On with Kara Swisher - Winners, Losers & WTF Moments: A Look Back at 2025’s Top Tech Stories

Episode Date: December 22, 2025

From an unenforced TikTok ban and a chatbot calling itself MechaHitler to mounting fears that we’re in an AI bubble, 2025 was another messy year for the tech industry. We watched billionaire CEOs fully align themselves with President Trump, Nvidia become the first $5 trillion company, and Elon Musk’s popularity tank, thanks to his DOGE antics (and yet he could still become the world’s first trillionaire).  Kara breaks down the biggest tech stories of 2025 with four journalists: Bill Cohan, a longtime financial journalist, author, and Puck co-founder who covers Wall Street; Casey Newton, founder and editor of the tech newsletter Platformer and host of The New York Times podcast “Hard Fork”; Joanna Stern, senior personal technology columnist for The Wall Street Journal and author of a forthcoming book about how she surrendered her life to A.I. for a year; and Charlie Warzel, staff writer at The Atlantic and host of the tech and culture podcast “Galaxy Brain.”  (Please note, this conversation was recorded before news broke that TikTok had signed a deal to spin off its U.S. business to a group of American investors, the Justice Department released a trove of documents tied to investigations into Jeffrey Epstein, and Waymo halted service in San Francisco because of power outages in the area.)  Questions? Comments? Email us at [e-mail] or find us on YouTube, Instagram, TikTok, Threads, and Bluesky @onwithkaraswisher. Learn more about your ad choices. Visit podcastchoices.com/adchoices

I am the only one among us probably that has done it. I mean, I'm not sure. How did it go? I had it on camera. I hated it. It made you feel isolated and alone and paranoid. And I understand him. You did it on camera? Yeah. Hi, everyone from the New York Magazine and the Box Media Podcast Network. This is on with Kara Swisher and I'm Kara Swisher. 2025 was dominated by stories about tech. We watched the industry's billionaire CEOs align themselves with President Donald Trump.

Chinese AI firm Deepseek released what looked like a major threat to U.S. competitors. And the Trump administration helped TikTok skirt a national ban on the app store. And that was just January. So we got four journalists with us today to talk through all the major tech stories of the year, from Grok's Mechah Hitler meltdown to the mounting fears that an AI bubble is about to bring down the global economy. Bill Cohen is a financial journalist, author, and Pup Co-founder who covers Wall Street. Casey Newton is the founder of the tech newsletter platformer, the co-host of the New York Times' podcast Hard Fork, and my former tenant. Joanna Stern is a senior personal technology columns for the Wall Street Journal.

She's writing a book about how she turned her whole life over to AI for a year. And Charlie Wurtsill is a staff writer at The Atlantic where he writes about tech and culture. He also hosts Galaxy Brain, a new podcast from the Atlantic, about the internet. We're going to look back at all the crazy things that happen in tech this year, and we'll also look forward to where those stories are headed in the new year. All right, let's get into my conversation with Bill, Casey, Joanna, and Charlie. It's going to be a wild ride, but also a fun one, so stick around. At Criminal, we've made it a tradition.

every December to dedicate an episode entirely to animals who are really going for it. And, Tony, what happened when you pulled over? Nothing out of the ordinary until I saw the cat on the back of the roof behind the luggage carrier. Listen to our fifth annual animals episode on Criminal, wherever you get your podcasts. It is all. Okay. Palooza, Bill, Casey, Joanna, and Charlie, thank you for coming on on. Thanks for having me on.

Good to be here. Thank you, Karen. So much has happened this year in tech. There's no way to get all the big stories. But if you each had to pick one word that best describe 2025 when it comes to tech, what would it be? Bill, I'll start with you, then Casey, Joanna, and Charlie. I'm in the sort of mind of 2025 was the year of a new hype cycle.

for tech. New hype cycle. Okay. Casey? I would say data centers. Data centers are at the heart of the AI buildout. The AI buildout is now on track to be bigger in terms of cost than the buildout of the entire U.S. interstate highway system. Okay. So I'm going with data center. Data center. Fascinating.

Joanna? I had hype. AI hype. AI hype. Okay. Charlie. Dojification of the United States government. Dojification.

even though it's sort of gone? Yeah. I mean, they screwed up a lot in a short amount of time. Yeah, they're very effective that way. So the biggest tech story, as we noticed, is AI, and it's also the biggest story in finance. And the question, of course, is there an AI bubble in the stock market or not at all?

And the effects on the economy bill, the Magnificent Seven make up more than a third of the S&P 500. So we saw a lot of volatility in the stock market based on how Wall Street was feeling about the AI bubble on any given day. But overall, 2025 was a banner year for tech giants. Invidia became the first company in history about at $5 trillion. Apple and Microsoft aren't far behind. So, Bill, what is happening here right now because of the amount of the stock market these companies take up? Yeah, I mean, personally, I think it's a big problem. I think it's a big risk that we've added to the

equity markets, and also there's huge amount of debt capital that is being used to finance the Casey's data centers, and also a debt that's being used to finance these AI companies. And they're all sort of getting new investments, many of them private at higher and higher valuations that absolutely just make no sense to me, Kara. I mean, I'm sure it's all great. I know it's going to be revolutionary and all. that. And Lord knows I use it pretty much every day, but I don't understand the valuations. So I suspect we're going to have some sort of shakeout like we had with Internet 1.0, where, you know,

the Internet itself, of course, is a great innovation and will last, you know, forever. But the valuations and the companies that were part of the first wave, there'll be a huge shakeout, I suspect, when it comes to this AI business. What's the shakeout look like? That was a huge decline, if you remember, and lots of deaths everywhere. Yeah, huge decline. Yeah, I mean, I think there'll be winners and losers, as there always are and there always should be. And, you know, I don't know who they'll be or what, but, I mean, you know, we're already beginning to see it to some extent. But they're not little companies that went out. These are the big ones. They're not little companies. Guess what, Karen? That doesn't mean just because they're big now that they can't tumble and fall. So it happens. all the time. Big, little, you know, when things go wrong, they go wrong.

Okay. So, Joanne, in January, the Chinese AI firm Deepseek released a free chatbot that rivaled the capabilities of its American rivals, but it cost a fraction of what firms like OpenAI spent to create ChatGBTGPT, if you believe, Deepseek. You tested Deepseek against ChatGPT and Claude as an AI work assistant, and you're writing a book about how you surrendered your life to AI for a year. You are indeed the real-life pluribus Carol. But based on your experience, what do the advances mean for the AI arms race that's propping up the economy right now? Well, I'd love to hear Casey's take on this too

because I think he's done some great writing on the open source world and what that future looks like. And I think that was the biggest fear back in January or February when it was, right? It was the end of January about how deep seek's going to come along. This is an open source model.

