Power Lunch - Stocks slide as a jump in bond yields threaten the bull market 5/19/26

Episode Date: May 19, 2026

Major averages coming under pressure from higher treasury yields as consumer and tech stocks lead the declines. Kelly Evans & Brian Sullivan break down whether the recent market rally may be in jeopardy. Meanwhile, media mogul, Byron Allen, and Google’s Chief AI Architect, Koray Kavukcuoglu, join the show for exclusive interviews on the state of their industries.  Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Government borrowing costs hitting their highest level since before the financial crisis in 2007. Welcome to Power Lunch, everybody alongside Kelly, I am Brian Stocks. Really, though, staging a turnaround how we finish today. Still an open question. Plus, media mogul Byron Allen making headlines again from a billion-dollar deal ambition to the future of local TV. What could his next move be and mean for media? He is here and on set. And Google's annual developers conference, it is underway.

CEO Sundar Pichai laying out the product updates, and we have the very, very late. And like you said, Brian, the big story today is this moving yields. The rise is not at all just happening here in the U.S., but is certainly dragging us higher. The 10-year, its highest level in more than a year. It's also impacting mortgage rates. We'll talk about that. Germany's 10-year boon, highest since 2011. France, highest since 2009.

In the UK, the 10-year guilt has climbed to levels not seen since 2008. And then in Japan, the 10-year yield reaching its highest since 1997 earlier this week. Those higher yields adding a wrinkle to the bull market with both the S&P and NASDAQ on pace for a third straight losing session. While off session lows, some of the biggest names in the AI trade are leading the markets lower. Our next guest still believes it's a long-term story but says the blistering pace of this rally is set to slow and is advising to add some protection. Osang is chief equity strategist at Wells Fargo. It's great to see you again. Welcome. Just talk through, wait, what, so you know, you're an equity strategist.

You know, hot trades act like this all the time. So is this kind of just a momentary piece of caution? Or do you think there's a larger frothy action here that you were a little concerned about? Yeah. So I think last Thursday was essentially a local peak. I don't think it was the big peak in this bull market, but a local peak. And we're probably going to slow down in terms of this rally going forward. We were very bullish the past month.

We were calling for the sugar high for the next three months, calling for $7,300 on the S&P by July. And that happened. So we're... We got to almost $7,500. Yeah, exactly. So that's been breached, actually. Now we're at a point where we had this massive rally. Our sentiment indicator, our contrarian sentiment indicator, triggered a cell signal for the first time since November 2021.

Wait, what is that based on the contrarian cell indicator? There are five variables that go into it. So price momentum, put cost. ratio, the skew between call and put options, ETF flows and the futures positioning. So it measures basically the overall sentiment and positioning and flows into the market. And that last Thursday triggered the first cell warning since in five years, basically? Two weeks ago.

Two weeks ago, it triggered a cell signal basically in four and a half years. And the last time it triggered a cell was November 2021, which was right before the bear market of 2022. So sentiment has gotten pretty euphoric. And going for our guess is that the speed of the rally is going to slow. Okay, but where's the selling net? Because, yeah, we might be down a little bit right now. We'll probably end the day higher. Who knows, the NASDAG is down three one hundredths of a percent at this point, right?

By the way, I have to say you are a very well-dressed man. Thank you. If you're on the radio, he and I are basically wearing the exact same outfit. I want to make that clear. But if this is the sell-off, half a percent, that's what we get. Yeah, so we don't see that much downside risk. There's been none. The market's up 5% this month.

Yeah, so because the rally has been really driven by fundamentals rather than just multiples and speculation. I mean, there was a little bit of that, but largely it was driven by fundamentals and earnings that we don't see that much downside risk. And also, I do think the AI cycle is actually getting healthier. So over the past two years, the AI bull market has really been driven by this KAP-X cycle,

But I think that's now maturing to a monetization-led bull market. So I think there was something to monetization-led bull market. So hyper-stallers are starting to put up good numbers and starting to monetize on their business. 800 billion estimated in capital spending over the next year. That's the number. Everyone's got their own number, but it's 700 to 800 billion. That's a lot of money. Yeah.

That's one and a half times Singapore's GDP. Right. But then demand still well-off-stripped supply, and will probably continue to do so over the next two years at least. And if that's the case, they get the pricing power. They're starting to put up ROI. Yeah, Free cash flow is really what was concerning over the past two years, and that's why hypers underperformers perform. But they are starting to show acceleration in cloud revenue.

