Closing Bell - Closing Bell Overtime: Sundar Pichai On Search; Vivek Ramaswamy On DOGE 12/4/24

Episode Date: December 4, 2024

Alphabet CEO Sundar Pichai talks how search is going to change in the next year, plus its latest efforts in AI from Dealbook Conference. Morgan sits down with Vivek Ramaswamy at CNBC CFO Council to talk what his plans are for the Department of Government Efficiency. Synopsys CEO Sassine Ghazi on the latest quarter and the macro spending environment. Mirion CEO Tom Logan on partnering with big tech to use its nuclear technology. 

Well, that bell marks the end of regulation. Decker's ringing the closing bell at the New York Stock Exchange. Robinhood doing the honors at the Nasdaq. Stocks storming higher for yet another record close for the S&P 500 and Nasdaq. And the Dow in uncharted territory. Well, it's charted now as well. Closing above 45,000 for the very first time. That is the scorecard on Wall Street, but winners stay late.

Welcome to Closing Bell Overtime. I'm John Fort. Morgan Brennan is at the CNBC CFO Council Summit in Washington, D.C. We're going to hear from her shortly. Now, get ready for some retail and tech earnings from Five Below, PVH, American Eagle, and Synopsys. CEO of Synopsys will join us before the call right here on Overtime after those numbers are released. But now Alphabet CEO Sundar Pichai is speaking at the New York Times Dealbook Summit. Let's listen in. I was talking earlier about AI, but I wanted to specifically sort of start with you

on AI and technology, but maybe in a bit of a different place. And I wanted to read you, if I could, a quote from a competitor of yours that I don't think you're going to love, but I think you'll have a good answer to, which is your friend Satya Nadella at Microsoft said that Google should have been the default winner in the world of big tech's AI race. He said, Google's a very competent company. Obviously, they have both the talent and the compute. They're vertically integrated. They have everything from data to silicon

to models to products and distribution. And I'm so curious, he sort of put down the debate there or tried to launch a debate there, about where Google is. You guys were the originals. You were the OGs when it comes to AI. And sort of where you think you are in the journey relative to these other players, and whether you think you were or still are supposed to be, as he says, the default winner. I would love to do a side-by-side comparison of Microsoft's own models and our models any day, any time. They're using someone else's models. The gloves come off early, and you're throwing down the gauntlet.

I have a lot of respect for them and the team. Look, it's such a dynamic moment in the industry. I obviously, when I look at what's coming ahead, we are in the earliest stages of a profound shift. We have taken such a deep, full-stack approach to AI. We do world-class research. You know, we are the most cited, when you look at Gen AI,

most cited company, institution in the world. Foundational research. We build AI infrastructure. And when I'm saying AI infrastructure, all the way from silicon, we are in our sixth generation of tensor processing units. You mentioned our product reach.

We have 15 products at have a billion users. We are building foundational models, and we use it internally. We provide it to developers, over 3 million developers. And it's a deep, full-stack investment. We are getting ready for our next generation of models. I just think there's so much innovation ahead.

We are committed to being at the state of the art in this field, and I think we are. Just coming today, we just announced groundbreaking research from a text and image prompt, creating a 3D scene. It's pretty amazing. And so this is, the frontier is moving pretty fast. So looking forward to 2025. But do you think to yourself, I mean, you guys were doing this with Demis and Google

Mines super early. I mean, you were the first. Yeah. And do you look and now say everybody, and this is actually something we talked about a little bit earlier this morning, everybody's catching up and getting to a similar place? In the current, you know, the current generation of LLM models are roughly, you know, a few companies have converged at the top. But I think we're all working on our next versions, too. I think the progress is going to get harder.

When I look at 25, the low-hanging fruit is gone. You know, the curve, the hill is steeper. I think the LE teams will stand out in 25. So I think it's an exciting year from that perspective. You think it's slowing down though? So that's interesting because I will tell you that your competitor slash peer in Sam Altman said there is no wall. There is no wall. You think it's maybe slowing down in terms of how quickly it scales? I'm very confident there will be a lot of progress in 2025. Okay. I think the models are definitely going to get better at reasoning, completing a sequence of actions more reliably, more agentic, if you will.

I think so you will see us push the boundaries. So I expect a lot of progress in 25. So I don't fully subscribe to the wall notion. But when you start out quickly scaling up, you can throw more compute and you can make a lot of progress. But you're definitely going to need deeper breakthroughs as we go to the next stage. So you can perceive it as there's a wall or you perceive it as, you know, there's some small barriers.

