people have said AI is going to eat media, I mean, you know, media, so this is not what they meant. And And yet this is kind of central to the whole story in a way. Yeah, I don't think it's a case of AI eating media. I think it's a case of money being made on the AI front flowing through to media and rescuing the investors in Warner Brothers who have been holding on to this stock which has been in the toilet for years. And they're getting bought out. So, you know, the rich is buying out the poor, in a sense.

7:40So that's what makes this a great story. It kind of unites the two stories of AI, which is on the rise, and entertainment, which is on the client. So that's big picture what it is. It just granularly, I mean, explain to us tactically what you mean here. How is that money going from AI into media? Explain to us sort of the Oracle connection to all this. Sure. So Paramount is now controlled by the Ellison family, the same family that controls Oracle. The news this week that Oracle had signed these big new AI cloud deals, sent their stock soaring and enriched Larry Ellison. So he now got even more financial resources to invest into Paramount and to finance Paramount's purchase of Warner.

8:36Okay. So we have this news, you know, this deal could go through. We're not really sure, you know, where it is in the process yet. But my question for you is, is this a good deal? Is this a good idea? I mean, it could be a great deal for Ellison, depending on if it works out. I think it's a risk. It's a very risky deal. Warner and Paramount, I mean, he only bought Paramount a month ago. And both of these companies are very reliant on cable, which is a declining business. So, and also Warner has a huge amount of debt. So if he buys Warner and puts it together with Paramount, the combined company is very exposed to a declining business.

9:25I mean, you have to remember, anybody could have bought Warner, but no one wants to pay up for those declining cable channels. So, you know, there's a big risk here. But although one of the things that you've written about is the potential for Warner Brothers Discovery to split its business up. And one of the questions that we were talking about this morning is why not wait for that split to happen and then buy the part of the business that isn't declined? Yeah, you would think that Ellison would have waited. But the problem was that split isn't due to happen until next year. And then I think there's a tax reason that after that split, when he could have just bought the streaming operation and the studio, which will be carved off.

10:10But I think he has to wait two years for tax reasons. So if he waited, he'd be probably waiting until 2028 or so. Okay. And I think he obviously wants to sort of move faster. Okay. So now we've talked about sort of the risk here and the bear case, I guess you could say. What is the flip side of that? What could be the bull case for this? Yeah, well, the upside is that he manages to figure out, I mean, firstly, putting them together catapults him into a very strong position in Hollywood. I mean, Warner and Paramount together, the two studios are very well positioned. The streaming operations would also be very well positioned.

10:56They'd be able to compete more effectively with Netflix, Amazon, Disney. The only question is he's got the debt he'll be taking on and he's also got the declining profits. And the studio and the streaming don't make that much money. So can he improve the profits from those fast enough to offset the declines in the cable business? And that's the question, which is very hard to sort of... Right, right. Very quickly, before we let you go, but we also had the big Murdoch news this week that Lachlan Murdoch has control of the family empire. You didn't write about it this week in the briefing that much, I don't think at least.

11:41I wrote a small bit. Small bit. Give us the big bit. Give us the big bit. Well, look, the interesting point about that story, everyone has been writing that, oh, Lachlan won this long, tense battle. I would argue that the winners were really James Murdoch and Elizabeth than Prudence who are getting out. I mean, who wants to bet on Lachlan who now has to run Fox and News Corp on his own? That's not a bet that I would take. So I think actually they are the ones who are making the money. They're getting out, so. So you think it's, okay. Yeah, that's a good way to think about it. The way I was thinking about really is just, I mean, how much news we got this week with the landscape and media shifting once again.

12:31I mean, new owners, new conglomerates. Seems like it changes every couple of years. And so the story will continue to unfold. Martin, thank you for coming on the show. Martin is our co-executive editor and he is the author of our nightly newsletter, The Briefing, here at The Information. Okay, the Winklevoss Brothers crypto exchange Gemini is going public today. It is the third big tech IPO to happen this week. We also had Figure, another crypto company that debuted yesterday, and Klarna, which listed shares earlier this week. I want to bring on a close watcher of both the crypto and the IPO markets.

13:08Jay Doss is the president and co-founder of Sapphire Ventures. He is a longtime friend of the show. Jay, welcome back. It's great to have you. Yeah, Coach. Thanks for having me. It's great to be back. So we had a busy week this week in IPOs, which is not something that we've said in a while. looking back on these three tell us about your reflections and what we learned about the IPO market more broadly yeah so the first thing is that these are all mostly fintech IPOs of course black coffee was uh uh you know a chain but you know even then if you look at it uh they're very different companies Plana is actually a consumer lending company which is very different than than figure uh which is you know doing a lot of things with blockchain but I almost think that they are more of an enterprise company that is trying to redo how mortgages are sold and loans are given on a blockchain.