So cloud revenue last quarter accelerated by 6 percentage points. So, yeah, we're starting to see that monetization playing out as well. I'm not sure. It's remarkable. So you mentioned the last time you saw some of these sort of positioning indicators, November of 21. We had an awful market in 22. We fell basically right from the start of the year. We were down, what, 20% by the lows in the fall before chat GPT came out. So, but it doesn't sound like you're expecting or warning about anything like that right now. I mean, Paul Tudor Jones says a year or two left in this rally. I mean, others would say three, five or ten. But when I listen to you, I think, is it possible that that we are closer to kind of an, so just talk a little bit about like the downside. that you're talking about. Yeah, so our call is to own the underlying equities in AI and sell calls because call premium is very rich right now because everyone was piled into this right-tail risk.

The consensus playbook of the market the past few weeks was this is 1999. We're going to see AI bubble forming. And everyone piled into buying calls. So call premium is very rich. So our view is if the speed of the rally slows, then it makes a lot of sense to sell calls and buy and own the underlying assets. You know, the 35-day move in technology has been the best in years. In years.

And the NASDAX's up 16% this year. This year. It's actually, despite everything else, we're on pace for one of the best years, or at least best few months of all time for technology. You're starting to sound a little bit cautious, but you are still overweight technology. how do you square the two? Yeah, I don't think you can really bet against this AI cycle. Like I said, I do think AI cycle is actually maturing and getting healthier because of monetization.

And the reason why semis didn't really go up much back in Q4, despite hyper-scalers raising numbers, KAPX numbers, was because of the concerns around overinvestment and monetization. And if hyperscalers are starting to monetize more, then that concern of over-investment subsides, which is actually a healthier cycle for semi-es. even. So I think, I don't think you can really bet against AI, at least yet. That being said, I think what could potentially derail this AI cycle is the economy. Because I think this AI

cap-back still hinges on the economy. And AI is getting, it's becoming huge. So semi-zan hardware, right now it's at about 2.5% of total GDP. That's going to go up to 3% based on hyperscalage consensus estimates. So that's actually above the internet bubble peak. and it's basically in line with the railway investment peak back in the 1850s. So unless my concern is that if the other parts of the economy don't really grow anymore and don't really support the investment cycle,

at the end of the day, AI CapEx is still being funded by Hyperscaler's core businesses, which are tied to the economy. So I think ultimately it hinges on the economy. Quick final one on this as well. I mean, to talk a little bit about the Russell 2000, you think AI actually drove some of that outperform, and do you see risk there in particular now?

Yeah, so Russell 2000 has massively outperformed the S&P so far this year. And a lot of people still look at the Russell II as some sort of a value or cyclical index. That's not the case anymore. A lot of the big stocks in the Russell 2 are actually AI-related. It's actually power stocks, quantum stocks, things like that, and AI infrastructure overall. And those were really the biggest drivers of the Russell 2 outperformance over the past year. And there is a Russell Rebalance coming up in June. And the biggest stocks are probably going to move up to the Russell 1000 from Russell 2000.

So if those top 20 stocks move up from the Russell 2000, excluding those, the Russell 2 actually underperform the S&P. So I think that could actually be negative for the Russell 2. Well, and we're going to let you go. he said that, the NASDAQ turned positive. See, I don't know about the Russell. But by the way, his tie has a little bit more of like dark blue dots against the light blue. It's fair. Well, his is more expensive than mine.

He's the equity strategist. I'm just a beloved TV host. You tell that just by looking at it? I could just, I could tell you's a very dapper individual. Osung, thanks very much. Oh, Sun, Kwan. All right, coming up, is it time to start thinking about trimming some of those big tech winners? We just talked about it a bit, but your next guest

considering taking some profits. from Google. We'll find out why. Next. Alphabet, Google's parent company and Focus today with its annual developers conference kicking off just about an hour ago. Kelly let her show with it. Stock is so far down a little bit today along with the broader tech sector, but let's be clear, it has been an unbelievable year for Alphabet investors. Here's Alphabet CEO Sundar Pachai taking stage moments ago. There he is. And now let's praying a longtime Alphabet shareholder for more reaction on what we've heard so far, what is to cover. Nancy Tangler, CIO and CEO of Laffer Tangler investments, whatever today, yesterday, tomorrow. Who cares? Alphabet shareholders, Nancy, like you, have printed money for years. Is that going to

continue? Thanks for having me, Brian. Love the tie. I think, yes, I think the company will continue to outperform. But the question is at what pace. And so what we do as portfolio managers is we look to be buying from eager sellers and selling to eager buyers. That's how we figured out was the most productive way to make money over the long term. So the stock's gotten pretty stretched. It's not super overvalued, but it's somewhat, I'd say, fully valued. And so we will be taking some profits. It's become an outsized holding for us across all of our large cap strategies. So you're not selling it. I want to be clear because you're kind of implied you're selling, you are selling, but you're selling simply because it got too big. It did too well, and it became too much of a weight. So it's not like I don't like it. I'm going to sell it. It's just it's big and we need to trim it.

