If one of their bots creates a virus and kills people, they need to go to jail. Period.
Hi everyone, this is Pivot from New York Magazine and the Vox Media Podcast Network. I'm Kara Swisher and I'm Scott Galloway. First, Donald Trump is dismissing AI concerns. This is the story, Scott. We really have to talk about this. He's saying any efforts to limit the technology are part of a sick conspiracy against AI and data centers. We're part of a sick conspiracy, Scott. According to Trump, the only control or guardrails that AI needs is a strong and smart high-IQ president, and the USA has that in spades. This comes after Anthropic CEO Dario Amodei published an essay over the weekend calling on the AI industry to slow the pace of development. He wants independent evaluators to review companies' safety practices and is urging a global effort, including China, to create standards and limits. In a rare and convenient show of unity, both Sam Altman of OpenAI and Elon Musk of xAI are backing Dario's call for a slowdown. Dario also spoke out on Face the Nation about wanting the government to be involved in all of this. Let's listen.
Anthropic, and I would hope other companies, have done everything we can to try to have legitimate oversight mechanisms. I think the government and the public need to have a stake, and this is why we've supported regulation of the technology. Regulation constrains the private companies. Regulation allows the public and its elected representatives to have a say and limits what the private companies can do.
So, will the government get involved? Let's listen to what Speaker Mike Johnson said on Meet the Press.
I think we've got to have meaningful regulation that does not smother American innovation. Now, the how to solve that equation, no one knows yet. We're not going to rush in and pass a piece of legislation that would do harm to the country and put China at an edge. And so we've got to do this carefully. Of course, we'd vote on that tomorrow if we had the solution. And I'll call everybody into session any day, but we have to make sure we're doing it the right way. I think the first step is to have a meeting of all the platform providers. I've talked to the president about this and others. I believe that's the next step. I think we need to do it quickly.
Of course, they've been meeting with them for months. They were all at the inaugural mic anyway, and they've been meeting tons. What do you — a lot of people are wondering what's called for this slowdown now? Did they see something, or is it just money? David Sacks, who is Trump's guru who's been pushing just sort of wild-west kind of policies, posted, 'Stop pretending the motivation to slow down is purely altruistic,' noting they want to slow down open source. In this case I agree with David Sacks, although I think he's been a menace because he hasn't allowed any reasonable legislation to go through. Nvidia CEO Jensen Huang responded to AI fears last week saying, 'What better way to create a demand than to create a problem.' We'll get to AI — OpenAI not going public until next year, which we have said many times, but talk a little bit about this because there's a lot happening here and suddenly all at once, you know, spurred by that Anthropic researcher who said it's going to kill us, which I think got everybody's attention.
Well, Speaker Johnson's response to this what feels like existential issue was basically to send out a save-the-date to say that we should meet about it, and to say that if there was anything that reasonably threaded the very fine needle of not wanting to get in the way of innovation and create shareholder value and economic growth, which is key to an economy, at the same time recognizing the externalities, which by the way is a tension we have dealt with for centuries with respect to economic progress versus not capturing too many gains for shareholders and outsourcing all the problems to the public that at the end of the day, other than keeping the nation safe, is what governments do through thoughtful legislation and local officials. Do they make that gavel for a man or a leader?
Governor Newsom has passed 12 laws attempting to regulate AI in California. They have laws that say no bot can pretend to be a person for longer than a certain amount of time because they realized in their nation and other nations adults and kids were falling in love with Cersei Lannister, right? They have certain hours that kids cannot use AI. They have all sorts of limits around how AI can be used for information. Now granted, they're using it for control, but they have a lot of regulation. And they understand the dangers, they understand the threat. And what do you know — 75% of the Chinese populace is optimistic about AI, and we haven't done anything right now. America isn't engaging in an experiment, it's the following: we have never in the history of the modern economy had a trillion-dollar industry that is totally unregulated.