But this is what we were actually talking about this morning. How much of it is processing power, meaning just physical processing power? If you can buy enough NVIDIA chips, that that changes the dynamic for you versus how much data is coming new data is coming in whether it's real or digital data or synthetic data versus tweaking and changing the algorithm look everyone is going to get more compute right you know for compute you're only constrained by money right like capital to do it I think there's a lot of capital around. So I don't think processing power, I mean, I think, first of all,

the current amount of compute we are using is just an arbitrary number. It's not like we're using a lot of compute. There's no reason why we can't keep scaling up, right? So I think everyone will get more compute. But I think where the breakthroughs need to come from, where the differentiation needs to come from, is your ability to achieve technical breakthroughs, algorithmic breakthroughs.

How do you make the systems work from a planning standpoint or from a reasoning standpoint? How do you make these systems better? Those are technical breakthroughs ahead. Right. How much do you think that, again, I mentioned Google sort of an OG in this space and now there's been this big competition.

How much do you think that you didn't move, you were not the first mover, I mean, you were the first mover and then not the first mover in this sort of uniquely interesting way. How much is that a function of protecting the current business? I think one of the great challenges that people talk about all the time is you have this enormous business called the Blue Link Economy, Search, and that's $307 billion of revenue, not all of it, but a lot of it. And you do not want to hurt or cannibalize that business.

Do you say to yourself, as you're thinking about these different AI possibilities and other things, you know, we've got to be a little bit slower because we've got to protect this other thing over here. Well, look, I think the area where we applied AI the most aggressively, if anything, in the company was in search. The gaps in search quality was all based on transformers. Internally, we call it BERT and MUM, and we made search multimodal, the search quality improvements. We were improving the language understanding of search. That's why we built transformers in the company.

And if you look at the last couple of years, we have, you know, with AI overviews, Gemini is being used by over a billion users in search alone. And I just feel like we are getting started. Search itself will continue to change profoundly in 25. I think we are going to be able to tackle more complex questions than ever before. I think you'll be surprised, even early in 25,

the kind of newer things search can do compared to where it is today. So there's always, you know, when you're running in technology, there's some version of an innovator's dilemma somewhere. Right. There's only one answer to an innovator's dilemma every time. You lean into that moment. Anything else, you won't be around in a few years' time. Do lean into that moment. Anything else, you won't be around in a few years' time.

Do you think you've leaned in enough? And I'm going to read you something, and it's a little tough, and you've probably read it. Christopher Mims in the journal. He says, quote, the company's core business is under siege. People are increasingly getting answers

from artificial intelligence, and he's talking not necessarily about Gemini, and he's saying younger generations are using other platforms to gather information. And that longer term, he suggests that some of the results that are going to be delivered by search engines, the value of them is going to be deteriorating because they're going to be flooded. The web is going to be flooded with AI-generated content. Look, there are multiple parts to the question. In a world in which you're flooded with a lot of content, sifting and providing people, if anything, I think something like search becomes more valuable.

In a world in which you're inundated with content, you're trying to find trustworthy content, content that makes sense to you, in a way reliably you can use it. I think it becomes more valuable. To your previous part about there's a lot of information out there, people are getting it in many different ways. Look, information is the essence of humanity. We've been on a curve on information. When Facebook came around, people had entirely new way

of getting information, YouTube, Facebook, I can keep going on and on. Through it, I think the problem with a lot of those constructs is they are zero sum in their inherent outlook. They just feel like people are consuming information in a certain limited way and people are all dividing that out, but that's not the reality of what people are doing, right? So if you were going to stack rank the competition now, if you were to sit on a

Monday morning... All right, we've been watching Alphabet, Google CEO Sundar Pichai speaking with our Andrew Ross Sorkin at the New York Times Dealbook Summit. He was there defending search in the age of AI, saying it's still going to be essential and that we're going to be surprised early next year by what search can do. Also talking about how it's going to get harder to advance AI models beyond the year 2025 and saying that he'll do a taste test against Microsoft's AI models anytime. Let's bring in Mike Santoli for some reaction. Mike, there's a lot of the economy, certainly in software, riding on AI advancement. Yeah, for sure. And pretty assertive, I guess, statement of confidence that, you know, Google did not squander some kind of head start.

And I do like the idea that, you know, search is is not something that's just this fixed pool of activity that it's theirs to lose. Now, the market is treating Alphabet shares as if there is perhaps something to lose on a net basis market share wise, because it is unwilling to pay up as much for each dollar of current earnings as it has been historically. But, you know, that's that's for the company to prove at this point as the volume of information scales, whether, you know, Google remains somewhere essential to it. Although, you know, Alphabet stock, parent of Google, was down around 130 a share at the lows of the year. And now it's up 174. And then if we look back last year, I mean, there were a lot of people betting that AI was going to take them out. For sure. I mean, it's obviously not been left behind.