Yeah. Then that's what we call it trimming. But if you look at the company, I mean, they are probably the most dominant provider of cloud and AI services. They've got vertical integration. They've got in the full stack is really dwarfing what the neoclots can do. So I think, the announcement today with Blackstone will be a problem for CoreWeave and some of the other providers over the long term. That's why we like to own the industry leaders, buy them when they're out of favor. So we added Google right after the Bard debacle and everybody wrote them off and the stock was a member of the Russell 1,000 value index. And so we owned it in growth for seven years, but we put it in our value portfolio. And we've been pretty happy with the returns annualized and cumulative, and we think they continue to dominate. We just know the market has a short attention span and will be turning its attention to other places. And that's where we will put the

proceeds as we sort out where that is. Yeah, I'd love to hear about that, Nancy. It's good to see you because we were just speaking with Osung, who thinks the near-term peak was 7,500 on the S&P, 50K on the Dow, 30K or so on the NASDAQ. What do you make of that? And where would you be looking now for, if we need them, new or different areas of leadership. Yeah, well, Kelly, I do think we're due for a correction. We know that the market likes to test the new Fed chair. We think that may be in the offing. You know, I think I was on your show. We were buying calls at the end of March on the S&P. And now we're looking to put in place some protection for our clients because we think we're going to get a pretty decent correction. That said, we're still in a bull market. And so you want to

use it like we did during Deep Seek and like we did earlier this year to add to some of the leaders across the spectrum. So we had in our 6 for 26, we had added D.R. Horton as a housing play. We own Home Depot, for example, in our 12th best. So it's not always just the big winners of recent times. Same with Walmart. You know, we've been talking about that stock for six years as an old economy company pivoting to the new technologies. But that said, if I get an opportunity, To buy more Nvidia at $108 a share like I did during Deepseek, I'm going to do that. So tech still going to lead, still going to dominate. But we've shifted our focus to software.

And we'd said a couple weeks ago, we thought the shift was going to move back in their direction. And so we've been adding to names like Microsoft at Service Now, CrowdStrike, Palantir, and we'll continue to do that. There was a debate on halftime earlier about Palantir and whether it was showing enough leadership and whether you could sort of say that for the software group as a whole. And it sounds like you would say it is or it will. Yeah, it just depends on your time horizon. You know, we're investing for the next three to five years.

A lot of the hedge funds are investing for the next three to five weeks. And we want to own the names, the winners, the ultimate winners. Sometimes we get it wrong. And that's what investing is. It's about being mostly right. But I will say that the names that we are buying, we think are going to be winners. We are not in names like Workday, CRM, or even Adobe.

We exited those names a while ago. So we don't like the seat model. We like the vertical integration, stack model like you get in service now. We think they're an ultimate beneficiary. And same with cybersecurity. That's going to be a boom for CrowdStrike. All right. Nancy, it's great to see you talking about some new areas to luck.

Really appreciate it. Thank you. Tangler with Lavertangler investments. Don't miss, speaking of which, Andrew Ross Sorkin's sitting down with Jeff Bezos from his Blue Origin Rocket Factory at 8 a.m. Eastern tomorrow, they'll be down in Florida. Be sure to tune in. And coming up, we'll go live with the CEO of one of the companies on the CNBC Disruptor 50 list and get into their plans to scale autonomous vehicle business to compete with Tesla. Stay with us for more right after this. CNBC's annual Disruptor 50 list was unveiled today. Spotlighting promising venture-backed companies,

a lot of them innovating with AI, number 27 on the list, Wabi. The Canadian, is Canadian companies eligible? The Canadian startup is in the autonomous vehicle race, pushing to make freight trucks driverless. Founded by former Uber execs, the company raised $750 million in its January funding round. CNBC's Julia Borson heads our list and is here with you. us today. Hello again. It's great to be here and it's great to get an opportunity to talk to Wabi's CEO, Raquel, or Tassoon. Raquel, thank you so much for joining us. You've had a big year, raised a ton of money and your big rigs are already operating on the road in Texas. But there are people behind the wheel. Give us a sense of the timeline to expand beyond Texas and also to remove those

people who are there just for safety. Yeah, Julianne, really a pleasure to see you and to be in the show today. Yeah, so, you know, our software is really, you know, very perform and ready. We are waiting for the final validation from the Volvo platform and then launching, you know, driverless trucks really, really soon. So that's basically what you're going to see over the next, you know, a little bit. But it's really, really imminent from that perspective. And today we operate in Texas. That's where I am, where I am today. And, you know, expanding over the full Sunbelt really on the one to two years timeframe. So there are other companies in this space.