It has been trained at the, what, a fraction of the costs of all of these giant models. And yet, nothing changed. changed this year for the big AI companies in the US. And so I'm not sure that the deep seek moment was a huge moment other than a news headline that we could all freak out about China and push the fodder even further that the US versus China, this is the AI race. This is the AI war is like we've got to be China. We've got to be China. I think it was only talk about hype. It was it was the way we got to

hold up the China flag and say, look, look at China. It's even. Deepseek. Go, go, go. Casey, I mean, I'm interested in your take here because I love what you've written on this stuff. Thanks. I mean, you know, people were very concerned about deep seek, even though at the time it was not a frontier model, right? What was interesting about deep seek was that it was a little bit less behind than the Chinese models have bent up into this point. It is also true, though, that when NBC went out and talked to startup founders in November, what they found was there are a lot of startups that are adopting these free, open source Chinese models because they are free. And so I do think you're seeing that on the rise in Silicon Valley. At the same time, go to, you know, the Fortune 200 companies, ask them who they are using. It is not, for the most part, these Chinese open source models. There are real security risks in these open source models that I suspect we'll be learning a lot more about in the years to come, and they're behind the frontier. So if you want the best, typically, you're still going with an American company. I want to actually get back to data centers with you, the most, you called it the most iconic technology of 2025. Harvard professor Jason Furman projected that investments in computer, equipment and software account of more than 90% of economic growth in the first half of the year. It's either Envidia stock or this spending. How long can that spending last? And it looks like sometimes they're trading money the same $5 back and forth. How does it slow down? Because data center is becoming a huge political issue as we headed to the election year, both negative

and positive economic growth and also pollution. Absolutely. I think you're going to see the rise of a big NIMBY movement next year when it comes to data centers, you're already starting to see it, and if we'd learn anything about American politics over the last 20 years, it's that NIMB movements

are extraordinarily effective. But I think you have to ask yourself, why are these data centers being built? It is anticipation of demand. And here's the point that I feel like a lot of reporters who cover AI miss out on. The demand is real, right?

Anthropics start of the year under $1 billion in projection. Now they're projected to make $10 billion in revenue. So, you know, the data centers are being built in anticipation that those numbers are going to keep going up. And I haven't seen a single explanation for why we should expect that to slow down next year. So I do agree with everyone who says,

hey, there's a lot of really strange deals here. And it seems almost certain that some of these companies are going to flame out, but you have to remember the demand is real. How important are the political issues? I think they're going to be increasingly important next year. You know, you have this really interesting opportunity, I think, both for Democrats and Republicans, because Trump has thrown in his lot with the accelerationist Andresa Horowitz's wing of the party and has said, hey, you guys do whatever you want. I barely understand this stuff to begin with. So if you want to say, hey, we're going to create a moratorium on all state level AI regulation, sure, go nuts. That's a big problem if you're a state representative for Missouri or Arkansas and everyone in your community who you're expecting to vote for you hates the data center that's getting built in your backyard. So I'm really interested to see next year who in these midterms comes out and says, hey, we're going to oppose these things and we're going to try to chart a different. course for AI. So, Charlie, you wrote recently the American economy, quote, appears to be at the moment in a sort of benevolent hostage situation when it comes to tech spending. Let's say there isn't a bubble in these companies managed to live on what Casey was just talking about on these massive investments. That probably means huge job losses, too, which is always being discussed. Talk about the upside for the average American aside from the economy doesn't collapse. I mean, I think it's a little bit hard to see what the upside is outside of like, you know, productivity gains and, like, you know, using these tools to vibe code, your own apps, things like that. You know, AI is a great technology in theory for bosses, right?

Unleashing a whole bunch of productivity, unleashing, you know, whole kinds of, you know, industry changes, making companies leaner and more efficient, right? I think that for the average person, there's certainly tons of creative uses for these tools. There's ways to use them to streamline your life to, you know, imagine things differently, recommend new things, use them as internet search, create interesting, you know, mock-ups of stuff like that. But I think that this idea of the hostage negotiation for me feels like when I look at the amount of money that is being invested in all of this infrastructure, which is real,

which serves very real demand like Casey's saying, you have to think in order for this to pay out at the kind of scale of the investment that's going into this industry in general, you are going to need to see some kind of disruption, some kind of either, you know, super-intelligence or just, you know, things that are so good and so immediately plug-in-play useful to companies that it will cause, in some sense, a real economic disruption, whether that's jobs. Versus the small disruptions. Right. And I think it's very silly to say that, you know, this technology is not rewiring the world, right?

It certainly is right now. But I think that if you go out and listen to CEOs talk and things like that, the justification is still not exactly. there, right? They're still buying, but they don't know what. And you're planning for some kind of future. You're investing, this is where you want to put your money towards this stuff. But I think in terms of the massive disruption, we haven't quite seen that borne out yet in all of the data. And so I think that that's what's interesting, right? Like, is it going to justify and cause this disruption? Right. So 2025 is also defined by the ways the tech industry and the Trump administration got in bed with each other. I'm sorry for that

metaphor. It begins with CEOs lining up at the inauguration, but that was just the beginning. And earlier this month, Trump announced that he would allow NVIDIA to sell its second-best AI chip to China. In return, the government would take a 25% cut on the sales. He said the same deal would apply to other AI chip manufacturers like AMD and Intel. We should note the U.S. government is also Intel's largest shareholder now, thanks to the socialist Trump administration. Bill, is this the new normal for the tech industry profits over national security, a heavy dose of industrial policy, will the government gets a cut? I mean, and Trump has gone even further.

I mean, he said that this should be literally part of our industrial policy. What's wrong with the U.S. government taking stakes in American companies? We should do more of that, which strikes me as socialism, communism, I don't know what. I mean, it doesn't feel very American to me and feels very opportunistic. And, you know, they took a position in a mining. company. And, you know, he hypes it up, part of the hype cycle. There's so many different aspects of the hype cycle. But, you know, it hypes that up. The stock trades way up. People pile into it. Then the stock comes down. You know, core weave, you know, stock gets hyped up.

Stock trades way up. Stock trades way down. Small investors get hurt. I mean, over and over and over again. And he's just hyping these things. And who knows? I mean, I get. But, you know, I get emails every day from traders on the floor of the exchanges telling me that they don't understand the trading in these options they've been doing for 45 years. They don't understand. It sure looks like people are getting inside information and trading on this news that not very many people are getting and making fortunes. And they can't, for the life of them, explain it.

And, you know, all roads seem to lead back to Donald Trump and his desire to kind of manipulate the markets. And now we have, thanks to Nira Tendon at the Center for American Progress, there's sort of like a tracker now of which people should go to because it's really pretty interesting. Right. So the corruption in plain say, let me ask you very good. How do CEOs think of this? Besides, I got to do it for the job, but this is what we must do.