He's like, 'We'll get to it,' but it's like, it's been 20 years, Mike, in case you're interested, and we've seen the damage done. Let me just posit something. I wrote to a pretty prominent technologist in this area and they said, you know, I was trying to compare it to nuclear proliferation, right? This is a very similar kind of thing. It should be global. I've been pushing for global standards for years and years with both Democrats and Republicans. And he said, right, if this was back then, Truman would have nationalized AI at this point. You know, that Trump — I know — and this was someone who does not believe in a lot of government things. He goes, 'We're at the — it needs to be nationalized.' Now, Donald Trump saying any efforts are part of a sick conspiracy, data centers, like do nothing is politically — this is like not a good idea, I think. And especially since he's utterly incompetent. He is not the person who is high IQ that can handle this. He is someone that gets bought and paid for by the technology industry. Talk a little bit about the feeling also on the flip side that it's all these companies trying to zone out open source by solidifying because they can — if the government passes regulations, they can pay for them as opposed to small startups and things like that.
There's just so much here. The first is I'm going to bring up a name here. Diane Whipple. Do you remember that name?
33, a former all-American lacrosse player. And in January 2001, about 25 years ago, she was mauled to death in the hallway of her Pacific Heights building by two Presa Canario dogs.
Dogs. Oh, I was — yes, I lived there then. It was terrifying. Yep. Crazy neighbors with a crazy dog. Yeah.
And the neighbors were prosecuted. One is still in jail. One died in jail because the law decided if you have these agents, i.e. vicious dogs, and you know they're capable of this, and they're off leash and they run into a hallway and they maul someone and kill that person, you are on the hook for this.
And now we're seeing across the nation that if parents don't secure their AR-15s and their mentally ill 15-year-old gets a hold of one of these things, we're going to prosecute the parents. And this goes to incentives. Right now, I would argue the majority of senior-level executives claiming, 'Oh no, Frank's out of control,' as in Frankenstein, is an attempt to absolve themselves of the responsibility that they own these pit bulls. And by the way, pit bulls are lovely dogs. I don't mean to malign pit bulls. I don't think they're — okay, maybe some of it is concerned. Fine. But this is about incentives. You are responsible for these agents and these bots, and while you want to capture the shareholder and the upside of the good things they do, that means you are responsible for the downside.
The fact that we have nothing resembling the IAEA. We have cooperated multilaterally. We have not blown ourselves up with enemies on nuclear weapons for 80 years. The first thing that needs to be done by Trump is he needs to get into a room with Xi and say, 'We have a mutually vested interest in no one developing a supervirus. We have a mutually vested interest in our youth not killing themselves. We have a mutually vested interest, especially China, in some level of control of misinformation that enrages people and gets them to start demonizing and going after and lynching people when a rumor there.' There is so much overlap in terms of mutually vested interest that they could get in a room with talented technologists and diplomats and say, here's a two- or three-page quote-unquote memo of understanding around what we are going to do. But the incentives need to change, and there needs to be something regarding leadership and a regulatory body here. And by the way, there are examples of how to do this. And I was thinking about this last night. The model, in my opinion, is the FAA, where the SEC after 1929 — pre-market safety review for all models above a certain capability line.
Never going to happen with Trump, just like never.
Well, okay, then that void will be filled by Newsom or someone else. And I do think this is about to become a campaign issue. This is about to bubble up from data centers to the danger of the models in terms of an election issue. Pre-market safety review for models above a certain capability line before deployment. Mandatory incident reporting similar to the NTSB, the National Transportation Safety Board. Anything goes wrong, you have to file it immediately and make us aware of it. And then you fund it with a transaction fee on frontier compute, the way FINRA works. That way, taxpayers aren't subsidizing oversight of the richest companies in history.
Correct. And taking the one thing every fast-moving industry hates and needs — an independent evaluator with, wait for it, subpoena power. Not what Dario's recommending. These embedded evaluators. 'Hey, Susie works for the government and makes 1/50th of you and she has no real teeth. But take her out with you tonight to the all-hands.' Dario is literally asking for this — zero teeth, zero juice.