It's not been declared a loser. But it's been kind of a market performer on a year to date basis. And again, it's really much more about relative to things like Meta, which used to have pretty similar valuations. It has lagged behind, at least in this phase. So obviously, you know, there's a there's a big regulatory shadow out there and who knows how much that's playing into it. Indeed, Mike, we'll see you again in just a bit. I want to get to another big interview. Vivek Ramaswamy, president-elect, Trump's pick to co-lead the new Department of Government Efficiency, or DOGE, is speaking at the CNBC CFO Council Summit. Let's listen in. For cutting the federal budget, did he give you any guidance? Well, first thing I'll say is I'm really grateful to President Trump for the opportunity

because I think it does take a different kind of president to prioritize this kind of project. This has been attempted before. Conservatives and even some Democrats have talked about for a long time the importance of downsizing the federal government, mostly just to the tune of rhetoric without a lot of action to go with it. And if you do something the way it's always been done and it's failed before, it's going to fail again. President Trump had the creativity and vision to set this up differently, to say that if you want a different result, you've got to do this very differently.

From the outside, you've got two outsiders, Elon, myself, President Trump in many ways is himself an outsider as the president. And to be able to look at not just, and I think this is important to understand about the project here, not just about cutting cost, but about restoring true self-governance in America, where right now the people we elect to run the government, they're not the ones who actually run the government. It's unelected, and even in many cases, unappointed federal bureaucrats that make most decisions that impact everyday Americans, both through the regulatory state and even in many cases, unappointed federal bureaucrats that make most decisions that impact everyday Americans, both through the regulatory state and even through expenditures that go out the door from people who were not elected and not even appointed by those who were

elected. So it's a massive problem we're taking on, but I do think that it is something that we're taking on in a very different way, and I hope to have, therefore, a different result. Is there any one thing that President Trump has told you, I want you to go cut that? I think he actually is looking at it the other way, in terms of keeping a very open mind, top down, to say that let's start with a clean slate. Let's start fresh, and everything is in scope

for what counts as government efficiency. It's not one part of the federal government or the other. It's actually looking at it as a whole. Are we respecting the taxpayer dollar? In some ways, to use an analogy to those in this room, in the way that a CFO would look at their shareholder capital. In many ways, you are also stewards of capital, right?

Your shareholders are the owners of your companies. I assume these are mostly publicly traded company CFOs in the room. Your duty is to the shareholder to ensure that that money is spent to maximize the size of their pie. You're their steward. I don't think the people who have historically spent government dollars view themselves in that same type of stewardship relationship with the taxpayer who is like effectively the shareholder of the federal government. And I think if you view the way the federal government spends the taxpayer money, and that was in a corporate context, that would be a fiduciary

breach hands down. I think the officers of that public company would at minimum be fired and at worst be civilly or worse in court over the dereliction of duty they owe to the people who have entrusted them with that capital. And so I think that that's what this exercise is about, is to restore that sense of duty that the government officials, and in many cases elected officials, owe to their stakeholder, which is the taxpayer. It's the equivalent of your relationship to the shareholder. And I think many of you probably, I know for sure because the bar is so low on the other side, do a better job than the government officials do in looking after the fiduciary obligation, you could say, to the taxpayer. And that's a big part of what we want to restore.

Let me ask you this then. You know, I've been watching Elon Musk's feed on X, and he's been spotlighting, you know, these little ridiculous things that the government spends money on, you know, squirrel mating research, or whatever comical thing you can come up with. And those are small dollar, they're funny, and they illustrate the point, but they're small dollar things. Can you give us a big dollar thing that you'd be willing to cut on day one? Something north of $10 billion, say. Well, first of all, we're three weeks in,

and January 20th is when we officially start. So I'm actually not to cut to trillion. I'm going to resist the idea of cherry picking something in advance. We'll have much more to say when we get started on January 20th. But what I will say is people who run companies or CFOs of companies will be sympathetic to this. Sometimes it's actually the aggregate of a bunch of small things that you can actually use to drive a big change rather than take aim at something big but might be strategic in the corporate context to a company. So a CFO is an interesting position because the CEO sets a strategic mandate, and that often requires

expending capital in order to advance that strategy. But that means that everything else that doesn't align with that strategy is on the chopping block. And so when you look at, for example, the NIH and the fact that for every grant the NIH makes, it has to give 70% of a true-up on top of that to the university slush fund for overhead. When private institutions that donate to those same scientific causes only have a rake of 10%, that's a real problem. For taxpayer perspective, it's not a huge dollar amount individually that NIH grants add up to. You're talking about billions of dollars in aggregate. It's not small money. But in terms of moving the federal budget, is that one cost going to do it? No. You talk about foreign aid to China, which I posted about yesterday, right? We're giving foreign aid to