I mean, I think that they are, they're hating it, Kara. I think they really do not like the fact that they have to do this. But on the other hand, there's real politique involved in this. And, you know, if you want to get a merger through, I mean, since when, if we ever heard of a president putting his thumb on the scale of two buyers of potential Warner Brothers Discovery deal, I mean, it's happening almost every day with him is putting his thumb on the scale. And I think they absolutely hate it, but they feel like, you know, what choice do I really have? My first obligation is to my shareholders, my employees, and I can't get in a fight with the president of the United States. I think it's worth saying that in some ways they hated Biden more. Like, this is another. thing that I think gets missed. If you ran Amazon or Google or meta, Joe Biden was trying to break your company apart. He was trying to make sure that you could never make another acquisition, and he wanted you to spin out the most valuable parts into separate companies. I happen to think there were a lot of

good reasons to do that. But if you ran one of those companies, you really did not like Joe Biden. And so as hard as it can be for us to believe, Trump has been a breath of fresh air for people like Mark Zuckerberg. Well, they get what they want. I mean, it's not just a hands-on prose. It's political muscle just a few days ago, Trump signed an executive order, which I think is worth the paper to print it on, to block state regulations on AI. It's something that tech CEOs lobbied him hard on. It's reasonable not to want 50 different AI policies. I get it. I've said it over and over on pivot. But this is really about politics and stopping all regulation as far as I can tell. So I don't think it's going to work, but what is unregulated I mean for the U.S.?

I think they just don't want regulation at all on any level and just to freeze it in some fashion. That's right. You know, when you look at the state-level AI regulations had passed, this was pretty basic stuff. Like California has a law that says, hey, you have to have a plan for what you're going to do if your AI causes a catastrophe, and we want to create some protections for whistleblowers at your company. That's the sort of standard that I think we want our frontier AI labs to be able to meet. But you have, you know, David Sacks renting his garments and screaming saying that, you know, we're going to lose to China if states adopt similar measures to that. And, you know, Kara, as you know, the first few efforts to pass this moratorium failed, what finally got it through

was that after they heard a lot of complaints from Republicans, they came around and said, okay, you're still allowed to pass laws that have to do with child safety and maybe safety in general. Marsha Blackburn. Yeah. But the real question is, I mean, one, I'm not sure that this moratorium that they pass is even constitutional. But the other thing that they've said is that they're going to try to get some sort of framework from Congress. And it's like, good luck with that. We've been trying to get frameworks from Congress for a lot of tech-related issues over the past decade, and we haven't had a lot of success. One thing that they did

make a big deal of, and the stuff he was muskling around was TikTok. Remember TikTok? Suddenly, there was a ban, then there wasn't. Last year, Congress passed a lot of force. The sale of ban from app stores, President Biden signed it to law, and the Supreme Court unanimously upheld it this year. And then nothing happened, largely

because Trump intervened and did nothing. Possible investors include a lot of Trump allies. It's not going to be a Larry Ellison-owned TikTok. It's Oracle is one of the partners, I think Oracle have 15%. I mean, Oracle is Larry Ellison, but it will own 15% of this consortium, and there's others in it. What happened to the national security concerns and when TikTok was digital fentanyl? I think it goes to what Casey was just saying. Think about

where we were in the Biden administration. I mean, last year, we would have been sitting here wondering if Google was going to be broken up this year. That was like the real thing. We were like, Will Chrome have to be its own business? Will TikTok shudder? And we will, you know, oh, maybe someone will get the algorithm. Oh, the whole app is going to be dismantled. It's like, no, none of that happened this year. And I don't think any of it's going to happen.

I don't think there's any true fear that, you know, yeah, I mean, like, TikTok is digital fentanyl. I mean, I think you all agree on that. It doesn't love it. But, like, is there really any potential of this happening? Is there potential maybe what we've already had two, three years ago, which is the thought that stuff will be moved to U.S. servers. It will have some U.S. involvement, but still China's pretty much largely control of parts of this algorithm and part of this app.

They own under the amount. Go ahead, Casey. I really think this is one of the most important stories of the year because it was basically the first thing that happened in the Trump administration. And as you just said, Kara, this was a law that had been passed by Congress and upheld by the Supreme Court. And Trump just decided he was going to ignore it. write a little memo called an executive order and just say,

no, actually this ban doesn't apply, and he got away with it. How many more times did we see him run that same playbook in 2025 to the detriment of all of us? So, you know, this is one way... Like, the can kept getting kicked down the road. Like, oh, we're going to deal with it now. We're going to deal with it now. And that's what I'm like... But to be fair, Biden also ponded he could have let it go into effect for at least a few days.

And he didn't. Right. True. He waited. Go ahead. Bill. I'm sorry. Just to Casey's point about, I mean, yes, Trump has been... fabulous for crypto

entrepreneurs, CEOs, for tech CEOs, all those bros who put on suits and lined up behind him at the inauguration, who go to those dinners. Yeah, okay, they are richest creases, they're just getting richer and richer. And so, of course, they're loving Trump,

okay? And they're enabling his graft as well. But, you know, that's like a tiny subset of CEOs and companies in this country. The rest of them are feeling ignored. They're feeling put upon with tariffs. And, you know, they do not like the idea that they have to kiss up to this guy to make sure that he doesn't, you know, whimsically do something that is going to hurt them.

Well, the tech companies are getting what they want. What could lead to a breakup that he doesn't have influence anymore, right? Then they'll move along to whoever does, presumably. But you know them better than most of me. Yeah, they'll move on. Your buddy Elon has sort of moved off center stage, but he's still getting everything he wants, like a trillion-dollar pay package, SpaceX being worth $800 billion. Right, right. He is.

He's getting what you want. Now, you mentioned crypto-charlie. 2025 was the year of Trump meme coin. Reuters reported the Trump organization took in more than $800 million during just the first half of the year. It was a stunning display of corruption. I've called him a coin-operated president, Big coins, by the way. Anyone wants something you can buy influence if they have the money, yet the first half. public, do they care? You've written about how the crypto capitalized on a cynical culture that

distrusted institutions. You know, you see stories about the military trading crypto, although let's be clear, Bitcoin is really far down. Talk about the cynicism that's bled into the sort of corruption. Well, I will first say, I don't know if the crypto bros actually are that happy with Donald Trump. Right now, it is not the bonanza that was promised from the crypto president. Right. I think a lot of retail investors. are really hurt by that one. I mean, I think what was interesting

to me in the 2024 election was how we had for a long time said crypto is looking for a use case, right? It's technology that doesn't have this great use case. It's basically a speculative asset. But what it was for the people who got in early

who had these big investments is they basically used it in 2024 to buy a bunch of influence, to lobby, to, you know, they got rid of as a U.S. senator in Ohio. And also just to show Donald Trump, I mean, is there a piece of technology more right for Donald Trump, right? Here's a thing that's like basically you can do shady financial crimes with, right? It is an instrument for graft. And I, you know, it makes complete and total

sense that he saw this and said, hey, let's go and do this. But I think like World Liberty Financial, the Trump crypto, you know, company there that has made them, you know, billions. I don't think a lot of people really understand it. I feel like that's one of those stories of this year that if regular people really were able to grasp onto this, people who aren't paying attention to Twitter and sort of the tech stuff like we are, I think it would be a genuine outrage.

