บทถอดความฉบับเต็มของ AI Trade Bounces Back as Investors Buy the Dip

IPO marketing and a roadshow and go public before Thanksgiving, they have to decide that in the next few days. 00:05:28 Speaker 5: This is a company that notably has not aligned itself with the Trump administration over the last few years. There was the dust up between the Department of Defense and Anthropic a few months ago. Is that looking like it could work to the company's advantage now that we're seeing a real change when it comes to Congress come 2027? 00:05:43 Speaker 6: I think a. 00:05:44 Speaker 7: Lot of the tech companies are going to have to kind of deal with being between a rock and a hard place right now. You're right that they've sort of had a better type of relationship with some Democrats. In fact, they've hired some former Biden administration officials. But Trump is still the president. This is still the regime that they operate under. And I think they've still received a fair amount of scrutiny. even from Democrats in recent months as they prepare to go public and as their models are sort of at the vanguard and having safety issues. 00:06:06 Speaker 5: Seth, we're talking about Anthropic and we have been, but where does all of this leave Sam Altman and OpenAI? 00:06:11 Speaker 7: Well, it's funny because now we're hearing these two companies are kind of potentially right around the same valuation as one another. And they're both trying to tap the markets for another key fundraiser, one in a private market, one in a public market. And I think they're kind of stretching, at this point, potentially the appetite for how much more capital they can get to meet their needs at a moment when the market might feel a bit more jittery. 00:06:29 Speaker 5: Yeah, it will be really interesting to see how both of these play out, especially the open AI side. 00:06:32 Speaker 1: Appreciate it. 00:06:33 Speaker 5: Seth, always good to see you. Check out Seth and his team's reporting, too, on the Bloomberg Terminal. It has been a bumpy few days for the AI trade and uncertainty still remains. Our next guest says, though, the fundamental demand for AI is still there. Michael Monahan is a portfolio manager at Founder ETFs. Michael, thanks so much for joining us. I want to start with Anthropic because you love founder-led companies. That's your entire thesis. You want to invest in companies that are still led by the people who founded them. Will you be buying into Anthropic's IPO? 00:07:04 Speaker 1: Thanks for having us, Tim. And that's right. We do invest in founders. 00:07:07 Speaker 8: We believe that they have an extra level of vision, execution and grit. And we will be buying into Anthropic day one. It meets our investment thesis. The fact that the company is the fastest growing company from a billion to 40 billion in the history, we think is very exciting. We think the revenue growth continues and we think that this anti-data center pushback, which is a luxury belief, we think when the fact that these recently elected politicians need to create jobs for the constituents, we think that there'll be a softening of this anti-data center posture. 00:07:40 Speaker 1: Okay. 00:07:41 Speaker 5: You call it a luxury belief. Explain what you mean by that. 00:07:45 Speaker 1: It means that. 00:07:47 Speaker 8: It's a belief that you can have when you don't need to put food on the table, when you don't need to pay the mortgage on your home for three or four kids. But if you're someone who has to pay a mortgage, take care of your children, it's a belief that goes against the wages that you're going to earn as an electrician or a sheetrock hanger or a coal miner to fire a facility to create electricity. 00:08:09 Speaker 5: I think the challenge, though, Michael, and perhaps we'll see more data in the coming days as the dust settles from this election, is that when it comes to the cost of living, the people who you're talking about being able to afford putting food on the table, they do associate data centers with increasing their own cost of living from the perspective of electricity, for example. They've seen their bills go up and they hear about these data centers that are so energy hungry. 00:08:33 Speaker 1: How do you square that? 00:08:35 Speaker 8: Yeah, I think that's been rhetoric that's been pushed by politicians, but I'm not sure that the data supports that. Most every one of these data centers either brings their own power or pays to increase the capacity on the system they're tying into. So, I think the data shows, if anything, they're lowering the power costs in the areas that they're going into. 00:08:58 Speaker 5: Given that you think this is a luxury belief and that you think these politicians who've spoken out against data centers in some of these states, such as Texas, Ohio, New York, when they do actually have to create jobs for their constituents, you think they'll ease up a little bit. Does that mean that you think the sell-off as a result of this is overdone? 00:09:19 Speaker 8: That's our viewpoint. We think this sell-off was very overdone. We saw the relief rally on Wednesday, markets like certainty. So once the certainty was put in place, we saw a relief rally. We got some selling going yesterday morning. I know Congressman Ansari put out this tweet about how she was going to block all data centers. But I think that now that we've worked through that, you've already seen a softening here in Texas. from the governor, Tirico has softened his position. And I think it's the reality that what Americans really want, they want to take care of their family, they want to earn a good living, and I think that progress is going to go forward to help support the voters that just want to take care of their families. And one of the ways they can do that is to lean into this American innovation. And there's also a patriotic element to it. We need to maintain super intelligence dominance to not fall behind another country. 00:10:17 Speaker 5: What's the risk in your view if the U.S.