There's Tesla, Aurora Innovation, which is already driving their autonomous vehicles from Texas through the southwest to Phoenix. Are you concerned about this competition, and how do you differentiate? So definitely not concerned about the competition. I think Wabi is in a tremendous position. We have the most scalable approach that really is,

as I mentioned before, is really ready to launch and scale significantly. And unlike some of the competitors, we really truly believe that we should do a safety first approach. So we are not launching driverless before the redundant platform from the OEM is ready. And that's what we are basically waiting for, which is that final validation that already a couple of quarters ago,

Volvo has publicly said that we'll be ready over the next couple of quarters and they also have said that, you know, we will have hundreds of tracks next year. So that gives a sense on the time frame and, you know, how imminent the launch and the scaling is. And, you know, Wabi wants to do this the right way, which is not do something with prototype parts where, you know, partners like OEMs and Tier 1s don't agree that that's the, you know, that's safe to put on public roads and instead really wait for the final, full validated platform to launch. Raquel, it's Kelly here. There have been so many companies in this space

trying to make either freight autonomous or driverless or both. I think that's the same thing. Or electric, that's another one. Has that helped in some ways cleared the way for you and what makes your company different? Or does this just remain a very difficult area to solve? Yeah. So the first wave of cell driving vehicles was really about proving

that this technology can indeed arrive to a point where you can launch a driverless product, and that has happened. That's a reality today. And then the second wave is really about, do you have technology that can truly scale? And the only way to really do this is with an AI-first approach. And that's what really sets apart Wabi from the rest of the competitors,

where thanks to our technology, we actually can have a much better product, where we can go to, you know, direct to customer. We don't need terminals, for example. that removes a massive friction and really for the first time provides a product that is really what the customer wants. And what is also really excited is that this technology enabled us to have the same brain that can drive, you know, robot taxis together with cell driving trucks, you know, providing or putting, you know, really setting apart Wabi from the rest of the competition. So yes, the first wave was necessary to get us here. but we feel very confident in terms of where we are leading this second wave. You mentioned robotoxies.

I know a key reason you raised that $750 million in January was to fund your expansion from these freight vehicles to robotoxies. Robotoxies are also an incredibly crowded space. I live in L.A. where I could take Waymos around. What's the strategy to compete there? And how important do you think partnerships will be to enable you to really get a foothold? Yeah, what has been very interesting is that, you know, today in our trucks, we have capabilities to drive in what we call, you know, general surface trees of our urban environments. So the system is already ready to be, you know, directly translated into the robot taxi market.

So we, you know, we are very confident that we can do a very quick entry with a, you know, with a product that can drive on highways, etc. That is something that will be, you know, something that the customers, your consumers really will like to use. So from that perspective, you know, this technology really is a massive differentiator that enables us to get us there. And then to back this up with a true commercialization, we have a massive partnership with Uber, where we are, you know, they have committed to deploy over 25,000 robot taxes from Wabi on the Uber network of the next couple of years, right? So if you fast forward, we definitely, there is a space for a second, moving beyond Waymo, and we are very excited to really lead the way at the scale. Right. Raquel, thanks for making the time. Congratulations, and best of luck.

Appreciate it. Raquel, or DeSan, Julia. Thank you as well. Very much for bringing us this list. All day today, you can be sure to catch CNBC Disruptor 18, which is Time Care and its CEO. They'll be on Fast Money later today. For the full list, go to CNBC.com slash Disruptor. And you know who's hosting Fast Money 5 p.m. tonight? Is it a tall, handsome TV person? Yes, but I'm also going to be there. We're looking forward to that. Aren't you nice? Coming up, the main.

Behind some buzzy recent media deals. Byron Allen will join us in studio. We'll talk the entire media and TV landscape. The whole thing. Stay tuned. The Weather Channel, BuzzFeed, and maybe stars. Your next guest is buying them all. And once more, his company has just taken over the late spot from CBS and Stephen Colbert.