I mean, it's like if you look at some of the meme coin stuff that happened in those first couple of weeks or like right before the inauguration, retail investors got absolutely murdered by that. I mean, nobody made money except. for the Trump's and World Liberty Financial. And so I think that, you know, that is, it's really set the tone for what the rest of the year was going to be in terms of, you know, kickbacks and kissing up and enriching oneself.

Right. But, you know, it's crashing. If any of you, if it does crash or even kind of crash, will the Trump family feel the effects of that? How exposed are they? That's the great thing about running multiple graphs at what. You know, they sort of have a diversified portfolio of graphs. so if any one of them goes down,

they can always fall back on the other ones, you know? Yeah, that's how one invest. Diversity. Diversity, because that's wrong. We'll be back in a minute. Support for this show comes from LinkedIn. You know how important it is to hire the right person for your small business.

That's why LinkedIn Jobs is stepping things up with their new AI. assistant so you can feel confident you're finding top talent that you can't find anywhere else. The best part is those great candidates are already on LinkedIn. LinkedIn says that employees hired through LinkedIn are 30% more likely to stick around for at least a year compared to those hired through the leading competitor. That's a big deal when every hire counts. Hiring doesn't have to be complicated. And let's be real, when you have a business to run, you don't have the hours to spend on hiring. And you can't just wait around hoping the right person stumbles upon your job.

That's why LinkedIn Jobs's AI assistant suggests 25 great-fit candidates daily so you can invite them to apply and keep things moving. Hire right the first time. Post your job for free at LinkedIn.com slash Kara, then promote it to use LinkedIn Jobs's new AI assistant, making it easier and faster to find top candidates. That's LinkedIn.com slash Kara to post your job for free. Terms and conditions apply. Support for this show comes from Quince. Every now and then, you find a gift so good you almost always want to keep it for yourself. This holiday season, Quince has cold weather essentials that might tempt you to do exactly that. Quince is offering $50 Mongolian cashmere sweaters that look and feel like designer pieces, silk tops and skirts for any season, and must have down outerwear built to take on the cold.

Not to mention beautifully tailored Italian will coats that are soft to the touch and crafted to last for years. As always, Quince is able to offer quality that rivals high-end brands without the high-end markup. I've gotten a lot of quince stuff myself over the years, including a really comfortable down cape, which I look fantastic in, obviously soft pants and a wide variety of things that are really warm that I can layer on, a vast, things like that. I really enjoy the stuff I wear from Quince, and I wear it every day, and it's really good because it's not very costly. You can find gifts so good you want to keep them with Quince. You can go to quince.com slash Kara for free shipping on your orders and 365-day returns,

now available in Canada, too. That's Q-U-I-N-C-E.com slash Kara to get free shipping and 365-day returns. Quince.com slash Kara. If you're looking for the perfect holiday gift and you want to give something more thoughtful than another gadget or pair of socks, here's my suggestion, a subscription to New York Magazine. I've been part of New York Magazine for a while now, and I can tell you it's some of the best journalism out there.

From AI in classrooms to the future of media, New York Magazine digs into the stories, ideas, and people's shaping culture today. And right now, when you subscribe or give an annual subscription, you'll get a free New York Magazine Beanie. New York Magazine is the gift that informs, entertains, and keeps on giving all year long. head to NYMag.com

slash gift to get started. So let's get to our favorite character this year in tech, I think. It has to be Elon Musk. It kept on giving in 2025. Bill, the year started with Musk taking over Doe. She threw USAID into the wood chipper. That was very kind of him to say it that way.

He made what appeared to be a Nazi salute on... No, it didn't... Why are we saying it appeared? He made a Nazi salute on stage, not once but twice. He proposed to slash trillions of dollars, in federal spending. Tens of thousands of federal workers lost their jobs. He compared them worse than Hitler and Stalin, cruelty on all ends a bit. In the end, he only managed to get a fraction of the spending, he promised, if that, and we're probably never getting our doge

checks. He might have actually cost money, and the first buddy clashed openly with President Trump and sort of launched the refocus on Epstein. He attacked Trump over that, over the handling of the files and made a U-turn to get back in his good graces. We're have Musk's events. We're Have Musk's adventures left him from a business perspective, Bill? What do you think it is? Because Tesla stock is raging right now for some reason, even though the car sales are declining. Well, we live in an era of the cult of the billionaire, Kara. I think you early on predicted that he and Trump would have a clash of egos, which of course they did. But he's come out, you know, just getting richer and richer and richer. I mean, the obscene trillion-dollar pay package

that he managed to, well, was it really a surprise, get out of his very pliant shareholders? I mean, it's grotesque in every way. I think I wrote a time's op-ed about that. And, you know, now SpaceX, which, you know, actually he's probably had a bit of a more hands-off approach about, you know, is actually run by somebody who knows what they're doing. Now he's going to reap the benefit of, you know,

a huge valuation and a huge IPO. The banks are, of course, lining up. I mean, Larry Ellison, you know, his Oracle spaceship has sort of come down to Earth a little bit, and he's, you know, struggling to prove to the Warner Brothers Discovery Board that his equity is real in the Paramount deal, which is a fascinating failure to communicate, it seems to me. You know, he's come down to Earth, but Elon is like getting ready to go to Mars, at least financially.

I mean, a guy gets richer and richer and richer. In July, Musk's AI chatbot Grok started calling itself Hitler and Mecca Hitler, and it went on an anti-Semitic tirade. Casey, on Hard Fork, you called Grok both a joke and a warning. Explain what you mean by that and what did it reveal about how difficult is to control these chatbox, even when you're the richest man on Earth. Yeah, I mean, the joke of it is that while, of course, you know, Mecha Hitler, Grok said many offensive things, as did white genocide grok, another Grok character from earlier this year, a lot of the output I did find just sort of comical, right? This is a chatbot that is trained on the internet. Most of these chatbots, when you do it and you try to make them helpful, they do come across

as kind of a little bit liberal, and that keeps infuriating Musk. And so he goes in, and he tries to adjust them and rewrite them and make them sort of, you know, more anti-woke. And this has reliably turned them into Hitler. I do think that's funny, unfortunately. Now, the, The warning, though, is that it speaks to how difficult these things are to wrangle. And if you assume that these models are going to become more powerful over time, as I do, the warning there is you want to figure out how to align them to what we actually want them to do. So I do think we will look back on the Mecca-Hitler era of Grok as a moment when more folks should have started paying attention to AI safety. Although they're way behind, presumably.

Go ahead, Joanne. Yeah, I would say, like, X-AI seems to be way behind. But I think what we've seen from Tesla, and you know, you'll hear all the Tesla bulls talking about how they're the biggest AI company. And that is because they could arguably be the biggest physical AI company when you do think about where they are in self-driving. And that's likely where the stock is up right now, right? Is that, you know, okay, he's promised you can turn every Tesla into a robo taxi. No other car company can do that. And so you've got them on that.