Well, they all are like Mike Johnson. Let me just interject. Mike Johnson's like, 'Let's get together with the leaders.' Why aren't they making the decisions?
Why the — they love meeting at those dinners. Have you been invited to one of those Trump dinners? Because we know a lot and we've been actually accurate. We have been talking about this issue for years. One thing we have to do, and I think we've done a very good job here on Pivot of articulating, is people hate you guys. Like, let's bring citizens in. Let's bring regulators in. Let's bring academics and scientists in that do not agree with you and your outlook. Bring people that don't have a financial interest in what's going to happen here and change the narrative. This is one that's driving me crazy. 'AI will kill 10% of humanity.' No, they won't. The people who make AI will kill 10%. Let's stop talking about AI being — it's like saying electricity will kill whatever percentage. And they did, by the way.
Didn't kill 50,000 Japanese. We did.
We did, that's correct. Like the bomb is just the tool, and for them to characterize it like — I don't know how mistakes were made by whom, and they need to be absolutely liable up to and including jail. Like up to — if one of their bots creates a virus and kills people, they need to go to jail. Period.
Period. And the fact that they're characterizing it as 'AI will do this' — AI will do nothing. They will do something by creating it. And the fact that Trump is in there — he is a bought-and-paid-for, corrupt, demented person. And we should not put this in his hands in any way whatsoever. He's not just ignorant, he's corrupt. And so it creates this terrible — we are now in a position as if someone crazy was in charge when nuclear proliferation, the nuclear bomb happened. Can you imagine letting companies figure out whether we should drop the bombs? It just — we can't do that. It makes me crazy.
Well, okay. So again, the analogy I've always used is: would we let Oppenheimer raise money from Andreessen Horowitz and start selling nuclear bombs into PNG and UAE? We wouldn't.
However, practically, we're not nationalizing companies that offer this much shareholder value and innovation. They're just not now. And by the way, I just want to correct the record. I said 50,000 Japanese. 50,000 people died on the initial burst. Somewhere between 150 and 250,000 Japanese died between the explosions and the fallout in Hiroshima and Nagasaki. Okay. What are the known knowns here? Trump's not going to do anything.
He doesn't understand it. The people around him are too old and too incompetent to understand these technologies. This would take real work and real effort. They're also all in the take.
Their view is: how do we get a vig? How do we let these companies make trillions of dollars and we get a vig either directly or indirectly? 'Hey UAE, you can have GPU chips if you fund my kid's crypto venture or my son-in-law's golf course.' So that's not — RFK Jr. He's the head of HHS. He is charged with the nation's health. He oversees something called the CDC, the Center for Disease Control. The greatest threat right now to the control of disease is the possibility that one of these frontier labs has tools that develop a supervirus.
That's right. If we had anybody resembling a competent head of HHS, they would already have 40 or 50 scientists and technologists sitting in the offices of these companies with full force of law saying, 'I'm not asking you what I need to see. I am telling you.' And then asking these models before they are allowed to be touched by anybody, 'How do I make a supervirus?' and see what they say. The CDC should be on site here.
This is a guy that just eats steak for breakfast, works out in jeans, and is really quite mentally ill from what I can tell. Like in all ways, he has no interest in this. The people that should have interest in this do not, or they're on the take, or they're just — they're on the take. I don't know what else to say about this government at this point because this is incredibly — if you were — and by the way, Mike Johnson, you are actually a representative of the people allegedly, and instead you are hand-in-glove with Trump. You want to have dinners with these people, and let me tell you, the minute you lose the midterms, you're going to be a footnote to these people. They will move along from you because you didn't get them what they wanted. They're not interested in you. They think you're an idiot. I know this for a fact. And your job is to represent the American people, which nobody is representing them here. And again, let me stress: AI is not going to kill 10%. There's a 10% chance that AI will kill us. There's a 10% chance that these companies and their leaders will kill us. That is what we have to keep saying over and over again. So the known known is the Trump administration will not be effective here, nor do they have any desire. So what could be done?