The ratings are showing it is winning. Byron Allen went from TV star to TV mobile. He's got a grand vision for how you will watch TV and video for years to come. He is the founder and CEO of Allen Media Group. They own the Weather Channel. 813 local TV stations, probably 18 soon. And now BuzzFeed, Byron Allen, joining us now on set for a couple block. Abundanza, Byron Allen, great to have you on set.

Thank you. Well, thank you for having me. It's great to see you. So great to see you as well. So the question is you got Weather Channel. Look at, you got BuzzFeed, 13 local TV networks, stars. Got 11% there, I think. What is your grand vision?

What are you trying to build and why, based on where you see the media puck going? I'm building the world's biggest media company. That's what I'm doing. That's all. That's it. I started, when I met my wife, she asked me, Jennifer said, what are you doing? I said, I'm building the world's biggest media company. When I met her, 25 years ago, that's what I said. Listen, beautiful.

I hope you enjoy the ride because it's going to be one hell of a ride. So I started from my dining room table and calling people and asking them to carry all the TV stations and asking them to carry my television show, literally called all 1,200 of them and asked them to carry this once a week, one hour show entertainers with Byron Allen. And I got about 50 to 100 noes from each of them. And after I worked through about 50,000, 100,000 nose, I got 150 yeses, which was, yes, we will carry a once a week show. and then I proceeded to sell my seven minutes of commercial time inside of the show. You don't give up. You didn't give up. Okay.

And then after about, I kept smiling and dialing and smiling and smiling. And next thing you know, I had about 75 shows on the air and one of the largest privately held television libraries in the world. So I don't need to tell you. You live in L.A. So you're in the mix. Yes. You know this. Better than I, I'll tell our audience.

I'm not telling you because you know there are a lot more sellers in media than there are buyers. but you are a big buyer right now. Why? What value do you see in what we do? I'm getting great prices, tremendous prices, the best prices. As Walmart likes to say, everyday low prices. They've done okay. They've done okay.

I love the prices that I'm getting. You know, when I bought the Weather Channel from Bain and Blackstone and NBC Universal, I bought it at a 1.85 multiple, okay? I had the money back in my pocket in about a year and a half. I've owned it for eight years now. We've pushed the revenue up 40%, epa dot up, 30%, ad sales have gone up. I love buying from private equity, you know, because I'm an owner operator. I'm one of the few owner operators.

I understand their mistakes. I see what they're doing wrong. I'm able to buy it and make it more efficient, more profitable, quickly. So I love the prices that I'm seeing, and they are phenomenal prices. You know, media is going through this transition. I think that's a good thing. I'm told, right? As you know, better than anybody.

And I love the transition because the transition creates great opportunity, enormous opportunity. So that's what I'm taking full advantage of. And what I'm doing is I'm buying these assets and I'm having them work together in symphony. I like being, see, I don't think it's just either or. I don't think it's just linear cable or linear broadcasting or just digital. I think you have to be a part of the entire landscape. I'm very comfortable with our Big Four network affiliates. I've invested about a billion dollars buying ABC NBC, NBC CBS affiliates.

I've invested in buying cable networks. I've invested in digital properties. I'm invested in sports. That's a big growth area. I control 97% of all HBCU sports. So this is something that I believe it works together. What do you see that the sellers don't, especially the ones who are selling now, a lot of these businesses, they've been in these businesses for years. And we can all see the numbers. But is that decline arresting? I mean, is it getting, do you think there's a growth part of the store? I mean, when you say you're buying it cheap, you've got that margin. You can do well, you know, even with the viewership being what it is. But what do you see that they don't? Well, first of all, they are good businesses. You know, sometimes they are for sellers. Sometimes they have to sell things to get deals approved. You know, so when I invested a little over a billion dollars to buy Big Four Network affiliates,

I bought them from a terrific company. Great. They are phenomenal, but they needed to get their Quincy deal approved. And the government said, you need to go sell these stations to someone. So I sold them, they sold them to me. I just recently sold back some of them, and they're happy and I'm happy. Where were the other people who could have bought them when you did? I mean, why weren't these prices being way bit up? You know, I don't know. I know, you know, it's one of those things When you're running on the track, don't look behind you. Just go straight for the, just go straight for the line. Just win.