You've got whatever is happening with these optimist robots, which are clearly far away, but they are putting a lot of effort in towards. And I've spent a lot of my year talking to companies about robots. We're going to talk about robots. Yeah. I mean, the hype is exactly there. It's hype, but he's encouraged people. And so I think the, like, the AI focus for Musk is really in Tesla, even though he may not be spending much of his time there. He's also used AI this year.

Don't forget, this was the year that he somehow. pulled X or Twitter, whatever we're calling it, Twix, off the edge of going bankrupt because the banks couldn't sell the 13 billion of debt. Next thing, you know, he's calling it an AI company. He's merging it with XAI and the debt flies out the door and the valuation is going up. I mean, he's Mr. Hype cycle on the AI front, both in Tesla and in X. So he's pulled X back from the brink. They'll merge Tesla into that, too.

And for what it's worth, when I talk to people who work at the leading labs, they do not count out X-A-I, right? Like, they do basically count out meta at this point. But when it comes to X-A-I, people say, like, that guy still has the infrastructure. He has good engineers, and there is still enough that is unclear about how to get sort of all the way to superintelligence

that they still think that Elon has a shot. So, Joanne, in November X, rolled out this location feature to help users verify the authenticity of some content. It revealed some popular mega accounts that claimed to be run by proud Americans are actually based in places like Eastern Europe, Island, Nigerian, Bangladesh. It's not exactly shocking. It's certainly not unique to X, but talk about the takeaway from that. Did it have any impact? Was anybody surprised, one, and two,

so what? It's funny because I did not realize that was just like a few weeks ago. It felt like a year ago that that happened. I know. So much of what we've talked about right now feels like it happens like two years ago. The bigger story, I didn't follow specifically so much on what happened there with the geographic location. I guess it wasn't shocking to many people that, like, guess what, these accounts say something that's just completely untrue. I mean, that's the story of the platform, really. I think what was interesting about X's years is how many different product things they were trying, especially around, you know, on either sex bot. And like, they just were, like, throwing random things into the product all the time. It's also the year that Linda Yaccarino

has left, and there seems to have been the real changeover in terms of, like, product change. Yeah. Gosh, I miss her. Me too. I forgot about Linda. Bring her back. I had a real takeaway from the location thing, which is more and more of the time we are, I think, coming to grips or trying to suss out, like, how much of the internet is just, like, absolutely synthetic or fake or not what it says it is, or, you know, tricking people who aren't paying as close of attention. And I think that something that has come up multiple times this year along this same vein, right?

whether it's the cracker barrel thing or the Sydney-Sweeney-Genes controversy or whatever with these internet-inflated, social media-inflated discourses is that these platforms are really inflating culture wars in interesting ways.

It's not that we're not divided, that we're not polarized, that we're not arguing with each other in the ways that we are and that these fights aren't actually playing on the social media that are real

and have some kind of meaning and political jockeying. But there is a lot of these controversies, these little things that are bubbling up synthetically that everyone, news media on down, is taking the bait from, right?

And there's this whole economy that people play into of, you know, aggregating and moving it up there. It's sort of snakes on a plane, but you called X a worthless poisoned hall of mirrors. I believe I call it a Nazi porn bar in the wake of that. Yeah, I've said that too.

So is that, so it doesn't matter is what you're saying. No, it absolutely matters because, like, X is going to be the place until the people who are obsessed with it and stay on it all the time, a lot of people whom are political elites, cultural, you know, influencer-type people, tech elites, the people like, you know, building all of the, you know, the AI tools that we've got who are, you know, have their own niche communities like media, Twitter did. Like, all of this, it matters because this is the place where those conversations are still

happening. There's still a vibrancy to it, and that is part of this problem. right? Like, I mean, I would urge a lot of people, a lot of journalists to get off there. I see a lot of people cooking their brains on there. And, like, people who I used to talk a lot with and who I think are probably decent people who are just talking now with, like, race scientists and being like, hey, that's a super fascinating, interesting blog post. And you're like, maybe I'd get off there. But it still has this outsized influence in our discourse in sort of helping, you know, assign what is popular. I'll say I don't post a lot anymore. I don't post a ton, but the conversation happening around tech elites, happening around new releases, I read it a lot and I keep up with it there. And also the algorithm's just really good. I mean, the algorithm really will just give you, it's just comparatively to the other things. It gives you what you want. And whether that's good or bad, I mean, my stuff is very innocuous. It's, I can say I'm on your show. It's tech news and, like, lesbian shows. It's great for me. Yeah. I find the others getting better and better, though.

I find them getting just as fine. And there's the mental health of not being called a cunt every five minutes like I do over on Twitter. By the owner hasn't done that yet, but he's thinking it. Casey, are you still on it? I mean, I have my account, but I haven't posted there in more than two years for all the reasons that Charlie outlines. And, yeah, I just applaud everything he said. So, Bill, getting back to what you were saying, this chance of becoming the world's first trillion, let's be clear, he doesn't get the money unless he delivers, right?

Although they could change the parameters in any time. He's got to make that an $8 trillion company. He's got to deliver. Right. But he's proven he has. So why did the stink of Doge come off him because he does other things? Do you think it put any stink on him? Because he still seems to have an impact on the tech industry and not American culture more broadly,

because he's quieted down in that area. You know, I think Doge just turned out to be, you know, sort of this sort of PR flop. I mean, it was just all, again, hype. It was all like, oh, Trump, of course Trump and Elon are hanging out together. You know, two guys who deserve each other. And, you know, I'll bring my kid to the Oval Office and he'll, you know, make a scene. And, you know, and it was so obvious that Trump was going to get tired of this guy trying to steal the Trump limelight because he can't share it with anyone ever.

And we know that. So to me, it was just like this phase. it had to be gotten through, and it ended up being quicker than I thought, and much less effective than I ever thought it would be. I mean, how many times in my lifetime have we tried to, you know, talked about cutting the budget and actually trying to balance budgets and be fiscally responsible, you know, and it never happens. Even the times that it happened, they reject it because they don't want to do it, and you're going to have this guy come in with a chainsaw and do it, who's erratic and on ketamine, apparently, according to Susie Wiles. So,

I say, you know, it was just a matter of something we had to get through. Get through. It was relatively quick. Each of you, what do you think the impact he has? First you, Joanne and then Charlie in case. Like long-term impact of Doge? Yeah. Or just of his...

No, of him. Where is he as we end the year? Because he failed at Doge. He called it a failure himself. He failed at Doge, and I think, look, I look at the Tesla sales and what happened there. I mean, there were all of those protests in front of Tesla. There was damage to the Tesla brand. There is absolutely no doubt there was damage to the Tesla brand.