does fall behind? 00:10:20 Speaker 8: I think certainly if we don't invest in these data centers, we don't let the magic of capitalism work, will fall behind the Chinese and some of the others. And I think that the American middle class has been a great beneficiary of the US being a dominant superpower for at least since World War II and maybe. 00:10:40 Speaker 1: A bit before that. So I think we want to continue to maintain that position. 00:10:44 Speaker 8: One of the ways that we can do that is maintain our dominant position in super intelligence. 00:10:49 Speaker 5: So were you buying yesterday? 00:10:52 Speaker 1: So we were a buyer yesterday. 00:10:55 Speaker 8: Certainly it was a great opportunity to own some of these generational companies. 00:10:59 Speaker 1: Like what? 00:11:02 Speaker 8: So, you know, SpaceX we think is a great opportunity. Nvidia was a great opportunity yesterday. And I also bought more of the fund personally for myself. 00:11:13 Speaker 5: The fund that follows founder-led companies. So we could, you said you've already, you already will be buying into Anthropic on day one. I want to go back to where we started. As we just heard from Seth, the concerns about the valuation in recent days from this company, or not from this company, from investors, I should say. Now thinking $ 1. 2 trillion to $ 1. 5 trillion, down from $ 1. 5 trillion to $ 2 trillion. 00:11:38 Speaker 6: Is $ 1. 00:11:38 Speaker 5: 2 trillion where you'd want to buy in, in terms of a valuation? Do you think that's a better value for you because it is less expensive and gives more of an opportunity for a runway and growth? 00:11:51 Speaker 1: Yes, we believe that. 00:11:52 Speaker 8: We were excited to buy the company at $ 1. 5 trillion. And I can tell you, we're much more excited to have the opportunity to buy it at $ 1. 00:11:59 Speaker 1: 2 trillion. 00:12:00 Speaker 8: So we were favorable at $ 1. 5 trillion. And I can tell you at $ 1. 2 trillion, we'll be working as hard as we can with Morgan Stanley and Goldman Sachs to get as much as we can on the opening print. 00:12:12 Speaker 5: I think a lot of people are concerned that this company grew so quickly in value in the private markets that there's not much opportunity left for growth as a public company. You know, double, triple, quadruple your money from an IPO. What do you say to them? 00:12:27 Speaker 8: You know, if you look historically, most of these great companies generate most of their value post-IPO. I know there's some concern in the recent term about how long some of these have waited, but the growth we think still remains. 00:12:42 Speaker 1: There's nothing that prevents these to be 5x, 10x from here. 00:12:46 Speaker 8: Looking at the adoption of how fast these companies are growing and how fast they're being pulled into workflows, one of the examples I like to give on where growth could continue is The top 100 law firms build $ 150. 00:13:02 Speaker 1: Billion a year. 00:13:03 Speaker 8: We think superintelligence is going to go after a big piece of that, and that's just one industry. So we think there's plenty of industries to keep stacking up to generate additional growth for companies like Anthropic. 00:13:15 Speaker 5: Michael Monahan, Portfolio Manager at Founder ETFs. Mike, always great to speak with you. Appreciate your time. Well, coming up, systemic funds cutting tech exposure, stoking a trillion-dollar tech stock route. 00:13:28 Speaker 6: We're going to have more on that next. 00:13:29 Speaker 5: This is Bloomberg Tech. Systematic funds slashing tech exposure, deepening the NASDAQ 100's plunge and wiping out $ 2. 3 trillion in market value as traders at Goldman Sachs say their phones are, quote, exploding with asks on how far the drop can go. Bloomberg's Natalia Kenijevic joins us now. Natalia, as the dust settles from yesterday's sell-off and we see a little bit of dip buying today, what are we learning about the technicals of yesterday's sell-off? 00:14:11 Speaker 4: Yeah, so it was driven by lots of technical factors. So, for example, we know that there is a whole universe of the so-called systematic funds. They don't care about fundamentals. They follow the stock market direction. So whenever we see a big sell-off, The index hits a certain threshold, so they have to sell. They don't manage that much money, but it is important because when they sell, they do it very quickly, and the market really feels that. So that is why we saw us selling across CTAs, wall controls, and it got like so good. 00:14:44 Speaker 5: Okay, so how exactly do these funds work and adjust their positions on a day like yesterday or a week like this week? 00:14:50 Speaker 4: Yeah, so some funds like CDAs, they just follow specific thresholds, whether it's like a big move to the upside or downside, so they have to sell positions. Then we have a big universe, it's called vol control funds, volatility control. So when volatility spikes, they have to get rid of their exposure to, for example, US stock futures, and go to cash. We saw a big spike in volatility index yesterday, the biggest move since the COVID-19, so they also had to adjust positions. So for example, some data points, we saw that world control funds reduced exposure by about 30 percentage points, and Barclays, Alex Altman is saying that Sell-off can continue for the next 24 to 48 hours. So it means that, yes, we do see a rebound today, but they can't adjust positions that quickly. So there is some kind of selling pressure potentially happening today. 00:15:42 Speaker 5: Hey, just 30 seconds left with you. On some of the biggest Wall Street trading desks, some of whom you've spoken to in the past 24 hours, they urge clients to buy stocks before the elections. What have we heard from them since? 00:15:53 Speaker 4: Yeah, we have not heard that much, but we know that there was a lot of anxiety in markets and only one big trading desk from Barclays was really careful. And Alex