China, but we're actually using Chinese borrowed money to do it. So we pay back the borrowed money with interest, but then part of that, even if it's a small amount, goes back to actually aid to that very country. So I actually think that there's a case to be made for looking at the aggregate of a bunch of the small expenses that fly below the radar of actual scrutiny. And in many cases, that adds up to something that could make a dent, even in the comparison to big expenses. So I wouldn't understate the importance of that. Yeah, and it's crazy, too, just because China is also our number one adversary from a national security perspective, even as we're having this conversation. So you've talked about it. You had this Wall Street Journal op-ed a couple of weeks ago. It's not just cost cutting. It is restructuring

to your point. It's regulatory rescissions, administrative reductions, and cost savings is what you said in that op-ed. So it raises the question, how much of this is going to be fundamental changes to, for example, the way the government is contracting and interacting with private sector as well and looking to glean services and looking to run itself? Yeah, look, that's certainly in scope. But I come back to the broad principle, the people we elect to run the government aren't the ones who run the government. And we fought a revolution in 1776 expressly to ensure that we had that chain of accountability. That's created this leviathan of an administrative state, the millions of civil servants who were never elected or even politically appointed to their positions. And part of the problem was you have millions of people who shouldn't have a job that are in that position.

They find the things to do. So it's not that their headcount costs are actually the major issue. It's the fact that they take on responsibilities that go beyond the scope of what the federal government should have been doing. And the best evidence we have for that is that you look at the Supreme Court over the last few years, effectively telling the administrative state that much of what you're doing is illegal. Countless cases in the last several years alone, including at the Supreme Court, the overturning of Chevron deference in the Loper-Bright case. Look at the major questions doctrine in West Virginia versus EPA that said many of the coal mining regulations were themselves unlawful because they never actually went through

Congress. What we have right now is hard evidence. This is not a political opinion. This is hard evidence delivered by the Supreme Court that much, if not most, of what the federal regulatory state is doing is on its terms right now illegal and unconstitutional. That needs to be fixed. Once you fix that, that's actually the root cause for a lot of the spending overruns and the carelessness with which the government's contracting with private parties who are charging the federal government far more than they would charge either another government or another private counterparty. Looking at the overruns of just raw waste, fraud, and abuse, the unauthorized dollars that are being spent, over half a trillion dollars a year of unauthorized expenses are still going out the door. You look at the way in which that money is being spent

through Medicare, Medicaid, Social Security. People love to have lazy armchair discussions about, oh, are you gonna make cuts to entitlements or not? When in fact, the dirty little secret is that many of those entitlement dollars aren't even going to people who they were supposed to be going to in the first place. They're hundreds of billions of dollars of savings

to extract just from basic program integrity measures to make sure that the person who's actually getting the money was indeed the person for whom it was intended or was supposed to be legally getting the money in the first place. And so my view is a lot of the political philosophical musing about whether or not we should be cutting entitlements or not or whether or not the president can impound funds, it's a deflection from the fact that there's a lot lower hanging fruit in terms of waste, fraud, abuse, error, program integrity failures that you got to go after first before we ever need to get to those harder political questions. And I hope that's a big part of what we're able to accomplish. Well, we have been watching Vivek Ramaswamy, president-elect Trump's pick to co-lead the Department of Government Efficiency, making the case for why that is necessary, that government efficiency is necessary at the CFO Council Summit, CNBC's. That is. Meantime, Synopsys earnings are out. Stock is lower. Sima Modi has the numbers. Sima. Hey, John. Synopsys delivering a solid

fourth quarter beat revenue in line, but looking ahead to the first quarter, guidance was light for revenue and earnings. And even for the full year, revenue guidance was a bit light. We're looking at shares down about 7%. This is a company that specializes in tools and services that are used by semiconductors and manufacturers with partnerships with NVIDIA, the likes of Taiwan Semi. So a key component and company to look to when thinking about the AI trade, John. Seema, thanks. And joining us now is Synopsys CEO, Sassin Ghazi. Sassin, good to see you. So a bit of a disconnect here in the guidance between what the street was expecting, what you're delivering.

But first of all, tell me about the quarter. Pretty much bang on, a little above with revenue and certainly with EPS. What's going right? And then in a challenging environment for semiconductor makers and some of your customers, what are you seeing from them? Hey, John, good to be here. First of all, we just wrapped up a fantastic FY24, 15% growth, 25% EPS. As far as Q1 comp, Q1 25 comparison to Q1 24, there are some mechanical accounting, I want to call them, that I'm looking forward to explaining on our earnings call to our investors. What we had in Q1-24 is an extra week. And our software security business, which we sold in Q2 and took it into discontinued operation, is throwing off the comparison.