I think that's the thing I hear about most, like, in the personal tech and automotive space is, like, I'm not buying a Tesla because I don't stand for him, even though it's the best car and truly, like, the most affordable EV you can get in America. I see it two ways. I see the first is that it was, I mean, I think he got a ton of egg on his face or, like, reputation or whatever. I mean, he looked like a complete and total boob. He failed. You know, it was sort of this idea of like if you apply the Silicon Valley like, you know, move, fast, break, everything, whatever, you know, cliche framework to this really complex institution, you're going to fuck it up big time, right? And it's going to be this huge problem. There's real lives at stake. A lot of real lives like around the world with USAID and all that stuff. I think the second thing, though, is that this was a rerun of the Twitter playbook, right? Come into this thing.

And, you know, basically say this doesn't work the right way, strip it for parts, turn it into a political weapon, right? And that, I think, he still gets a lot of credit among the right wing of the party for doing that. That is still a thing that a lot of tech bosses, frankly, I think look on and say, like, I want to run some version of that playbook. Maybe I don't want to screw it up the way he did, but I still think that that's innovative. So, on the whole, going up, Casey? He's gotten me really interested in ketamine as a performance-enhancing drug. You know, at some point, you think it would have cost him. But it's like the way that this guy has thrived since he made that a part of his lifestyle, potentially the world's first trillionaire. Suffice to say, I'm interested. You're interested.

You really got to keep your dosage good, though. Yeah. And it seems like, you know, it's kept the Elon birth rate up, too. So, you know. All right. I am the only one among us probably that has done ketamine. I'm not sure. How did it go?

I had it on camera. I hated it. It made you feel isolated and alone and paranoid. You did it on camera? Yeah. Well, that's not the fun way to do it. Where do we watch that? No, it was still ketamine, my friend.

It was still catamine. You got to be doing that late night. You got to take it to the club. That's Kara Swisher's jam. Where do we watch that? I'll send it to you. It's really actually very funny. It's very funny.

I didn't like it. I didn't like it. I found it boring. I was like, oh, are we done? But I'm not a big drug person. Marijuana, paranoia kind of thing? No, you're just alone in a dark place and floating around. It's stupid.

It's just stupid. I'm like, can I go watch Law & Order now? I mean, really, that's what I thought in the middle of it's like. I'm really bored. I'm tired of floating around. I don't have any big thoughts. I need to move along. We'll be back in a minute.

If you were a person online, around the turn of the century, there's a good chance that AOL Instant Messenger was crucially important to your social life. For a whole generation of us, it's how we learned to LOL and LMAO and JK and BRB. It's how we learn to be people online. And this week on version history,

a new chat show about old technology, we're talking about the whole story of AIM, from its skunk works era inside of AOL, to its very sad death at the hands of text messages. All that on Virgin History, wherever you get podcasts. Support for the show comes from New York Magazine's The Strategist. The Strategist helps people who want to shop the Internet smartly.

Its editors are reporters, testers, and obsessives. You can think of them as your shopaholic friends who carry equally about function, value, innovation, and good taste. And their new feature, The Gift Scout, takes the best of their reporting and recommendations and uses it to surface gifts for the most hard to shop for people on your list. All you have to do is type in a description of that person, like your parent who swears they don't want anything, or your brother-in-law who's a tech junkie,

or your niece with a sweet tooth. And the Gift Scout was scanned through all of the products they've written about and come up with some relevant suggestions. The more specific you make your requests, the better. Even down to the age range, every single product you'll see is something they've written about, so you can be confident that your gift has a strategist's still of approval. Visit the strategist.com slash gift scout to try.

it out yourself. So let's look ahead to 2026. Let's finish up, talk a lot about the grab bag of big stories that are coming. They're going to see, including with Elon, some huge IPOs, SpaceX is among them, also open-a-anthropic. Bill, first, what are you going to be looking at as these companies prepare to go public? Because they've stayed private for a long time, which is their preferred place of being.

Well, again, we're getting back to what we first talked to out of the beginning. hype cycle. I mean, IPOs are a perfect product for the hype cycle. And, you know, these things come and go. They're a little rollercoastery. And, you know, Wall Street underwriting of IPOs has been kind of cool for the last three or four years. I guess maybe because everybody in tech hated Joe Biden. I wasn't aware of how much they hated Joe Biden until this conversation because they weren't getting everything they possibly wanted and more. But now we're in the big-time IPO hype cycle, and Kara, nobody is better at hyping up and getting people excited about overvalued companies than the Wall Street underwriting machine. They get paid for it.

They're really good at it. They favor their institutional investors, and it's really a bonanza. And so the floodgates of all that are just going to open. And so that's dead for the year now. Which one do you think is the most promising? So SpaceX, I assume. Well, I think Starlink is a real thing. You know, a real thing that has helped people around the world connect to the Internet.

And I don't know if you've ever seen. I was up at my house upstate once and saw the whole Starlink constellation go by the night sky. It was unbelievable. It was riveting. I didn't know what the heck it was. Yeah, he was very early to that. So is that the one you think is, what about the old? Open AIs and Anthropics or Strips?

Open AI, you know, is it losing to Gemini 3? I mean, or Anthro-I mean, SpaceX is clearly unique and is clearly being hyped. And it's going to, I mean, that could be the first trillion-dollar IPO. Right. 1.5, that's what they're saying. Anybody else on where this is going? Which ones are you paying attention to them? I mean, look, I think everyone's really excited to see these S-1s, right?

Like, everybody wants to see the actual financials of these company. We want to see all of the debts. We want to see all of the risks. that they're pricing in. We want to see the real revenue number. So I think we're just going to get a lot of information from these IPOs that I think might help us understand the actual state of our economy a little bit better.

A little better, okay. And one thing about Anthropic that I just love this part of it is that, you know, San Bangman-Fried, invested in that and, you know, everybody thought he, you know, mishandled his financial, lost money, he spent 25 years in jail. And that thing, when it goes public, I mean, they've obviously sold their stake by now, but had they held it, it would have been a bonanza, and he was obviously right about that.

He was right about that one, yeah. Yep. Got to hand it to him. So we're waiting also to see what happens at Netflix-Pro's acquisition of Warner Brothers and HBO. Paramount Skydance is mounting a hostile takeover bid for all of Warner Brothers discovery, including CNN, though this recording, Warner Brothers is told its shareholders a rejection. Paramount is run by David Ellison with a backing his father, Oracle's Larry Ellison, Big Daddy. Both are allies of Trumps, who says he gets involved in the deal, but he's kind of been dissing them recently. This is important because they've been promising to techify media, but this sort of ownership by tech companies, Charlie, how do you look at that?