Altman was saying, you know, these elections will not resolve fundamental questions about the AI trade. So most of them were like super bullish, but we saw a move yesterday. 00:16:13 Speaker 5: Everybody check out Natalia Kinijevic's story. We barely scratched the surface here. That's Natalia Kinijevic from Bloomberg News. Thanks, Natalia. Well, coming up, Lyft shares dropping after posting a surprise loss in its third quarter. Shares down about 5.5% right now. David Risher, the company's CEO, joins us next. This is Bloomberg Tech. 00:16:33 Speaker 9: Bloomberg. 00:16:41 Speaker 6: Thank you. 00:16:45 Speaker 5: Record rides, but a surprise loss for Lyft. The company reported its third quarter earnings yesterday after the bell. Shares today are falling. The company missed revenue expectations. It posted a quarterly loss. Shares down just a little over 5% right now. Joining us now is Lyft CEO David Risher. David, thanks so much for joining us on Bloomberg Tech. I want to talk potential M & A because you guys have been on this buying spree over the last couple of years. The European taxi app, FreeNow, TBR Global Chauffeuring, Get UK, you've been integrating them into the platform, some bike businesses too. What is your appetite for further M & A over the next year or two? 00:17:21 Speaker 10: So we're in a nice position right now, and here's why. We have a very strong core business. Bookings are up 17% year-on-year, as you just mentioned, active riders. It's sort of, I don't know how many quarters of all-time active high riders we've had, including, by the way, even on Halloween a couple days ago, again, new all-time highs. So that's a very good core business. It produces about $ 1. 1 billion of cash, trailing 12 months. It's in that five quarters in a row. So that unlocks opportunities for growth. And as you said, we've purchased a couple of European assets because we were primarily a U.S. 00:17:50 Speaker 1: Company. 00:17:51 Speaker 10: Now we're a much more global company. And so today, for example, a year after that acquisition... If you're a Lyft rider, you can open up your phone in London or Paris or Barcelona or Amsterdam and you can get a Lyft ride. So that's an amazing place to be. I don't want to speculate so much about future acquisition. We've got a lot on our plate right now just to integrate the current ones we've already got. But it's nice to be in a position where we can... You know, deliver a great service for our riders every single day in North America and then use that to sort. 00:18:17 Speaker 1: Of expand overseas. 00:18:18 Speaker 5: So where are you on integrating the recently acquired businesses? How is that going? Where are you in fully integrating them and how do you see that affecting headcount? 00:18:28 Speaker 8: Yeah. 00:18:28 Speaker 10: So I'd say it's sort of still early days in integration. So anyone who's run a company before knows that you buy a company and then the real work starts. You think it's all about the deal, but it's not about the deal. It's about how do you bring the two teams together? How do you bring the cultures together? And really importantly, how do you bring the technology platform together? So I think the biggest milestone we've just reached, as I say, is now across 120 cities. You can now use the Lyft app. But still, FreeNow, which is the brand that we acquired and the company we acquired, that brand still exists. So we have to bring that under the Lyft sort of name across all of Europe. There's still more people integration work to do. But here's the last thing I'll say, and this is maybe kind of interesting in an AI world. These projects, integration projects, they're quite complex to bring two companies' platforms together. We actually achieved that ahead of schedule. And the reason we achieved it ahead of schedule, or at least the first couple milestones, was largely because of AI. So we're seeing some productivity there, which is kind of fun. And any CEO and certainly CFO likes to see productivity based on all this. 00:19:31 Speaker 5: So on the core driving business, how are you doing at attracting drivers to the service right now? Because I take a lot of Lyfts, I take a lot of Ubers, I talk to the drivers a lot. Their common refrain is that the companies are still taking too much of the share of the fare. 00:19:47 Speaker 10: Yeah, so our driver supply and our driver hours are at an all-time high, an all-time high. And this, again, is as recent as a couple of days ago for Halloween. We tracked this very, very closely because it was such a big night. Now, we, about a year ago, put in place something called the Lyft fee cap because I hear this feedback as well. In fact, I'm a driver on the Lyft platform, so I feel it on the other side too. We've said, and this is sort of worked into the arrangement we have with drivers, We will never take more than 30% of the fee that a rider pays after insurance is taken out. And the average is 15%. And literally, we rebate to drivers every single week. So I think the real answer is as long as we do a good job by our drivers, they'll stick with us. And the proof is in the pudding, though. Record all-time high riders or drivers. I will say, I'm going to brag for a second. We have about a 50-point preference gap versus drivers who drive for the other guys. I think you mentioned their name a couple minutes ago. So that's kind of cool, too. 00:20:41 Speaker 6: I'm pretty proud of that. 00:20:42 Speaker 1: All right. 