And I never miss a chance to sound more important than I am. I'm having dinner with James Milliken tonight, who is the president of the University of California. Before that he was the president of the University of Texas. This is —
This is your board meetings. Congratulations.
Thank you. This is a talented man with very strong political connections and an unbelievable command of education. This is my idea. He needs to hire Alex Stamos and he needs to assemble a collection of technologists, legal scholars, social scholars, philosophers, and put together what is the largest evaluation committee and blue-ribbon panel and immediately offer it up for any frontier model. And then get — and I think they would sign up Altman, Amodei, maybe Musk. I think they would agree to this thing. I think they are actually at the point where if we don't get something resembling a thoughtful, credible committee or evaluation board that serves as a model for a regulatory board — there have never been companies with multi-trillion-dollar valuations that are this vulnerable.
I'm sitting here in my hotel room looking at the intersection between Santa Monica and Wilshire Boulevards. If it got out that a rogue bot started messing with the traffic lights and there were some car accidents here, the next day these companies might have to actually shut down.
These companies are so — the media is waiting with bated breath. Anything regarding a rogue action here.
And can I say, again, I cannot stand David Sacks. I think he's been a menace here. But he is correct. The motivation for the slowdown is not altruistic. They are trying to solidify their strengths against open source at the same time that they're warning us. They're trying to protect their businesses so they can dominate and nobody else can. They're trying to quash innovation. And this I agree with him. That said, he's been incredibly irresponsible in blocking any kind of guardrails. And so just because you say one right thing, David, doesn't mean I think you're not a menace. In related news, speaking of money — because money, you have to pay attention to the money here — Sam Altman has confirmed that OpenAI will not go public this year. He told Fortune that given the current safety concerns, this would be an ill-advised moment to take the company public. This is just an excuse. This is — they can't go public. Scott, I'd like to play a prediction you made back in February.
I think that what was supposed to be the most anticipated IPO was the IPO of OpenAI in 2026, sometime this year, early '27. I don't think that's going to happen. I think this company has now gone into full — I don't call it panic mode, but it feels as if the momentum has a habit of creating more momentum, and I think the momentum is really negative around this company.
So meanwhile — and I want you to get your thoughts on that. Meanwhile, Anthropic has probably chosen NASDAQ for its IPO, even as it's scared of killing everybody. It will be going public, which could happen as soon as October. So how does Dario move forward with the IPO if he's so concerned? He's so concerned, Scott. And what does this mean for OpenAI? Because these are the two twins — you know, everybody else is peanuts compared to these two. Well, except for Gemini, but that's within Google. But go ahead.
Well, we're never ones to feel bad for tech CEOs, but Dario and Sam are being pulled in a lot of directions right now. They have society and philosophers and media saying, 'Okay, do you really want to be the guys in your late 30s known as the greatest murderers in history?' Or be the ones that are responsible for hospital systems having their power cut off through a cyber attack that thousands of people die because they can't get their kidney dialysis or the emergency room lights go out. I got to believe that unlike Mark Zuckerberg or even Sheryl Sandberg, who I think is given way more credit for not being psychotic than she should, I get the sense — I'd like to think these guys are actually concerned for a lot of reasons. Having said that, their job is to get their A, B, C, D, E, and F-round investors who have invested hundreds of billions in their companies a return. And a lot of these guys, the VCs and the employees, have already started spending that money. Housing prices are up 24% in San Francisco. You don't think that all the engineers at OpenAI have already gone out and started shopping for houses? You don't think that the Series C at OpenAI is counting their money and saying, 'I have this basket of stuff in my portfolio and my stake in OpenAI is going to pay for the entire fund and I'm going to be a successful VC finally. What? You're delaying it? What the — No. Get out.' These guys are being pulled in a lot of different directions, and because they have a total lack of leadership of a regulatory body that says this is what is required to have a safe frontier model, this is what is required for a company like yours to assess the risk and put out an S-1 and actually go public, they're in no man's land. There's too much — look what's happened in the market just since the kid came out and said there's a 10% chance of an extinction event. This is what's happened. SoftBank is down 15%. Nvidia's down 3%. CoreWeave down 9%. Oracle down 5%. Microsoft, Amazon, and Alphabet all lower. NASDAQ futures down 1.5 to 2%.