You know, I see great opportunities and I take advantage. I'm not publicly traded. I'm able to move quickly. I don't like to do the analysis paralysis. I think there's a lot of analysis paralysis at these companies. You know, I should not have been able to buy the weather channel at a 1.85 multiple. But there were a lot of companies, you know, one of the big issues, they said, well, Comcast has the ability to drop the weather channel. That's why people wouldn't step in and buy it. And I had to go,

okay, if you think Comcast will drop the weather channel, then you're a brain donor. Because the weather channel is not a cable network. It's a technology platform that has proprietary software at 3,500 cable head ends that allows us to tell you the temperature and the weather in your zip code. As a matter of fact, the weather channel is hardwired into the Penadon as a matter of national security. I love the retrocast, the retrocast, the Weather Channel, where it looks like what we saw as a kid very quickly. We got to go to break, but I want to, and then you're going to come back, but I want to ask you,

you put your hat in the ring for Paramount. Yes. Some people on Wall Street said, don't know if he has the money. You had the money. Yes. And you know that because you do know I had the money. Yes. You know, it was easy.

What people didn't realize about Paramount is they got a double downgrade. And once they got that double downgrade, 15. billion of debt became portable, assumable. And not only that, you could prime the debt and put billions in front of the $15 billion they had put on the company. So I had a line around the block

to put debt in front of the $15 billion that I could assume. And we had another private equity company, private equity firm, come in and say, here's $4 billion sale lease back of the real estate. And a streamer came along and said, we will take everything out of that library

site unseen for $10 billion. We had way more than enough to buy it. But that was Sherry Redstone's call. And I think Larry and David Ellison, I think they're terrific. I think they're going to do a phenomenal job. And she wanted to sell to them. There were a number of people she just didn't want to engage with. And I'm glad she sold it to them because it needed to transfer.

And I think they're going to do an amazing job. Byron, thanks for now. Don't go anywhere. We're going to circle back and continue the conversation. Brandon Gomez is standing by with the CNBC News Update. Brandon, what's the latest? Hey, Kelly. The U.S. reportedly seized an Iranian-linked oil tanker in the Indian Ocean. The Wall Street Journal reports the tanker was sanctioned by the U.S. back in March for transporting Iranian oil. The journal reports the ship was likely loaded with more than a million barrels of oil from Iran's Kard Island in February. Meanwhile, are reportedly opening a new criminal probe into former Venezuelan leader Nicholas Maduro. CBS News reports the new investigation came about over concerns the current case against him is weak. Maduro and his wife were captured by U.S. forces in January and currently face drug trafficking

and firearms charges. And the NAACP is calling on black student athletes to boycott southern colleges in the wake of last month's Supreme Court decision weakening the Voting the Voting Rights Act. The civil rights group is urging black recruits to withhold commitments from a list of schools primarily within the Southeastern conference. The SEC has yet to comment on the campaign. Kelly, I'll send things back to you. All right. Brandon, thanks. Up next, Byron Allen is here. he's staying and he'll be joined by a special guest on how they are going to use AI to change your viewing and reading habits. You're going to want to hear this.

All right, for years, BuzzFeed was all the buzz in media, but changing viewer and reader habits hit the company and the industry. But Byron Allen, who's still with us, sees huge value in the company and just bought it. He believes AI can help revive what and how we can sue media. We're joined again by Byron and Jonah Preddy co-founder of BuzzFeed and Huff Post. We'll work with Alan on the company's AI strategies, Jonah. Welcome into the conversation as well. What is AI going to do to change the way we consume information? What will it look like in a year or five years? The way I think of AI, it's a new computing platform. So every time there's a new computing platform, eventually media really changes a lot. So we had mobile as a new computing platform. And at first,

all of the focus was on the handsets and the operating system, but then eventually it was all about the apps. And now we have AI as a new computing platform. And so it's all about data centers and the models. But now these models are becoming really powerful and they allow us to do things in media that we wouldn't be able to do before. So for example, we used to have programmers that would make formats like quizzes and games and things like that. Now writers can make that themselves using cloud code and can generate all kinds of new interactive formats. So if they have a creative vision, they can make that creative vision real without even being an engineer. And Jonah, you perfectly captured kind of that era of apps and how to make all of that work.

But your time, but your BuzzFeed period is sun setting, is it not? And I just want to talk a little bit about it. But what do you see now as, or do we have to wait and find out what the next chapter of media on this AI platform looks like? Well, I think there's huge opportunity and we're seeing benefits already to the AI investment. And a lot of it is, is this new era where people, people are going direct to websites. So it used to be people would come through Facebook.

They would come through Google. That has declined a lot. First, Facebook declined a lot. Then Google declined a lot. That's true across the industry. But what we're seeing is that people don't love what, you know, doom scrolling all day on what is essentially like an ad network. If you go to Instagram, it's like the creators are making ads. The ads between the content is.