00:20:42 Speaker 5: I think we know the other guys you're talking about. Let's go from cars to bikes, something that's certainly close to me here in New York as a city bike user. E-bikes, one of the most popular parts of city bike. But you keep raising the price. to use them. This year, e-bike went to 27 cents a minute for members, 41 cents for non-members. Talk to me about pricing power. How much do you have here before riders start saying an e-bike trip just. 00:21:03 Speaker 6:

Isn't worth it? 00:21:05 Speaker 10: So it's a great question. So first, let's sort of scale this thing. 00:21:08 Speaker 1: I mean, you're right. 00:21:09 Speaker 10: We do in the summer 200,000 rides a day in New York City alone. And we've reached, I think it's been five months of active, of all-time high rides here in San Francisco on our bike system here as well. So clearly there's a lot of demand for the product. When it comes to price, it's actually quite carefully regulated. In New York City, you know, we have a contract with New York City to provide the bike service there. And we're only allowed to raise it little bits and pieces. It's still a really good value. We do everything we can to keep the prices affordable as possible. I was actually just here with Daniel Lurie, the mayor of New York City, of San Francisco a couple weeks ago, on stage with him talking about affordability and how important the bike is as an affordable option. I hear you fares do go up a tiny little bit, but still, if you compare it to most other options, it's still a pretty great deal for riders. 00:21:52 Speaker 5: Just very briefly, in the last few years, there was concern that, and there was even talk of you guys actually divesting from the bike business. Since then, it seems like your strategy has changed. Is it safe to say that right now City Bike is not for sale and will remain part of Lyft? 00:22:07 Speaker 10: It is now for sale, and it is a great, great product for us. Every time I'm in New York, I use more city bikes probably than I use Lyfts or Subways just because it's so fast. 00:22:13 Speaker 6: It feels great. 00:22:14 Speaker 10: You feel like you're Superman on the bike, kind of jamming up and down the avenues. Look, it's a really important part of a lot of people's lives. And maybe I'll just zoom out and just say we don't control the economy, of course, but we have to recognize that you have some people who've got all the money in the world. 00:22:29 Speaker 6: It's crazy, right? 00:22:30 Speaker 10: And so we've got great products for them. Some people who want to save money, get on a bike and have the time of your life. 00:22:35 Speaker 5: David Risher, CEO of Lyft, joining us from San Francisco. 00:22:38 Speaker 6: David, good to see you. 00:22:39 Speaker 5: Thanks so much. Well, coming up, hedge funds reeling after a painful October. We're going to discuss that next right here on Bloomberg Tech. Welcome back to Bloomberg Tech. Let's get a check on these markets with Bloomberg's Isabel Lee. Isabel, what are you checking out? 00:23:07 Speaker 9: Hi, Ed. 00:23:07 Speaker 11: We're seeing a reversal of a reversal today because on Wednesday we saw this massive rally and on Thursday we saw an even more massive sell-off. But today, Friday, you're seeing U.S. stocks kind of stabilizing thanks to the rebound in chipmakers. You're seeing the S & P there higher by nearly 1%. The Nasdaq also up 1%. by nearly two percent and the philadelphia semiconductor

index higher by more than four percent there's a lot of interesting cross currents here earlier this morning we got data that that implied that the fed could raise rates in december and that's another story but for now at a tim i think it's worth repeating that yesterday was such a massive sell-off we saw nearly four trillion dollars wiped out in the entire u.s ecosystem i mean it was really a mix of conviction driven selling It was a mix of systematic unwinding and even hedge funds really dumping their AI bets. Goldman Sachs even said that the sell-off has pushed markets into this really massive extreme panic level, but the silver lining is that historically has preceded a strong rebound. And if you look at some of the big names, they're also green on your screen, Alphabet, Microsoft, and Amazon. all higher by at least 1%. So the irony of this all is on Wednesday, they rebounded because they thought the tighter regulation could maybe control the seemingly out of control AI ecosystem. On Thursday, it was that very prospect that caused the markets to really back down. But on Friday, again, it's the reversal of that. So it remains to be seen, Tim, whether this is the start of the recovery or whether this is just a temporary reprieve. 00:24:27 Speaker 6: Back to you, Tim. 00:24:28 Speaker 5: Bloomberg's Isabelle Lee. I mean, Isabelle, I get it. We both wear glasses, okay? 00:24:32 Speaker 6: I get it. It's all good. 00:24:32 Speaker 11: Yes, and you're both so smart and so fantastic. 