13:58And then we have Gemini, right? Which is kind of an exchange. You could also look at them as a digital asset treasury company. So, you know, it's going to be interesting to see what happens with the pricing for Gemini. Right. And I mean, we've talked about this a little bit on the show before, but is this the IPO window opening? Is something structurally changing about the IPO market in your mind here? Yeah, I do think that there is a lot of appetite for IPOs, not only just from institutional investors, but also a lot of retail investors, right? I do think that if you look at how the bankers are nowadays structuring the IPOs, I think I read somewhere that the Gemini IPO, 30 % of it is being sold to retail.

14:39Yeah, yeah, through Robinhood and some of these newer platforms. Exactly, which is very different, right? Typically, previously, IPOs, when you had, you would sell maybe 10 % to retail, if that. But I think what these companies and the bankers are doing is really giving this appetite to the retail investor. Because remember, a lot of the hedge funds, a lot of the crossover funds, they do have access to these companies, even when they're private, when they do these big tender offers and these big rounds. So this is really capturing the retail kind of appetite for IPOs. Right. And I mean, we're also at this place in the economy where rates are starting to come down.

15:21And there's also that factor there that, you know, that might be sort of opening up the markets a little more. Exactly. Right. People are looking for growth. And there's a lot of capital kind of sloshing around in the system. And some of that is also kind of pivoting back from, you know, more credit and debt-oriented instruments. because, as you said, the interest rates are coming down and people are looking for growth. But in general, I think the risk appetite has definitely changed. I know the economy is giving different signals depending on what you want to look at. But I think there are people out there, both retail and on the institutional side, really looking for growth stocks.

16:02And that is why the IPO market is doing so well. So now, over the last couple of years, we've talked about this bar that it's not just you, but everyone has said$500 million in ARR. That is kind of the bar for going public. I think that equates to something like a$5 billion market cap. Last time we had you on the show, when this is when Figma IPO, you basically made the point, I don't think the bar has really changed that much. With these structural changes happening in the economy, do you think that bar is coming down at all or are we still at 500? No, I think you like, look, even, you know, Gemini is selling 425 million at the kind of, you know, the IPO.

16:41But what has happened, I think, is, you know, if you look at the Gemini IPO, it's kind of interesting. They actually reduce the size of the offering. In some ways, you know, they basically decrease the supply so that, you know, they can get a really nice pop and have on the upside. But I do think that that is why since the way that these IPOs are being structured, I think a lot of institutional investors, even if they buy, if they get allocation, they're not going to buy in a huge way until there's more liquidity and float in the stock. Right. But what about, again, the size of the company, like 500 million ARR, does that threshold change for you?

17:19Yeah, I think that threshold is still there. Like you look at figure, that's like over 7 billion in market cap. I think Gemini is pricing at like three, three and a half billion. So I do think that, you know, maybe the bar, you know, is not at five billion. But also, this is the lifecycle of IPO. If you look back in history, typically at the beginning, you really get the tippy top companies, you know, the five billion, 10 billion mark is kind of what it's people looking for. But then over time, as there is more and more of these IPOs and more people want to buy them, companies go out with maybe not those kind of metrics and not as much float.

17:56So I can see that in the next 12, 18 months, there will be companies going out with much smaller market caps and the float will be like maybe$150 million. The question is what happens to those companies long-term? Do they end up in PE hell and get acquired or do they actually able to survive and execute and meet and beat the forecast and actually become a big company? Can you help me understand, you talked about Gemini's sort of people seeing it as sort of a digital asset treasury company. And that's kind of an interesting view because I was looking at the financials. I mean, I was comparing it to Coinbase, right?

18:34Coinbase, it's a profitable company. Revenue is growing at Coinbase. As Gemini revenue contracted, if you look at the first six months of the year this year versus the first six months last year, revenue's contracted and operating expenses have gone up. So I'm not really seeing what the case here is for the company. You're saying, hey, they have crypto on their balance sheet, just five stars. Look, there are companies like that trading out that strategy, which was previously called micro strategy, is a great example, right? Yeah. Basically holding Bitcoin, each dealer and a couple of other companies kind of doing that.

19:12So I don't know what the market, it's going to be interesting to see how it prices and what happens on the trading side, you know, because, you know, at some point, you know, if there's a lot of retail investment, it doesn't matter the fundamentals of the company. Just a meme stock at that point. Yeah, a little bit. I want to say complete meme stock. I know. Yeah, yeah, yeah. But, you know, loosely. Yeah, exactly. It kind of, you know, what people are talking about on Reddit, you know, all of these, you know, social network, that's kind of what drives the price of the stock and not the fundamentals.

19:47I think you find out the fundamentals over time in six to nine months when the lockup expires and the company has provided two or three kind of quarterly results. So yeah, it is very hard to sometimes analyze why a stock is kind of going up versus going down, right? And the thing about these treasury companies, I don't really understand that is, if the sort of competitive thesis for investing in this company is the fact that it's a treasury company, well, then really the treasury company, But it's hard to, you know, they're marginal, they're different businesses with the same cryptocurrencies on the balance sheet.