So true. And so people want to go directly to a place they can trust. Brands like HuffPost, brands like BuzzFeed. And they go there. And we can give them new experiences with this, with this AI. Because Byron, I'm curious where your involvement, you see BuzzFeed is still having value. I think you're now going to be the majority owner. And what's the value you see?

Well, you know, I invested a little over $150 million in an app called Local Now that uses artificial intelligence and proprietary software to curate, aggregate, and stream, super hyper local news, weather sports, and traffic, geo-fence to the user's zip code. Now, we won Best Streamer twice and no one cares. There's no branding, the brand, local now, right? But I said, this is an opportunity to take BuzzFeed and Huff posts and put them in the streaming business. The biggest streamer on the planet, YouTube, it's free and it's global. People love free and they love streaming.

So I said, okay, we're not going to change BuzzFeed. We're going to add on to the foundation. He has done a magnificent job, 20 years of branding, and now we're going to put them in the streaming business. So, you know, when I had dinner one night, would read Hastings with the founder of Netflix. I said, Reed, what keeps you up at night? He said YouTube. I said, YouTube. And he said, yes, Byron, if YouTube starts to deliver premium content, how do I get people to pay me X, Y, Z per month?

And I said, got it. I'm going to make it super hyper, local, and premium content for free. And a lot of the content sitting at YouTube, it's non-exclusive. So I say to the YouTube, you know, creators, the content that you put at YouTube, you're going to be able to put it at BuzzFeed and get not one check, but two checks. I think you read my mind because I want to follow up on the Google. Today is Google's Developers Day. It's called Google I.O.

And it's kind of a top story. Kelly Let her show with it just for this one. It's a big deal. They're getting now into sort of the AI chip market. We don't care about that here. But talk a little bit, Byron, about how powerful Alphabet Google is. It's 12 percent, I think, of the streaming market. YouTube. That's right.

Google, obviously, YouTube TV. Yes. Hugely powerful. Yes. What if they moved into like paid content? Then what? Is Google too powerful? No, no.

I mean, look, I'm a big believer in open the markets and the winners and the winners. Google, obviously, YouTube, they do a phenomenal job. And in full disclosure, I have a very large partnership with Google YouTube. It goes out about 10 years for cloud. and all types of services and helps us with ad support, you name it, AI. So we have a strong partnership with them. But listen, we live in a country where we have Coke and Pepsi. So you want to foster competition.

It's great for the whole country. It's great for the advertisers. So I think, look, they are doing a great job, but you have these creators who are saying, hey, where else can I go and monetize my content? BuzzFeed has a great brand, and we're now going to put them in the video streaming business, which they haven't been.

You're now going to be president of BuzzFeed AI, correct? Yep. And so again, are we to think about this as a video pivot or just? Well, I think AI is completely multimodal. So it goes across everything that an organization does. So we use AI in how we operate the business. We use AI in how we help creators make content. We'll use it for video, for text.

But we're going to use it where AI is a new medium. We're not just going to create the old content using AI. and put AI slop out on the internet, we're going to give tools to really creative people to do their best work using these new technologies. And I think that's the thing that is so exciting to me about this new chapter for us. And we wanted to have a vehicle

where people could invest in the intersection of content and streaming and AI, because right now that's not available for most. Gentlemen, thank you so much. Both of you for coming in and making the time. Can't wait to see what these next chapters bring. Thank you, Kelly. If AI can do something about the commute across the drive,

George Washington Bridge, we're golden. You're golden. I'll come visit you wherever you are. Thank you, Brian. Thank you, Brian Allen, Jonah Peretti. Coming up, Mackenzie Sagallo, sitting down with Google's chief AI architect from Google I. Stay tuned. Google's annual developer showcase is taking place this afternoon.

Mackenzie Sigalos has been reporting live with all the latest developments and announcements for us. Let's head back out there where she's standing by with an exclusive interview with Google's chief AI architect. McKenzie? Hey, Kelly. So I'm here with Cori Kavichulu, Google's chief AI architect, DeepMinds, CTO. And I really want to start with the Wall Street reaction to the announcements today. We're not seeing shares move much. I'm hearing that people read these more as minor upgrades instead of the generational leap that they were looking for.

But it was only 10 days ago that you eclipsed Nvidia in terms of market cap because of the confidence that the street has in your AI strategy. So talk to me about what investors are missing today. Thank you very much for having me. What you are seeing here today is what we have been developing for building intelligence, and these are the recent developments that we are sharing. And I think there are two main updates in terms of AI technology that we are bringing. First one is Gemini 3.5 Flash. It's making a huge jump in terms of capabilities in coding and agentic workflows.