00:24:35 Speaker 1: Oh, okay. 00:24:36 Speaker 5: I guess we have to go. Isabelle Lee on the markets. Another industry in a vulnerable position, hedge funds, a tech sell-off, and various geopolitical events leading to a painful October. Firms such as Walleye Capital and Kirkuswald Asset Management recording some of their worst months ever. Bloomberg's Hema Parmar joins us with more. Okay, so this is irrespective of yesterday. This is not taken into account yesterday. Walk us through the pain that we saw in October, and then connect it to what we saw yesterday. 00:25:05 Speaker 9: Yeah, so it's been an incredibly difficult and trying period for all these hedge fund managers, right? You saw pain in October, and then you saw some of the fallout from that. in the past couple days. So you are seeing some of the worst numbers that these funds have seen in their history. Kirkus Wald had its worst month ever. You're seeing funds down 9%, 12%, 16% in one month alone in October, and that is before the pain that we saw from this week. Now this week what we have seen is some funds then firing teams, selling off a lot of their equity positions, focusing on other areas, like Sandlin now focusing on financials, and basically firing the staff and selling off all of its other equity positions. Because of the pain, you're seeing this tumultuous domino effect of all of these firms just seeing the prices fall, needing to fix their books. 00:26:01 Speaker 5: I'm confused because these are called hedge funds, right? They're supposed to hedge so they don't experience the same pain that other owners feel in times of market turmoil. 00:26:14 Speaker 6: Exactly. 00:26:15 Speaker 9: And so the problem that we're seeing here is correlation. We're seeing crowded trades. Structurally, a lot of these firms have similar risk parameters. At a multi-strat, for example, when you've lost money, once you hit a certain threshold of losses, you then are put in a position where you need to exit. The firm might have to then kill some of your book or put you on hold. And so what happens is you see big firms selling off these shares. in a big way, forcing down the stock price, more selling, and you see this really rippling through the space, and what this calls attention to is the leverage we see in the industry, is all the crowded trades, is all the firms kind of doing the same thing in a big way, and the pain being so widespread that it's taking down a lot of firms. 00:26:58 Speaker 5: Okay, very briefly, you mentioned Multistrat, stock pickers, how are they doing? 00:27:01 Speaker 9: So not great. Typically, we expect them to produce at least a 1% gain a month. Right now, you're seeing losses of 2%, 3% in a month, which is not good for them. And the equity funds also struggling significantly, too. 00:27:14 Speaker 5: So unless things change this month, November is going to be another nasty month for these companies? 00:27:18 Speaker 9: Beginning of the month so we will there still is three four weeks to see how things play out right now it's looking like the pain is pretty bad at least in the first beginning of this month of November things can change fast but you know a lot of these firms now will have to dig themselves out of a pretty big hole that they've made in October and and then the additional hold that they're probably seeing this week too. So we're going to be watching very closely to see how this plays out through the rest of the month. It's not looking great. 00:27:44 Speaker 5: And we'll be watching your reporting closely. That's Hema Parmar. She covers hedge funds over at Bloomberg News. Well, shifting gears a little bit, Tesla will pare back production of its Cybertruck next year. That's according to CEO Elon Musk. It suggests that the slow-selling model could actually be phased out. 00:28:01 Speaker 6: This is the company. 00:28:02 Speaker 5: Reorients around driverless technology and AI technology. In a post on YesX, Musk said output will wind down at the company's Texas plant in mid-2027 to free up capacity for the CyberCab autonomous vehicle. Well, coming up, the blue wave might be giving some AI companies the blues. Many of the top tech names are bracing for new challenges after the Democratic sweep in the midterm elections. We're going to dive deeper into AI policy with the former chief innovation officer at the Federal Reserve when we return. 00:28:32 Speaker 6: That's next. 00:28:33 Speaker 5: This is Bloomberg Tech. House

Republicans are requesting a briefing from NVIDIA CEO Jensen Huang on his dealings with an executive from Neocloud provider Megaspeed that's under investigation for spying on the U.S. on behalf of China. Joining us now with more details on the story is Bloomberg's Maggie Eastland out in Washington, D.C. 00:29:11 Speaker 6: Maggie, what's going on here? 00:29:14 Speaker 12: Yeah, let me break it down. So three Republican lawmakers are asking Jensen Huang of NVIDIA to come in for a briefing. specifically on his interactions with one executive at Megaspeed. That's a woman that the New York Times reported last week that U.S. officials are actively investigating as a Chinese spy. So lawmakers are clearly not satisfied with what NVIDIA has said so far, which is that all of its interactions with this company and Wang's interactions with NVIDIA with this executive were normal. And they're looking for more information. 