Violent Backlash: What the Sam Altman Attacks Signal for AI

0:00Support for this show comes from Virgin Atlantic. A lot of people dread flying. I've been on some bad flights and I've been on some truly miserable flights. But it's a whole different story when an airline shows up for you and the crew treats you like a VIP. Virgin Atlantic offers warm, one-on-one service from the moment you step on board. Its upper-class cabin features four-course meals, fully laid-flat seats, and drinks delivered on demand. Make the journey as exceptional as the destination when you fly Virgin Atlantic. Go to virginatlantic.com to learn more.

0:34Every soda's got something to say. Colas brag. Probiotics give TED Talks. Then there's Mr. Pibb. No pitch, no promises, just bold cherry bringing the flavor. Bold kick of cherry. And you're Mr. Pibb.

0:48Bradley Tusk:This episode is brought to you by Palmolive. Family time isn't just the big moments. It's weeknight dinners. Sitting around the table, everyone talking all at once. So when the plates are empty and the sink is full, use Palmolive Ultra. Palmolive's most powerful formula removes up to 99.9 % of grease, leaving your dishes sparkling clean. And the new convenient pump makes cleaning even easier, so you can spend less time tackling dishes and more time together. Shop now at palmolive.com. Today's number, 12 ,000. That's how many comments Trump received on an image he posted of himself depicted as Jesus before it was taken down.

1:27According to the president, he meant to be portrayed as a doctor. That was right after he called the Pope weak and terrible. And in other news, panic on Wall Street as traders prepare for the rapture.

1:42Bradley Tusk:Money market's mad. If money is evil, then that building is hell. Show those up!

1:51Welcome to Prof. View Markets. I'm Ed Elson. It is April 15th. Let's check in on yesterday's market vitals. The major indices rose as President Trump signaled he was open to talks with Iran. That pushed the Nasdaq to its 10th straight gain, while the S &P 500 came close to a record high. Oil prices fell below$100 a barrel. Bank stocks were mixed after earnings. Wells Fargo fell 5%, while Citigroup rose nearly 3%. We'll be breaking down all those bank earnings on tomorrow's episode. And finally, Amazon shares rose nearly 4 % after the company acquired Starlink's biggest competitor, Globalstar. Okay, what else is happening?

2:33AI has a popularity problem, and it is now getting violent. Last week in Indiana, a local councilman's home was shot at 13 times after he voiced support for a data center project in his town, a sign reading, quote, no data centers was left at his door. Then Sam Altman, the OpenAI CEO, was targeted twice in the same weekend. A man threw a Molotov cocktail at his home on Friday and threatened to burn down OpenAI's San Francisco headquarters. Police recovered a document from the suspect warning of humanity's, quote, impending extinction from AI, as well as a list of names and addresses of CEOs and investors of AI companies.

3:15The 20-year-old has been charged with attempted murder and faces a second count of attempted murder for the security guard who was at Altman's house. Separately, two people were arrested for firing shots at Altman's house on Sunday. These attacks are an extreme manifestation of the rising anti-AI sentiment in the US. Among 31 countries surveyed, Americans reported the lowest level of trust in their own government to regulate AI at just 31%. And people are now acting on that distrust. In just two years, $64 billion of data center projects have been blocked or delayed due to local opposition. So here to discuss these disturbing headlines and AI's general popularity problem in this country, we are having another panel discussion with two experts.

4:04We've got Bradley Tusk, founder and CEO of Tusk Ventures, and Brian Merchant, tech journalist and author of the Blood in the Machine Substack. Bradley and Brian, thank you very much for joining me on the show here. Bradley, I'll start with you. I mean, this news of Sam Altman, two attacks in the span of just a few days, it really is just a striking example of this growing feeling in America that I've talked about, I know you've talked about, and that is a lot of people just don't like AI at this point. What do you make of this news and what does it say about this moment?

4:40Bradley Tusk:Yeah, I mean, I think people don't like a lot of things. And to be clear, regardless of what you think of either AI or Sam Altman, no one should be throwing Molotov cocktails at his home or at anyone's home. But I think you've got a combination of, one, just general distrust or happiness in this country, right? Whether it is the fact that we are 23rd in the World Happiness Report or 62nd for people under the age of 25. Whether it's the fact that our government seems to be hijacked by extremists on both sides of the aisle, whether it's the fact that we haven't regulated Internet 2.0 yet. So even things like social media have never been dealt with by Washington, let alone AI.

5:16Bradley Tusk:And then combine that with the fact that AI is really unpopular. I saw a YouGov poll that showed that people have a 47 to 27 by margin of people distrust of AI. People think that AI will replace a lot more jobs than it creates. almost every different survey mechanism out there shows that people are fearful. And then anecdotally, when you just talk to people, they feel the same way. Then you mentioned in the intro local opposition blocking the construction of data centers. I think that's often the fault of the hyperscalers who seem to think that it would be okay to pass along all of their energy costs to regular consumers, right?