And we are bringing that to developers by Integrity 2.0. It's our new agentic development platform. And then the other one is bringing Agentic World to use. users and consumers, as well as businesses via Gemini Spark. We also have Gemini Omni, which is a big step in world modeling, which is really critical on the path towards AGI, being able to not just understand our world, but be able to simulate it. And I want to talk about this focus on the consumer,

because you're doing that at a time where Open AI and Anthropic are very much targeting the enterprise buyer. And so talk to me, because on stage today, a lot of these new products that you're rolling out are free. So one, how can you afford this new AI strategy? but two, and perhaps more importantly, how do you think about the value of these billions of consumers over the long term? The capabilities that we are bringing with Gemini 3.5 Flash, combined with it is quality, but it's also its speed, right? Like what we are showing is Gemini 3.5 Flash is really a frontier-level capability model,

but it is four times more efficient, four times faster than comparable models. And with that, we are bringing that capability. to consumers across all of our surfaces, and especially with Gemini Spark, which is the exciting development for us. It's like an open claw equivalent. Is that overstating what it's capable of? What it is capable of is really this agentic world

enables all sorts of capabilities and opportunities for people to get help from agents. And being able to both push the technology, but bring that in a safe and secure way to users is what we are trying to achieve. And this idea of cost efficiency is something that's come up repeatedly today. You had Sundar Pichai talking about how

if companies converted 80% of their workloads to Gemini, they would save a billion dollars a year. So as you think about your model strategy, how much do you look to kind of capitalize on cost efficiency instead of prioritizing something like a generational leap that you might get with Gemini 4, which we did not get today? Even going back to Gemini 1, our strategy has always been pushed the frontier of AI research and capabilities,

but do it in a way that we define the Pareto frontier of performance and efficiency. And here, today we are again seeing the same thing with 3.5 flash. We are pushing that frontier for businesses and customers in a way that you get the frontier capability with that efficiency. Your team is rebuilding Siri for Apple. It's happening inside of your labs in London. Talk to me about how that effort is going and when we might expect to see the fruits of that labor. We have a good collaboration with Apple. We are working with them, but there's nothing more to share at this point.

Okay. There's a lot of that work happening in London? London and here together with the Apple teams. Now, Google already has a robust enterprise business. You see Thomas Currie and Cloud Chief really looking to sell more AI products through Google Cloud in the context of existing enterprise buyers. So you talked about some of the coding advances that came up today as part of the product releases. How do you divide your time thinking about how important the enterprise buyer is versus the consumer? Because it seems like there's such a focus for Google on the billions of consumers already across your different surfaces. I think it is important to think about when we define AI and what progress is to get the signal from both consumers and enterprises as well, because there are many different challenges across both surfaces. And we are always very excited to get that signal from across all surfaces.

that helps us a lot and bring that value back to both businesses. And that is part of why Flash is a critical part that we are bringing that frontier capability. And I'd be remiss not to ask you about the CAPEX budget because it's something that doubled this fiscal year, which had doubled from the year before that and the year before that. But as you talk about these cost efficiencies and what you can achieve with something like 3.5 Flash, do you see a world in which Google and Alphabet as a parent company can start paring back that budget, that you would reach a level of technological capability that you wouldn't have to spend so much money on CAPEX?

We see AI growing day over day. You can see the impact that it's having in the world. You can see businesses adapting AI more and more and consumers adopting more and more as well. So as part of this, we have a long-term investment in our hardware. You go back 10 years ago, we started AI hardware business back then, together with our AI research. So that investment is a long-term investment

and that's going to continue. Fantastic. Kari, we're going to leave it there. And Brian, I'm sending it back to you. All right. Great stuff. Out West, McKenzie. Thank you very much.

More power lunch right after this short break. Quick heads up. You don't want to miss this. CNBC exclusively sitting down with Jeff Bezos from his Blue Origin Rocket Factory at 8 a.m. Eastern tomorrow. Be sure to tune it. And very quickly, an RBI.

This one actually about sports, though not baseball. The Eastern Conference finals to the New York Knicks and the Cavaliers of Cleveland tipping off tonight. And if you want to go, you better get ready to pay up. The get-in price, Kelly, like the worst seat in the house, $549.49. Tick-pick says that makes it the most expensive Eastern Conference finals ever. It's that Manhattan premium. You're going to go? No. Who's your team? Do you have an NBA team?

Houston Rockets. Why? Because I know the owner. Thanks for watching, Power Lunch. Closing Bell Source right now.