Not only OpenAI — Nvidia and Anthropic are really fragile right now.
They all are. The entire economy is really fragile because Trump has made a giant bet on AI, which is responsible for 90% of the capex increase, 72% of the earnings growth, a ridiculous amount of the GDP growth. So our economy is now resting on a pillar where a guy showed up — a 28-year-old showed up and said, 'Oh, by the way, 40% of the NASDAQ is kind of engaged in this thing that might be an expected death of 80 million people, or 800 million if you believe my 10% calculation.' This is such a failure of leadership and governance. There should have been a regulatory body a couple years ago telling the media — the godfather of — I apologize, I'm all over the place. The godfather of AI right now, Hinton, summarized it perfectly. AI can go one of two ways. There can be a very good AI that helps solve pancreatic cancer, makes people more productive, lowers costs, figures out a way to take care of mom more efficiently, extend life, extend quality. There's all sorts — figure out threats of nuclear debt. Oh, there's all sorts of wondrous things AI could do. Good AI, bad AI. I won't even go into the specifics of bad AI. We can all imagine them. But the fork in the road — a big component of the fork in the road is thoughtful leadership and multilateral cooperation, of which there is none right now.
Right. And it's all coming. And that's the thing — politically, this is a powder keg. California Governor Gavin Newsom just signed a package of 13 bills targeting social media and AI use for kids, creating some of the strongest online protection in the country. The bills include a ban on platforms offering addictive features to users under 16, as well as stronger safeguards for AI companion chatbots. There's also a penalty of up to a million dollars per child if a social media company is found negligent in causing harm. Still a high hill to get to these people. In signing the bills, Newsom took a victory lap for California, saying we're going to lead the nation again. The bar is pretty damn low. Congress is not going to follow suit. Now, there's a lot of criticisms of some of these parts of these bills — they don't go far enough, they go too far, they do this. But you know, this is a patchwork regulatory scheme, even if it's California. And so we need to have Congress — Mike Johnson, who is going to be a footnote, by the way. You won't have your job. You better get on this like immediately on the thing because first of all, people care about it. You've seen them express it through data centers. We cannot have these companies making the rules for the rest of us. Let's just be dead clear about that. They can't.
When has that ever worked?
When has that ever worked?
The incentive when — let's let the car makers decide on, you know, seatbelts. Ralph Nader, by being an asshole, and that's great. We need that kind of — and that's what's coming for you people. You don't understand how much people hate you, and they feel it. They feel it in their bones. And so stop playing antitrust games. Stop trying to solidify by finding an opportunity when you are screwing us. You're always trying to find an opportunity, these companies. Anyway, I'm going to stop.
Yeah. The hate's about to bubble up from data centers to these frontier models. It's the same stuff. Well, you've really blown it, kids. Let me just say you've absolutely blown it. Anyway, let's go on a quick break. When we come back, Mark Carney's power move. Support for this show comes from NetSuite.