And I suspect Disney will get sold, possibly to Apple, possibly to Amazon. I'm just hoping that there's one company we can all just pay attention to, right? Just one. Just the mega, the mega blob. I think what's really interesting is when you listen to people, like, creatives or people kind of like, you know, high up in the entertainment industry talk about, you know, the potential, especially with Netflix winning out. And just how it's a lose, lose, lose, lose, lose, lose, lose, lose, lose, no matter who wins. Genuinely, nobody wins. And even the people who win, other than, you know, the top, top, top folks, those Sarandos, you know, even the winners start to.

lose in the end, right? Because, and I just, I feel like, I kind of get the, the, when I think too hard about this, I get the thousand yards stare of just like, you know, we started out with three DVDs in the envelope, you know, and it was like a pretty delightful thing. And then we got some other, some really delightful, you know, tech-enabled, like, streaming stuff. Oh, my gosh, we could binge this stuff now. This is so amazing. And you just kind of fast forward with this, like, rapacious style of like, we must scale, we must do this. You can never. rest on your laurels, you must keep going. And then it's like, well, maybe we won't have movies in movie theaters anymore, right? Maybe there, maybe there's just like, won't be a lot of

people working on movie sets. I'm sorry, isn't that up to the consumer? If they don't want to go, it's not Netflix fault that they don't. I'm sorry. I just, that one, when's the last time a lot of these Hollywood people were in movie theaters, except for Quentin Tarantino, I guess. I just sometimes, I don't know about that part. Consumers don't like it. Consumers don't like theaters. I don't think it's that. I mean, I think that there's a lot wrong with the movie theater business model with the industry. Like, it is way too expensive to take your kids and some other folks out for, you know, for a night at the movies.

That is a problem, obviously. But I think it's going too far to say that, like, that experience, we voted on it as a people. I mean, playing to convenience, I think always wins out because people. Yes, but I don't think, I think we did vote on it. But anyway, Bill, you've been writing a lot about the acquisition. What are you? You and I text a lot about it. And I thought Paramount was in much more, because they're naive and David is not as smart as he needs to be.

So what were you hearing as it's headed into the next year? I would be shocked if Paramount didn't raise their bid. And the bids are too ambiguously close at the moment. And so that gives the Warner Brothers Discovery Board all the business judgment that they need to go with Netflix. But if the Ellison's decide to really go for it. and I think if they probably will, they've already said their last offer was not best and final. I think if they make it clear that it's unambiguous that they have the highest bid, I think they'll win, and we'll know that sometime probably in the next few, in the next month or so.

Yeah, so that would be $34, very clean, guaranteed. Well, $34 was my, right now is $30. Yeah, if they went to $34, all cash very clean, I think it's all. over because I think Netflix can walk away with their breakup fee and some sort of content deal from, you know, HBO Max, et cetera, and be very, very happy. So the question is, are they going to raise their bid? There's this continuous disconnect, which I really don't understand, that the WBD people think Larry Ellison really isn't on the hook for all the equity, and they keep saying, he is on the hook. What more do we need to do to convince you?

that he's on the hook. I don't know how they're going to bridge this divide, frankly. I mean, they could come out and say, you know, here's my blood on a guarantee, and isn't that good enough? But I think in the end, they will prevail because their pockets are deeper and Netflix is going to have to drop

because their stock has been going down and they can declare victory and get what they want. The only outer thing is if Big Daddy decides, oh, what the hell? He's so smart. It seems dumb. This seems dumb. Yeah, dumb like a fox, Kara.

Oh, I don't know where he wins on this time. World Second Richard's Man. I get it, but on this one, I don't see him winning. There's no money. Anyway, Joanna, speaking of robots, one of your stories that got a lot of attention this year was how you tested a live-in humanoid robot 1X Neo. Pre-eaters are open. The cost is $20,000, and delivery is expected sometime in 2026.

In your story about Neo, you wrote the home is a trickier environment for robots, even trickier than factories and city streets. So what will you be looking for in this area of home robotics? Will we all have something like the Jetsons Rosie robot in our home by the end of 2026? Well, I think we're going to, the hype for humanoid robots and home humanoid robots, personal humanoid robots, is going to continue next year. And so, yes, that company 1X with their Neo robot has said they're going to start beta testing in homes. I have cleverly signed up. And so we will see if that happens.

and we will see if my year ends in divorce. And so they're one company. Another company, they're called Sunday Robotics. They're going to be launching a small beta program. And so where we're in in this moment or robotics is that these companies need tons of data. They cannot make these robots do things that we want in the home, fold your laundry, empty your dishwasher. Unless they have them in your home. Unless they have some in homes and these robots are doing the things, right?

And so went really viral about that. One next thing was like, the way that the robots have to be in your home is actually a human has to operate it back in their headquarters. And so it's really a puppet, right? So the puppet comes into your home. It's learning and doing these things in your home. And it's also like looking at you and all of these things. And it's crazy. It's just like a totally crazy dystopian idea.

But also these companies are very convinced that people will do it. And I think it's really, it's like it's crazy. It's fun. And we're going to see what happens. really when some of these get into these beta tester homes next year. And I expect there will be some crazy headlines. I hope I write them. Doing things badly is not really something you want in your home. That's my feeling. So Casey, this month, Australia became the first country to ban social media for kids, in this case for users younger than 16. Already, we're seeing other countries look at

similar legislation. You said Australia's ban has been the moment you realize quite all the bad choices. These tech companies made over the past decade have caught up with them, but the tech companies are going to fight it. There are some people who think it's a Mike Massey. is one of them who think it's a problem. Where do you see this headed? I just think a continuing of it. But what are your thoughts? I agree with you. I think this was the first domino to fall. Already, Malaysia has said that they're going to adopt something similar. Denmark and Norway are not far behind. So I predict that a year from now, a ban on social media for under 16s is actually going to be the norm. I don't think we will get there in the United States due to the

First Amendment, although there are many states that are trying. I mean, look, these companies had well over a decade to make these products safer children, and they still aren't. And every time that we get leaks out of these companies about what the workers who are working on this stuff actually say, is that they tried to make them safer and were rebuffed at almost every turn. So I just think that this is an argument that the social media companies have lost, and democracies just aren't willing to put up with it anymore. Yeah, I would agree. Also this month, Charlie, Disney announced it's investing $1 billion in Open AI and doing a licensing deal, its characters, the company's video platform, SORA, starting next year. It makes Disney the first major

Hollywood company to license its contact on an AI platform. Casey's written about how A.A. is winning in the copyright fight in music. Charlie, you've written about generative AI platforms like SORA, tools to crush creativity more than anything. They incentivize the creation of Slop. So I've talked to the Disney executive. They feel like they have to figure this out, and they did a short-term deal. I think it is the short-term deal. What do you make of this, what's happening here? And could you see a public backlash to AI Slop or more deals like this? I think there's going to be a big dance here between people who see these things as fun and silly and throwaway things

and people who are generally really worried, upset and frustrated by the copyright issues therein, right? Like people who make their living on this stuff. There's just so many different artists. and folks who are really mad about this, and also who are getting it from so many different angles, right? Like, you have a lot of artists, you know, clashing with big platforms like Spotify who feel like they've been taken advantage of forever

or, you know, who are having copies of their work come out and things like that against their best wishes. I think that this is all going to take a long time to really iron out here. Like, I don't think that there's like a really long-term vision with like the Sora style stuff. I think maybe some people will want to outsource their avatars if they're, like, very famous and there's some, you know, good reason for that. But I think for most people, it feels really to me like this flashy proof of concept kind of thing, right?