As well as, you know, we have a mix of our product portfolio that is upfront revenue. So quarter to quarter, there is a disconnect. However, if you look at the year, we're still guiding a double-digit growth for FY25 after a 15% growth in 24. And if you zoom out, after a 15% CAGR growth over the last five years. So we're very confident about where we are and where the year is going to shape out to be. Okay. You mentioned the full year. I do want to put that in perspective. The consensus on the full year revenue is at $6.905 billion, and you're guiding to a range of $6.745 to $6.805. So how much of that has to do with the fewer weeks that you pointed to in the first quarter? How much of it

is you also being careful given the state of the semiconductor market right now and how your customers are behaving? So some of it is the accounting. And the other part is if you look at the macro today, we are living in the semiconductor market in what I like to call in the tale of two markets. You have the AI infrastructure build-out. So any company who's delivering to the AI infrastructure build-out, meaning if you're a memory semiconductor provider or an advanced logic CPU, GPU, you're killing it. You're doing very well. And we're benefiting greatly from it. If you look at the second cohort, which is the mobile, the PC, industrial, automotive, they're still trying to find a way to leverage AI on devices and still trailing in the recovery.

Now, as far as Synopsys is concerned in serving that market, we do well. And the reason we do well, we're tied to R&D. But we're not growing as fast as the first cohort, which is the AI infrastructure cohort. Keeping all of that in perspective, I want to talk about systems. You expect your ANSYS acquisition, I believe, to still close in the first half of calendar 25. And that is more beyond just the chips themselves, envisioning how all of these systems are going to work together, being able to model that for customers. We've seen more of this AI expectation showing up in results of companies, even like Pure Storage here on overtime yesterday, and retooling the degree to which storage draws energy in the data center. How much are those changes, and even how

infrastructure is being placed in the data center, how much is that influencing the need for the type of software and modeling solutions that you expect for Synopsys to deliver? Yeah, that's such an exciting area for Synopsys. As you pointed out, with the Ansys acquisition, which is a $35 billion acquisition we're anticipating to close the first half of 2025, we envision Synopsys to deliver the engineering design solutions from silicon all the way up to system. Almost every hyperscaler that I have conversations with, they are looking to optimize at the system level. What does that mean? As they're building their data center, of course, it's filled with semiconductor components. But in order to optimize for energy, in order to optimize for security, for reliability of this data center,

you have to take into account the system performance, the workload, the software you're running on that system. So Synopsys plus Ansys will provide that engineering workflow from the semiconductor all the way to that system level definition. And that does not only apply to data center. Envision the future car or robots or industrial applications. They all need to optimize from that silicon to system. All right. Sassin Ghazi, CEO of Synopsys, thank you for giving us that perspective before the call on those earnings that just crossed minutes ago. Well, Cloud Week continues tomorrow on Overtime, and we're going to speak exclusively with ServiceNow CEO Bill McDermott. His stock is helping the iShares expanded tech software ETF hit a new high today,

also driving that gain. Salesforce having the biggest impact on the Dow up 11 percent today following its revenue beat, raising the low end of its full year sales outlook. Well, beyond tech in retail, five below earnings are out. Julia Boorstin has the numbers. Julia. That's right. Five below shares are soaring now up nearly 11 percent on revenues of 844 million, ahead of estimates of $799 million, with adjusted earnings of $0.42 per share. Now, same-store sales did increase 0.6%. That was a surprise gain because consensus expectations were a 4% decline in same-store sales. Five Below also appointing a new CEO, Winnie Park, that's effective December 16th. She had been CEO of Forever 21 since 2022.

Now shifting gears over to American Eagle, plunging on weaker fourth quarter guidance. You see shares are now down about 14%. Now the company did beat on the bottom line with adjusted earnings of 48 cents per share. That was two cents better than expected. Revenues of 1.29 billion, just a hair below estimates. The company's fourth quarter guidance of a 1% increase in comp store sales is also lower than the street account estimate, guiding to a 2.2% increase. So that's a big disappointment there. The company warning, we remain cognizant of potential choppiness during non-peak periods. Now, shifting gears over to PVH, shares of that stock are now down about 7 percent, earnings of $3.03 billion, beating estimates of $2.59 billion and revenues. I'm

sorry. Excuse me. That was an earnings per share estimates of $2.59 and revenues of $2.26 billion, beating expectations of $2.2 billion. But the company warning that it does see fourth quarter earnings in a range of $3.05 to $3.20. That's much lower than the estimate of $3.71. The company also guiding to revenues down 6% to 7%, which is lower than analysts' projection for a much smaller decline. Now, Sentinel-1 shares are also falling. Now we see Sentinel-1 shares down about 9.5%. That company reporting zero cents in adjusted earnings, missing expectations by a penny.

Revenues of $211 million, $1 million ahead of estimates. But the real miss here that's putting pressure on SentinelOne shares, the company's non-GAAP operating margin was negative 5% in the quarter versus the negative 2.9% street account estimate. Don, back over to you. All right. That was a lot of results, Julia Boorstin. Thanks. Well, Fed Chair Jay Powell making headlines at the New York Times Dealbook Summit in his last public appearance before the Fed's next interest rate decision. Our Steve Leisman has the highlights. Steve. Yeah.