00:29:44 Speaker 5: You know, this comes in the context of Democrats taking control of the Senate and the House this week. We're going to talk more about that in just a few minutes. They obviously have not taken power yet. But does the pressure, and this is House Republicans who are bringing this, does this up the pressure on executives like Jensen Huang and what he could face, a preview of what he could face come January. 00:30:06 Speaker 1: Absolutely. 00:30:07 Speaker 12: I think that a lot of tech executives are now preparing for this, and NVIDIA has actually shown some signs that it might be seeing the writing on the wall as well. Just today, one of my colleagues broke the news that in Southeast Asia, NVIDIA is asking that any customers there that want to get its cloud stamp of approval to be an official NVIDIA partner, have to actually enable something called fleet intelligence, which would allow Nvidia to track the chips, which is, of course, could be useful if they're trying to pay attention to potential smuggling. So that is one sign already that Nvidia is taking this seriously. And, you know, Democrats in January are set to up the scrutiny even more. 00:30:44 Speaker 5: What could be the ultimate implications of at least this potential spy probe here? 00:30:51 Speaker 1: I think it's hard to say at the moment. 00:30:53 Speaker 12: There could be official legal action, but we have yet to see that against NVIDIA yet. 00:30:59 Speaker 5: All right, Bloomberg's Maggie Eastland down in Washington, D.C. Maggie, thanks so much for joining us on Bloomberg Tech today. I want to dive deeper into the AI policy discussion. Joining us now is Sunaina Tuteja, former chief innovation officer for the Federal Reserve System and current senior fellow at the Atlantic Council's Geoeconomic Center. We've all been trying to figure out this week, and I think we're not the only ones. Investors are trying to figure out this week. What's actually happening when it comes to the data center build out over the next couple of years now that there has been this

big shift across the country toward Democrats, some of whom ran on tickets opposed to data centers? Is the market getting this right or are they over are they overcompensating here? 00:31:39 Speaker 6: Yeah, thanks, Tim. 00:31:40 Speaker 3: So the election certainly proved that AI is now a kitchen table issue for Americans, and that's going to continue into 2028. So you thought you were done talking about elections. 00:31:52 Speaker 6: You're not. 00:31:52 Speaker 3: Now we're right into 2028. But what's still TBD is how much of this election rhetoric is going to translate into actionable policy. You know, I've been talking to a lot of the candidates, you know, some that have been reelected, some that are coming into it new. It's clear that, you know, on the surface, AI was an important campaign issue for all of them. But then you get into the second and the third and the fourth derivative question, and you start to realize it's kind of like a Rorschach test. Yes, they all were talking about AI, but they were all talking about it from a very different vantage. 00:32:30 Speaker 6: Point, right? 00:32:31 Speaker 1: Right. 00:32:32 Speaker 5: Some talking about data centers, others talking about control, safeguards, guardrails, and the like. 00:32:38 Speaker 3: So that gets me thinking, sure, there is a coalition around the concern of AI, but as they all come into Congress, is there even a coalition around what the problem statement is, let alone the solution that they work toward, right? So what I'm watching for from a policy-making perspective is really three tiers. So there's a federal, there's a Congress, but then there's the states, and then what are the regulatory agencies going to do, right? So from a federal and a Congress perspective, I think it would behoove them to really look at can we put singles on the board, especially, you know, the White House is still a player in this. So some of the things that the White House had already been talking about, you know, whistleblower protection, independent evals, you think about, you know, testing on frontier models, If Congress can unify on that, get the White House on board, build momentum on that. 00:33:30 Speaker 5: I want to go back to data centers for just a second because they're such a big topic this week and yesterday that was a big part of the sell-off. We had an interesting conversation with Michael Monahan over at Founders 100 ETF earlier and he said, these politicians who ran on this and are pushing back on it are going to see the writing on the wall soon when they need to deliver for their constituents when it comes to affordability. You worked at the Fed. For us, tell us what the economic impact of a data center is on a local community. 00:33:58 Speaker 1: Is it positive? 00:33:59 Speaker 5: Does it bring economic growth? What do the data tell you? 00:34:02 Speaker 3: I think it's still being worked out. I don't know that everybody knows the full picture just yet. I think on the data