That it's just trying to continue the hype of these products are really good at generating synthetic, lifelike quality, whatever things. They're going to bring people like Disney on board and stuff. But, I mean, I really don't know. It's a brand safety nightmare. Like, I've already been watching so many videos. of Toy Story characters trapped in the Twin Towers on 9-11 on X

that have 91 million views. You know, stuff like that where you're just like, all right, I mean, Thunderdome is here. Yeah, it's pretty gross. It is a little hypey, but it's definitely brand safety

is a real issue, but they also can't avoid it at the same time. Okay, last question to wrap up in just a sentence or two, give us your top tech prediction for 2026. We'll start with Bill,

then Joanna, Casey, and with Charlie. bill i guess uh my top uh prediction would be that these tech bros uh are going to fall out of favor finally fall out of favor with who with all of us with with the american public with the you know maga crowd in washington with politicians i think they've just pushed things way too far right right because we've had far too much of them how can we miss them if they won't go way. Joanna? Can I have two? Yeah, go ahead. All right. Well, I think the one that we haven't talked about, and I think it's naturally going to happen, is that Apple will release a foldable iPhone. It'll be $2,500, and people will finally a foldable iPhone. It'll be fine. People will be fine with it. That's my team. Are you going to buy it? Will I buy it? I'm going to buy it.

Yeah, I'm going to buy it. Yeah, I'm going to buy it. Yeah, I mean, it's my job. I'm going to buy it. But it probably also is fine. And then I would say, you know, I've really been thinking a lot about Waymo, and we didn't talk a lot about it here, but I really think next year's a huge year for Waymo. And I hate to, like, say it in this way. And I talk a lot about this in my book, but something bad is going to happen with Waymo.

It's just eventually going to happen. And I think maybe next year, you know, just the odds are bad, meaning a crash. Yeah, like Waymo will get in a. a crash, and there will be a crash, and it will make a lot of news, even though there's been so much real data to show the real safety around it. But the more Waymo gets into these markets, they're going to launch in, like, five more markets next year, and some of them are going to be East Coast markets where, you know, they haven't typically been before.

Yeah, that's one of my predictions. Wow. Okay. I know, it's sad. I don't want to predict that something bad is going to happen, and I hope it's, you know, I hope it's not a cat. A foldable iPhone and dead people. Yeah, there we go.

Let's sum it up. Casey. I'll also go with two. One, I think Joanna's household robot is going to quit due to the toxic work environment in the home. So that's sort of prediction one. And then prediction to... My home? My personal home?

Yeah. Toxic environment in my home? Okay. That's right. Yeah, okay. Great. All right. So we're on the same page.

And number two, I think, like, I just don't think that the AI bubble is going to pop in a meaningful way. I think that the leading frontier last... are going to be making a lot more money at this time next year than they were this year, and that while we may see a flame out here or there, if you're counting on 2026 being the year where the whole AI thing

goes away in Silicon Valley moves on, I think that is absolutely not going to happen. All right, good one. Charlie? I think there's going to be two backlashes that bubble. I don't think they're going to completely foment and go over, but I think one is just like a scrolling

and phone backlash. I think we're seeing more phone-free spaces. I think we're seeing more concerts and things like that. And I just think that there's a like there's a lot of memes going around in like Gen Z and under spaces of just like scrolling's cringe, man, put it away. Like we just, we're tired of this, we're tired of all that. And I also think there could be potentially if there is any economic trouble related to AI. I think there could be like an AI populist sort of backlash, especially if there's any indications from the government that we're going to help, you know, maybe backstop this, that it's in national security interests to make sure these companies can just do whatever they want. And I think that there

could be a political sort of leap onto that from a populist lens. All right. I'm going to ask one more, actually, very quick. Key person in 2026 in tech or tech finance. It doesn't matter. Let's go backwards. Charlie, just one name. You can say why very briefly. And then Casey and then Joanna and Bill. Is it cliche or wrong to say Jeffrey Epstein maybe? You know, we're probably going to get, we're going to get the files maybe, hopefully. I don't know. Jeffrey Epstein. The dead pedophile for the win. Okay. Casey. Fiji Simo. Okay. Fiji Simo, the CEO of Apps at OpenAI. She has a lot of responsibilities that used to belong to Sam Altman. I think she's brought a lot of Facebook energy into the company, which can be good for growth and bad for other things. So what does Fiji do with

the apps under her control, I think is a huge question how you get it did next year? Nice one. Okay, good. All right, Joanna? I'll keep going on the Apple train, and I'll say Tim Cook. I don't believe Tim Cook will step down next year, but I think that it will continue to dominate headlines what's happening internally at Apple and how they fight the AI. Well, they're being behind in AI, and I think we'll hear a lot about more of the moves he's trying to make to move the company in that way and improve Siri and do the things he needs to do. God, if I were him, I'd quit and go home. He's done a good job.

You think he's going to do it next year, though? I don't. He probably feels frustrated now, just like Eiger. He should leave because he did a great job. I'd take a bow and let someone else deal with it. But these people can't do that. They can't just declare victory and go home. Bill? Well, since I had more time to think about it than anybody else, I'm going to go with the next speaker of the House of Representatives, Hakeem Jeffries, who I think is going to really change the dynamic in Silicon Valley and the cozy. that these tech bros have been enjoying in Washington for the last few years. Tchaim is on the march, I guess. He's going to come in here. Well, it assumes, of course, that the Democrats take back the House, which I think they will. Even Trump is making that noise, and he'll be the new speaker, and there'll be all sorts of interesting investigations, I think, that are long overdue.

I would say Representative Garcia is the one I'm paying attention to. Right. He could become a deputy, a key deputy. You're right. All right, guys. Thank you. So much. I really appreciate it. This was great. Full of information. Bill, Casey, Joanne, and Charlie, thank you for your time and see you all in 2026. Today's show was produced by Christian Castorosel, Michelle Eloy, Megan Bernie, and Kalin Lynch. Nishot Kerwa is Vox Media's executive producer of podcasts. Special thanks to Andrea Lopez Grisado and Eric Likki. Our engineers are Fernando Aruta and Rick Kwan, and our

our theme music is by Tracademics. If you're already following this show, your algorithm is full of tech stories and lesbian news. If not, you're doing ketamine, on camera, and with Casey. Go over you listen to a podcast, search for On with Caraswisher, and hit follow. Thanks for listening to On With Carous Swisher from Podium Media, New York Magazine,

the Box Media Podcast Network, and us. We've got a special show for you this Thursday on Christmas Day. We're bringing you an episode of The Long Game with Jake Sullivan and John Finer. It's a weekly national security podcast that takes a long view analysis of all the major news stories. Merry Christmas and happy holidays. We'll be back on Monday with more.