Hey, John. Fed Chair Jay Powell telling CNBC's Andrew Ross Sorkin that the Fed remains on course to cut interest rates towards neutral, but is going to be cautious about it. The economy is strong and it's stronger than we thought it was going to be in September. So the labor market is better and the downside risks appear to be less in the labor market. Growth is definitely stronger than we thought. And inflation is coming a little higher. So the good news is that we can afford to be a little more cautious in as we as we try to find neutral. But that hint of hawkishness looked to have a little effect on fairly confident pricing for a December rate cut in futures markets.

The December contract priced with a 75% probability of a 25 basis point cut. Pause remains priced in for January, you can see there. But the chance of a follow-on cut in March has kind of quietly crept up to now stand around 60%, 62%. Just last week, it was around 38%. So Powell was either unsuccessful at tamping down rate cut enthusiasm or he was content to let stand relatively dovish market pricing. Markets also today were reacting to the weaker ADP employment report than expected and ISM services both coming in below expectations. Add to that the Beige Book headlines from the Fed supporting a weaker take

on the economy, saying economic activity rose slightly in most districts, employment levels were flat or up only slightly, and prices rose only at a modest pace. Businesses were, however, optimistic about demand in the coming months. Powell's comments did give Bitcoin a boost when he said it was not a competitor to the dollar, but rather to gold. Bitcoin traders seem to take the comparison to gold as an endorsement, even while Powell called it a highly speculative asset that was highly volatile and didn't seem to mean it as a stamp of approval, guys. So maybe a little irony in that take there on Bitcoin, John.

Well, the Bitcoin crowd will take it as a compliment if it's at all possible. Yeah, right, exactly. Well, it's time for a CNBC News update with Kate Rogers. Kate. Hi, John. A manhunt is going on in New York City for a gunman who shot and killed United Healthcare CEO Brian Thompson as he headed into an Investor Day event in Midtown Manhattan early Wednesday morning. Police say the gunman was lying in wait and shot Thompson repeatedly from behind before fleeing on an e-bike into Central Park. New York's police commissioner called it a, quote, brazen, targeted attack.

Thompson's wife tells NBC News that he had been receiving threats. French Prime Minister Michel Barnier was ousted this afternoon in a no-confidence vote from Parliament just three months into the job, becoming the shortest-serving prime minister in French history, lawmakers voting him out after an attempt to push through part of his government's budget. And President-elect Trump's pick to run the FBI, Kash Patel, was the target of an Iranian cyberattack. That's according to two senior law enforcement officials who spoke to NBC News. The hackers are believed to have targeted some of his communications,

but it's unclear whether they accessed the info or how much data was viewed or stolen. John, back over to you. Okay, thank you. Up next, an under-the-radar AI trade. The CEO of Mirion Technologies on how bringing decommissioned nuclear plants back online to power AI is impacting his business. Plus, Jeff Bezos will take the stage at the New York Times Dealbook Summit just moments from now, and we will bring that to you live as soon as it begins. We'll be right back. Welcome back to Overtime. Meta announcing yesterday it's looking to partner with nuclear energy developers to pursue its AI ambitions as well as meet its sustainability goals.

This follows recent announcements by Amazon, Microsoft and Google that they're planning to help power their data centers with nuclear reactors. Well, Marion Technologies hopes to become one of the longterm winners of big tech deploying nuclear power. The nuclear and cancer care company projected a strong 2025 and announced a $100 million stock buyback program at its investor day this week. Shares are up almost 70% this year. Joining me now on set is Thomas Logan, the founding chairman and CEO of Marriott. Tom, good to have you. So which should investors focus more on here, the cancer care part or the nuclear power for data centers part? You know, John, I'd love to

say both. I think right now what's commanding the greatest attention certainly is the nuclear power play. The announcement a few months ago by Microsoft and Constellation about their deal to bring back the Three Mile Island power plant really has, I think, brought nuclear power more into the public consciousness, particularly within the investor community, but really fundamentally highlights the fact that the world simply does not have the electrical generating capacity, particularly baseload carbon-free capacity, that it needs to power the AI revolution. This is almost 40 percent of our total business supporting the nuclear infrastructure. And obviously, we're strong supporters of this. We think we have a very important role to play here.