20:21Why do you buy one treasury company versus the next? It's like, I don't know. Exactly. I personally don't know. But look, I think that is the challenge with fintech stocks, right? I think fintech stocks are not like software stocks where, you know, software stocks, you have ARR and, you know, GRR and NDR. Those are kind of, you know, margin. Those are the key metrics. FinTech businesses are always complicated because you'll have some things on your balance sheet. You'll have some stuff, maybe you're selling directly to the consumers as a lending instrument. You might be so. So FinTech, I think stocks are much difficult to analyze.

20:58And for a lay person, of course, bankers and - Financial statements are very confusing. You really have to get an expert to read the cash flow statements because it's not the same. Exactly. Exactly. Exactly. But look, I think overall, I think it's good for the market that there is appetite for all of these IPO stocks and that people are getting liquidity, people are trading, because you want that in some ways, as I keep saying that it is good to be public companies because you have the regulation kind of overview, but it just makes you operate as a much, much better company while you're public versus staying private.

21:38Right. Great. Well, Jay, thank you so much for coming on the show. It was a busy week for IPOs, and I suspect that things might get a little bit busier, so we will have to have you on more to make sense of all of these initial public offerings. That is Jay Das, the co-founder at Sapphire Ventures. Okay. Well, we all know just how important Bloomberg terminals have become to Wall Street. At least one company is hoping it can use AI to take at least some of that market share away from the terminal. Alphasense was founded more than a decade ago, but AI has helped lift its valuation to$4 billion last year.

22:12I want to bring on the company's CEO, Jack Coco, to help us understand how Wall Street is thinking about this emerging technology. Jack, welcome to TI TV. It's great to have you. Thanks, Akash. Thanks for having me. So explain to us what exactly the platform is for starters. Well, Alphasense is the technology behind the most important decisions in the corporate world across from Wall Street to corporate boardrooms. Basically, an AI search and market intelligence platform that puts the right insights at the fingertips of decision makers at our clients. Companies like Google, Pfizer, JP Morgan. And what they're doing with it is really analyzing what they need to know as they're acquiring companies, making investments, launching new products, entering new markets, all the major value creation activities in business.

23:03where it really matters that you have the confidence from having the right data and insights at your fingertips and really can make a difference in your competitive edge against companies that don't have this kind of capability at their hands. So you've got, I mean, I've actually used the platform. You've got not just the financial data, but you've got the research reports, you've got the SEC filings, you've got news, you've got expert analysis. You kind of put it all into one platform, and now you're using AI to sort of make sense of all that data for people. It's certainly a very helpful tool.

23:36My question is, why can't a company like Bloomberg, which I think is probably the closest analog people know for something like this, why can't Bloomberg just build something like this with AI? Well, I may not be the right person to ask. I should go ask Michael Bloomberg. We've entered this space to come and innovate something new that wasn't being done before. we saw these traditional legacy data terminals providing access to a lot of content. I used these as an analyst back in the day in my investment banking days. And they were great at providing access to a lot of information, but what they didn't do is actually help you really find it.

24:13You had to manually go and look for it and spend a lot of time. And what we wanted to build is a capability, a semantic AI-driven search capability that helps you find all the relevant data points and now uses AI to really pull it together, do chain of thought reasoning and kind of tie that together in a really strong narrative so that users can have the right information at the fingertips. And we haven't seen anybody else be able to do that yet. Right. With the AI component. And then the other piece is actually, sorry, it's the content where, as you mentioned, And our expert transcript library is actually a major proprietary content set that we acquired a company last year called Tegus and have been really aggressively expanding that.

24:58And that provides a lot of proprietary insight that you can't find anywhere else. So even these sort of very large legacy data terminal companies do not have access to a lot of this intelligence that you get through AlphaSense. And so tell me about that expert research that you guys are doing. Do you have a team of people that is then going and doing primary research with experts? Or are you using AI agents to run these interviews? What does that research look like? Well, the majority of it is still us facilitating venture capitalists, private equity professionals, and hedge fund people doing interviews every day.

25:3620 ,000 calls per quarter with experts in the corporate world across from former executives, customers, partners, competitors of every one of those companies to really dig deep into what matters for those companies. And now what we've added is to your second question, we've added an AI interviewer actually, where we're also able to send that to interview all of these experts and have really smart conversations, a half an hour conversation with an expert on sophisticated technologies, and that's able to add even more skill to the library. So you're kind of betting that this data, the primary data that you're collecting, that is kind of the moat, I guess, because it's the quality of the data, essentially, that it's better for anyone using this platform.

26:23The quality of the data, the breadth of the insights, the unique insights, and of course, the technology that ties it all together. We've seen, heard from clients where, you know, private equity firm told us recently they'd spent a month doing research manually on a market segment where our deep research product gave that same, or even better in their words, report in 10 minutes. So that's the kind of thing that really differentiates us. I do want to ask you, your company is more than a decade old, and you've sort of been around for, I guess, call it the pre-ChatGPT era, and then now you're building in this AI era.