center front, a lot of that work is actually going to happen at the state level. And if you look at it, 10 of the largest states, except for Florida, now have a Democratic governor And those governors aren't going to sit it out. So I think on a data center policy making, I would actually look to the states and the work that's going to happen there. Because a lot of permitting on the data centers is a state level discussion, the electricity, the power. And to your point about the data and the metrics to watch, again, less on the federal side and more on the state and the local side is what I would be watching for. And then that also brings us to the Regulatory agencies, there's been a lot of discussion and debate around how to regulate at the frontier model and lab side, but when you start applying AI in financial services or healthcare, a highly regulated spaces, those regulators are going to have a say anyways. 00:35:05 Speaker 5: What are the federal regulatory agencies, the dance that they're going to have to do, the needle they're going to have to thread? Pick whatever metaphor you want compared with what's happening in China. How do you do that in a way? And what's your message to regulators to do this in a way where they don't lose out to China? 00:35:22 Speaker 3: Yeah, I mean, the analogy I keep thinking about is a little bit of the aircraft carrier and speedboats. I think some of these issues are more longer term and the harder, gnarlier questions, and that will require more of a sober, slower approach, and that's okay. But then some of the things, like you think about agentic commerce, you already have a lot of financial institutions trying to grapple with the notion of if Tim is going to proxy his agent to start accessing his bank account, where's the liability? 00:35:49 Speaker 6: Not yet. 00:35:50 Speaker 5: I'll just tell you that. 00:35:51 Speaker 3: I think a lot of bank CEOs are going to tell you not to do it, right? But those are not issues that you can punt for later. Those are discussions that are being had now. So I look at those are speedboat issues that need to be tackled now. So I think... And again, I don't... I think it's a false debate to say if you start to tackle some of those questions, you're going to lose to China from a competitive perspective. I would also say... Listen, I come from an innovation world and you want to be pushing the boundaries, but at the same time, there's a misconception that regulation is anathema to innovation. 00:36:24 Speaker 5: Is there a misconception in your view that this technology could destroy all of us? 00:36:29 Speaker 3: I'm not a doomer. I'm a stubborn optimist. I think we can strike the balance. 00:36:33 Speaker 5: How do we do that? 00:36:35 Speaker 3: I think by setting up the right guardrails. We've proven that we can do that with previous technologies. 00:36:40 Speaker 2: And I mean. 00:36:40 Speaker 5: Well, okay, very briefly, the only one that comes to mind for me is nuclear weapons. What's another technology that I'm not thinking of? 00:36:46 Speaker 3: But I think even with that, we've got smart people who are working on it, and we've got smart people who are talking about that, hey, this is not all, that there are dangers around it, and we've got to be sober around it. You know, we were talking with a few of the builders around AI and the proliferation of fraud. And one of the things in talking to a lot of builders that I heard was, hey, the fact that we're talking about that fraud can be extended due to AI is something new that in the past with technologies we hadn't. There's more vigilance around it. And the builders are now building capabilities with AI to combat that fraud. Will the preventative measures be at the same levels And we'll, you know, at the same level, at the same pace? No, but we're going to play catch up and we will meet the moment. 00:37:35 Speaker 5: Sunaina, this has been a great chat. Appreciate you joining us on Bloomberg Tech. 00:37:38 Speaker 3: Thanks for having me. 00:37:40 Speaker 5: Sunaina joining us here. Sunaina Tuteja joining us here. Coming up, Jason Robbins, CEO of DraftKings, joins us on the tail of a disappointing earnings report. This is Bloomberg Tech. Cheers to DraftKings plunging on a 2027 earnings forecast that missed estimates. Investors also not happy with third quarter revenue and a cut outlook for the fiscal year 2026. Shares down about 26% right now. We're joined by the CEO of DraftKings, Jason Robbins. 00:38:18 Speaker 6: Jason, how are you doing? 00:38:19 Speaker 5: To me, this was all about prediction markets, not in the sense of competition, but in the sense of stepped up investment in what you're doing with prediction markets. You said $ 650 to $ 700 million as a result of this. Why are prediction markets so expensive for you to build out? 00:38:35 Speaker 6: Well, it's not the build-out. 00:38:37 Speaker 13: We actually built out, I think, what is now the best-in-class sports offering for much less. Most of that was marketing investment to acquire customers. We had unbelievable customer acquisition paybacks with the cohort we acquired so far this year. So we leaned in, and this was one where we messaged that we were going to follow the data and spend more. I think it was a lot more than people thought. There was some sticker shock. I do think if you look at our earnings materials, we disclosed a lot of data on the paybacks and what the returns on those investments are, and they look extremely attractive. So I think once that dust settles and we start to be able to show that, that people will see it was a good investment, but there was certainly a reaction to the short term, and I understand that. But I think this will prove to be the right move for us over the next year or two. 00:39:28 Speaker 5: Just a few months ago, you expected to invest only $ 200 to $ 300 million in predictions this year. Do you think some of the sell-off is because investors