Is there the danger of oversupply in this market? We've been talking to a lot of people who are kind of rushing into this, and there are some smaller scale nuclear options as well. I think it's highly unlikely, in part because the way that the new capacity will effectively be metered out as it's rolling out. And we'll see, I believe, a few successive waves. Firstly, through the installed base, which today is about 440 operating reactors, will generate more capacity through higher capacity utilization, capacity up rates, life extensions and the like. On top of that there are today roughly 50 utility scale new reactors under construction globally and then behind that we have what what

is even more exciting which is this wave of small modular reactor technology. So I think as the successive waves are metering out, really concurrent with the build-out, the roll-out of AI, I see limited risk of excess capacity. What positions you to have a better structured, better positioned business supporting this push into nuclear? Well, the great thing about Merion, firstly, is that if you look at our aggregate exposure to commercial nuclear power, it's almost 40 percent of our total business. So we believe we're one of the better ways to actually play this. But importantly, if you look through that, what you'll see is that 80 percent of our nuclear power related revenue comes from the installed

base. And so as a result, it is very visible, very predictable and very sensitive overall to these favorable dynamics that we see. When you look beyond that, the new build activity, both utility scale and SMRs, really is kind of an additive optionality to that overall business model. How much is software and data about how that power is available, how it's performing, how much is that going to be important? It's going to be hugely important because if you look at the legacy industry right now, it's still very much an analog industry.

But as we see the pressures coming from the SMR development community, where they demand digital solutions, software ecosystems that are more comprehensive on the front end, that I think is going to drive very rapid change. And then beyond that, as you look at the workforce within utility scale, there is a generational change taking place. Digital natives that demand digital solutions. I think it'll be hugely important. All right. AI is hungry.

Tom, thanks. Tom Logan from Merion. Well, investors are basically all in on stocks right now. Up next, Mike Santoli looks at what that could mean for the market. Plus, a new study reveals which popular weight loss drug appears to be the clear winner when it comes to people shedding pounds. We've got details coming up on Overtime. Welcome back. Mike Santoli is looking at whether investors are overly stocked with stocks. Mike? Yeah, John, I would say at least that in aggregate, they're amply supplied with equities.

J.P. Morgan Strategies put this together. It's the implied global asset allocation, basically toting up the value of all these assets around the world, subtracting what is owned by central banks to say that this is what's in private hands. And you see the top line there. That's equities as a percentage of the whole. And it's pretty much as high as it's been in more than 20 years. You go back to the very late 90s. It did get higher. Most of this is because equity values are up, right? The indexes are higher. Somebody owns all those stocks. And therefore, it's not as if it's all about new flows into that asset class.

Nonetheless, this is how it sets up. And you see pretty low when it comes to fixed income in private hands there, too. Higher for longer rates. Maybe that's what people expect. Cash is pretty steady there. So I would just argue that there very well may be upside to this measure. And this bull market certainly doesn't seem like it's over, but it doesn't mean that investors collectively need to incrementally buy that aggressively to maintain their exposure to the asset class. That's the financial crisis, that second

peak. It sure is. Yeah. Yeah. That's 2000. That's 07, 08. Yeah. OK, well, we'll have to be careful. Yes, that's right. Thank you. Well, OpenAI announcing a new partnership with a high-profile defense startup. Up next, how that deal could impact national security. And we'll bring you Jeff Bezos' comments from DealBook as soon as he takes the stage. Just minutes from now. Over time, we'll be right back. Welcome back. AeroVironment shares down about 7.5% in overtime on earnings. The defense drone maker beating sales estimates but missing on the bottom line

and issuing weak full-year profit and revenue guidance. Now let's get to Morgan Brennan with some more news on the aerospace and defense industry. Say, Morgan. Yeah, it's a busy day for the aerospace and defense industry and what is going to be a busy week in terms of news. So OpenAI, John, stepping further into the national security arena today, announcing a partnership with defense tech upstart Andrel Industries to develop and responsibly deploy advanced AI for national security missions.

And this brings together OpenAI's models with Anduril's defense systems and Lattice software platform. And if I put this really simply, this enables Anduril to leverage OpenAI models to assess drone threats more quickly, more accurately, give military personnel the information to make better decisions while staying out of harm's way. This is very much focused on counter drone technology capabilities. Now in the press release the company highlighting that in light of the race against adversaries like China this strengthens the US commitment to maintaining a technological edge and ensures that AI tools upholds democratic values while protecting military and critical infrastructure. Now OpenAI has been expanding into the space, it's been working with the Air Force, with DARPA. But this partnership represents the first for the

startup with another company in the national security space. Now, it comes after Palantir struck a recent partnership with Anthropic. And last month, Meta said it would make its Lama AI available to the federal government and defense contractors. It's a big shift in narrative from five years ago when Silicon Valley did not want to be affiliated with defense or national security work. But as the world becomes more dangerous, we've actually seen that completely change and being seen now as a means of deterrence and an ethical decision to actually work with the U.S. government in a geopolitical landscape in which China is also racing to deploy AI in these ways. Now, meantime, if we stick with aerospace and defense and we shift to policy,