When your competition is already implementing AI, your business needs NetSuite next to stay ahead of the game. You probably know NetSuite, the AI-powered business management suite that securely connects all of your data. It's a unified suite that brings together your financials, inventory, commerce, HR, and CRM into a single source of truth. Trusted by over 40,000 customers, NetSuite Next is the next big leap in how business gets done. Because AI is built into everything you do, it automatically surfaces custom insights throughout the day. The AI agents work alongside you to solve problems and handle routine work. And anytime you have a question about anything, you can ask just like you're having a conversation with a colleague. NetSuite is customized for a wide range of industries, so it supports the way your business truly works. Whether your company earns millions or even hundreds of millions, it's time for NetSuite Next, where your business meets AI. If your revenues are at least seven figures, get their free business guide, 'Aligning for the Agentic Era: How AI Is Changing Everyday Work,' at netsuite.com/pivot. The guide is free to you at netsuite.com/pivot.
Support for this show comes from Framer. Your website is the public face of your business, but that means any blemishes, typos, or ugly designs are showing up loudly and in high def. Not exactly leaving a great first impression. Framer is the AI website builder that helps creators, teams, and businesses ship production-ready sites faster than ever while getting every detail right. Prompt, inspect, edit, and publish in one place at a whole new pace. It's an enterprise-level solution powered by agents that provide premium hosting, enterprise-grade security, and 99.99% uptime SLAs. With Framer, agents and humans work in tandem. Agents bring speed and scale. People bring taste, judgment, and control. Framer's agents solve the gap between AI-generated ideas and production-ready website work. That's why leading brands like Perplexity and Miro build their websites in Framer. Learn how you can get more out of your site from a Framer specialist, or get started building for free today at framer.com/pivot for 30% off a Framer Pro annual plan. That's framer.com/pivot. Rules and restrictions apply.
Scott, we're back. While Donald Trump has been busy stoking the flames of a trade war with Canada, Prime Minister Mark Carney — we love him, he's a handsome man — is on the rebound, working behind the scenes to pursue a relationship with the EU. He doesn't want to fully commit Canada to membership and is asking to be an associate member of the European Union. I love that, just — they're going to date. Carney has reportedly been speaking to the EU heads of state about energy, AI, critical...
Minerals, defense, even visa-free travel for Canadians. He's scheduled to address the European Parliament later this year to flesh out his vision for how Canada would plug into its 21 trillion dollar EU economy. What do you think? You love Mr. Carney. So tell us.
Yeah, I think he's going to be Time's Man of the Year. I think he's going to win a Nobel Prize. I think he defines forceful yet dignified. And I think he's decided, I mean, effectively what you're going to have is, and this is the opportunity for the next presidential candidate, I'd get out ahead of this. I would propose an alliance, not only a free trading zone, but a diplomatic alliance between all democracies.
And Carney, what the EU is desperate for is a leader with some power because unfortunately, the whole is less than the sum of its parts because what they have is 27 veto votes. And their neighbor, other than saying what they don't like, it's very difficult for them to band around any sort of serious economic or technology policy because everyone has a say. And this is a big opportunity, in my opinion, for Canada basically to continue in an effective way to say no, that we're not going to be bullied. They have more in common right now with the EU than they have with the US. The US will be big losers here. We're under the impression that we can bully anybody. We're a third of the world's economy. We just can't do it.
I'm talking about my lunches. I'm having lunch with a guy who runs $150 billion and he always says, 'Where's the alpha? Where are your ideas?' My best idea right now is the following. There's $750 billion, or three-quarters of a trillion dollars, of commerce. I think every one of those $750 billion is looking for a new provider of those services. And I'm going to say to this guy, what European companies are poised to take a Canadian business away from America?
Yeah. Because Mexico could do the same thing.
Huge opportunity for Mexico. Quite frankly, I don't think it has the same... That's a correct question and a great idea. I would think that with Carney trying to cuddle up to the Europeans, that if you're some Finnish manufacturer of ball bearings and Canadian whatever it is, paper companies are buying these ball bearings from a US manufacturer in Minneapolis, I'd be like, 'Hey, it's going to be a little bit more expensive, but we make pretty good ball bearings in Finland.' I think there is so much business that is now up for grabs to the detriment of American employees, factories, and shareholders.
Trump always says we don't need anyone. I'm like, we need people.