might not have the confidence that it sticks to $ 700 million and doesn't become $ 900 million? 00:39:43 Speaker 6: I don't know about that. 00:39:44 Speaker 13: I mean, I think the fact that it was much higher than previous, if you look at what that means for earnings this year, it reduced our earnings. And that's what people are upset about. I hope people don't think we're going to spend more. That's already quite a bit of investment for this year. And we did also note that we are getting a great return on that. Of that money we're spending, the customers we acquired, not even the ones that we are going to acquire the rest of the year, just the ones we acquired to date, are going to generate $ 350 million of gross profit next year. So you can see why we're excited about the paybacks on this and feel very good about that investment. And this really is very similar to how we've always managed the business. There was a period where we were loss-making in 2022 and 2023, in the early 2023, where we were investing into things and people were a little skeptical of it. And now they're like, okay, I get it. Those cohorts look really good and are producing real cash flow in the core business, the non-predictions part of the business. So I think this one's a little unique because you have the regulatory uncertainty. So I think that's part of what makes investors skittish about it. But I think, you know, what made us feel some conviction that this was the right thing is that even in a regulatory scenario where sports contracts go away, we still will be able to pay back these cohorts in roughly four years, which is a bit longer than our typical three-year target. But still, you know, something I think that's acceptable given, you know, the paybacks are much, much better in the event that sports contracts do stick around. 00:41:14 Speaker 5: Do you think that sports contracts will go away? I mean, it looks like this is going to go to the Supreme Court. What chance do you think it has in surviving? 00:41:21 Speaker 6: Oh, I can't predict that. I mean, there's, you know, anybody who tells you they can't. 00:41:26 Speaker 5: But you're spending a lot of money on something that might not be a legal business, right? 00:41:31 Speaker 13: Well, yes, but we also have cross-sell into other non-sports contracts as well as other products we have like Lottery and Pick Six. So our 26th cohort is going to pay back before what the projected Supreme Court decision date is. So that investment we definitely feel good about. The 27th cohort we think will not entirely pay back by then. Obviously, the court date could get pushed out. but it will end up paying back just a couple years later. So we also feel like that's the right decision. But coming back to what you're saying about the uncertainty, here's how I would think about it. If you look at the landscape, whether it's the traditional prediction market operators like Poly Markets and CalSheets of the world, or whether it's our traditional competitor set and the legal regulated gaming market, the Fan Duels of the world, I think if you look at all of that competitive set, We are the only company that has positioned ourselves where we are in a great spot either way regardless of what the court outcome is. They are really counting on it. If you're a traditional prediction market operator, you're counting on that going your way. Otherwise, the whole business is in shambles. And if you're one of our traditional competitors who hasn't been able to make the progress that we've made on the predictions front, you're really counting on it getting shut down. Otherwise, you are way behind in something that now is a real part of the industry. So I think that's strategically how we've tried to position ourselves. And I don't think investors quite get that yet, but I think we should get credit for that. 00:43:03 Speaker 6: Sure, it costs money. I'm not saying it's free. 00:43:06 Speaker 13: We do think we're going to get a return on that money, but we've also strategically positioned ourselves so that we are not dependent on an outcome that we have no control over, whereas everyone else virtually in our competitive set is dependent on that outcome. 00:43:18 Speaker 5: Jason, we only have 20 seconds left. Just on prediction markets, what role do you think they played in this year's elections? I mean, Cal State and Polymarket are doing a victory lap right now, just very briefly. 00:43:29 Speaker 13: Well, I think obviously people looked at prediction markets to see how they were doing. We had a lot of traffic for that. There's not a lot of volume in terms of people actually going in there, though, so I'm not sure how accurate some. 00:43:42 Speaker 6: Of them are. 00:43:42 Speaker 13: The ones that have decent liquidity that really, you know, the bigger ones are probably pretty reflective of sentiment. But there's a lot of smaller ones where you can move the needle with very small volume. 00:43:53 Speaker 5: All right, Jason Robbins, always good to catch up. CEO of DraftKings joining us from Boston. 00:43:57 Speaker 6: Thanks, Jason. 00:43:58 Speaker 5: That is going to do it for this Friday edition of Bloomberg Tech. Do not forget to check out our podcast. You can find it on the terminal as well as anywhere you get your podcasts. This is Bloomberg.