former President Trump nominating Jared Isaacman, President-elect Trump nominating Jared Isaacman. This is a face familiar to overtime viewers to lead NASA as the administrator. Isaacman is CEO and founder of Shift4. He's also an astronaut two times over in historic space flights. He did that spacewalk just earlier in the year. Those space flights conducted with Elon Musk's SpaceX. Now, I had asked Isaacman about what the Trump administration might mean for the space sector and for policy in his last appearance here on Overtime just last month. One clear takeaway, SpaceX can expect a lot more government business. I'm not. And here's what he had to say. It's almost been like controversial to give

contracts to the company that has the best product at the lowest cost in SpaceX. Like you almost had to throw money away on alternative vendors just to appear not to be too Elon friendly. When really as taxpayers, what we should care about, you know, is low, you know, best product, lowest cost, you know, let's, let's get a lot of Starlinks up there and improve connectivity across the world. So I'm not sure that I would say the clear takeaway is that SpaceX should expect more business, but that this type of contracting and competition should expect to ramp potentially. So Isaacman writing on X in part that, quote, Americans will walk on the moon and Mars, and in doing so, we will make life better here on Earth. Shift four,

though, that stock actually tumbled on this news, closing down over 12% on that news. You can see right there on your screen. It was the worst day of the year, John. All right, a lot going on for sure, Morgan, thanks. Well, Lilly, one of the big winners in the S&P 500 today after a new study found just how effective its weight loss drug, ZetBand, is compared to rival WeeGovie. Details straight ahead. Welcome back to Overtime. Alphabet and Google CEO Sundar Pichai

speaking with Andrew Ross Sorkin at the New York Times Dealbook Summit this hour. Andrew asked Pichai for his take on the Department of Justice antitrust case against the company. Here's what he had to say. We are the cutting edge of innovation. We provide these services. We open source and publish most of this innovation. The entire AI, the transformer architecture, etc. A lot of our products are open source. So, you know, so I will vigorously defend. I can't comment on ongoing litigation, but I have deep faith in our judicial system. We're going to have more from DealBook coming up on CNBC, including Andrew's conversation with Amazon founder Jeff Bezos. Well, shares of Eli Lilly making big gains today, even as a new study shows its weight loss drug is more effective than its main rival, Wegovi. Angelica Peebles has the details. I got it that time. Yes, you did, John.

And so today, this trial is confirming that people lose more weight on Zepound than they do on Wegovi. And if you talk to any obesity doctor, that's what they'll tell you. But now Lilly can definitively say that with the results of this head-to-head study. So after about a year and a half, people who took Zepalin lost an average of 20% of their body weight versus 14% for people who took Wagobe. And those numbers are similar to what we've seen in previous studies of the drugs, so it's not too surprising. Now, Lilly also saying that the side effects for both drugs were similar to previous studies,

mostly GI-related and generally mild or moderate, but the company not providing a detailed breakdown at this point, so that will be one to watch when Lilly does share the full results next year. And Lilly in a statement saying that they conducted the study to help doctors and patients make informed decisions about treatment choice. But at least right now, those decisions are really coming down to supply and access. So we'll have to see how this moves the needle, John. Do you expect it to? You know, I really think right now it's about supply and access because, again, you know, you talk to doctors and they'll tell you, you know, maybe they want to put someone on one or the other, but ultimately it's up to the insurance coverage

that they have. So and also what's available at the pharmacy. Of course. Angelica, thanks. Well, data resilience company Veeam, which Insight Partners bought for five billion dollars five years ago, today announced a 15 billion billion valuation and a $2 billion secondary sale. I spoke with CEO Anand Eswaran about how it sets the company up for the public markets. We're an incredibly profitable company. And in fact, just this year, we're going to probably spit out, you know, half a billion dollars in cash. And so the secondary was not to raise cash or liquidity because we needed it. The primary intent was to welcome new investors who will stay with us, you know, for the longer term, for the journey,

of which the next milestone would likely be an IPO. Avim's public competitors now include Rubrik, which reports earnings tomorrow here on Overtime, and Commvault. Private competitors include Druva and Cohesity, whose $3 billion acquisition of Veritas could close any day now. Cohesity CEO Sanjay Poonen has openly said the Veritas buy is about scaling up ahead of its own IPO. So 2025 could be eventful for public debuts in resilience and recovery if the markets hold up. And the excitement around software, very much a part of the market story today, holds up as well. And speaking of software, I mean, the NASDAQ closed up 1.3 percent. We had the Dow close above 45,000 for the first time.

And then tomorrow, the story continues. One of the hot names in software, in enterprise software, has been serviced now. It's at all time highs above eleven hundred a share. The CEO, Bill McDermott, is going to join us right here on Overtime tomorrow as Cloud Week continues. Of course, we had AWS reInvent this week, which is a really huge event in the cloud world. We had Sundar Pichai, the CEO of Google and Alphabet on here as well. So a lot of news around AI and the cloud continuing here on Overtime. That's going to do it for Overtime today. Fast Money starts now.