5:55Bradley Tusk:And if you are living near a data center, the idea that your electricity bills should go up 30, 40 percent to subsidize Sam Altman or Jensen Wong or whoever it is so that they can become trillionaires is unacceptable. And in this case, I think it's actually elected officials on both sides of the aisle acting in the interest to protect their constituents. And so, yeah, when you have a government that consistently fails to regulate technology, when you have a government that feels run by the extremes and you have a society that's generally unhappy, these are unfortunately the kind of things that come from it.

6:26Brian, you've written about this before, and your book is about actually the Luddite movement, which is sort of the first iteration of technology coming along, people getting very worried about it and revolting, essentially. What do you make of the attacks on Sam Altman? What does it say to you? Well, I mean, what it says to me is that this discontent, these grievances that people have are real. They are pronounced. And we have to look at them as if, you know, some of these people are obviously on the extreme end of whether it's a political spectrum or an ideology. At least one of these shooters was, you know, one of these ex-risk AI safety advocates who's really worried that AI is going to rise up and become sentient and end humanity.

7:20And so if you believe that, then, you know, doing all you can, you know, may look like a rational outcome as abhorrent as it looks to everybody else. And to step back a second, we do have a long history, right, when there is a disruptive technology, number one, number two, that is being developed and sort of unleashed by a particular sort of group of interests, right? When you have in the Luddites' time, that was the factory owners who were spearheading factorization and automation. And they were doing it without community input, without asking what workers and communities, what they wanted. So we have a dynamic that looks an awful lot like what's happening here today, where you have a few industrialists who had the backing of the state.

8:10They had all the resources. They had all the capital. They had all the power. And they were saying, this is the way it's going to be. We're going to automate jobs this way, and you're either going to sort of work in our factory or you're going to get out of the way. And the Luddites, who actually registered, this is one of the things that people get wrong about the Luddites today, is they weren't dummies. They weren't backwards looking. They understood quite well what was happening. They were technologists. They used this stuff every day. They used the automated technologies in smaller iterations in their workshops and at home.

8:43And so they understood what the industrialists were trying to do, and that's what motivated their response. They didn't want to see their way of life subsumed by factorization, given over to a relative handful of interests. So it was really about power, it was about democracy, and it was about losing agency. And so today, a lot of the backlash we see against AI is motivated by these very same fears and concerns in no small part because the AI CEOs and tech titans themselves have come out and used this language, right? From the beginning, they've said, oh, this technology is so powerful. It could be big trouble for humanity.

9:23It could be the gravest thing humanity has ever faced if we're not careful with it. it's going to eliminate 20 to 30 to 50 percent of jobs, you know, depending on how Dario Amadei at Anthropic is feeling. And it is going to be this huge disruptive event. And that's how they're forecasting. That's how they're describing their own project, their own business. And so, again, why would anybody, you know, not take that seriously, right? We take it serious at different levels, and some people will attach themselves to the X-risk element and say, well, we don't want to exterminate humanity. And most people will say, hey, I'm out here listening and you're saying, you want to automate all the jobs with AI tools.

10:06You want to automate, why would I be okay with that? Why would I trade that? Why would I allow a data center in my backyard to help you in that project? So to me, all of this backlash, I'm honestly a little surprised it hasn't arrived a little bit sooner, just how aggressive the industry and its leadership has often been. Yeah. Yeah. This gets to the sort of the PR and comms point. And Bradley, I mean, you've worked in exactly this sector. You've worked in politics. You've worked in tech and politics and how they come together. And there is this interesting question, which is like, well, all of the big AI CEOs are telling us that this technology is in a lot of ways quite scary and in some cases bad, like it's going to destroy things.

10:52It's going to destroy white-collar work. It's going to completely disrupt the economic model as we know it. And they've done it in a way that is legitimately quite scary. And I guess it does beg the question of like, I mean, why say that? If you're the CEO of a technology company, why would you come out and say this technology is going to be really bad and it's going to really negatively impact a lot of people's lives. I mean, what do you make of the comms strategy there?

11:22Bradley Tusk:Yeah. I mean, I think that, keep in mind, from their perspective, comms is a couple of different things at the same time. It's the way we're talking about it right now, which is how the public might perceive something, how regulators and lawmakers might perceive it. But it's also fundraising, right? So OpenAI and Anthropoc are still both privately held companies with giant valuations. OpenAI is nearly a trillion dollars at this point. And as they raise money, a lot of what you just said interpreted slightly differently is very appealing potentially to investors, right? So when you're talking about, hey, this is going to wipe out lots of jobs, what investors hear is this will be the tool instead that businesses are going to use to replace workers.

12:04Bradley Tusk:And instead, they're going to pay money to OpenAI, to Anthropic, to all of these different companies. And so I think that the language that you use potentially to recruit employees. So the New Yorker has a great piece this week on Sam Altman and a lot of the recruiting that he did at OpenAI was around the idea that he was the responsible person trying to protect humanity from the potential perils of AI. That clearly does not seem to be the case, but he used that language to incentivize people who did care about this issue genuinely to come work for him. There's language they use with investors.

12:39Bradley Tusk:And I think what they're finding right now, and I think sometimes this is sort of the both naivete and arrogance that you will see in the tech world, which is a lack of understanding of how their words then land with real people or with people in politics and government. And a lot of what they're saying is now coming back to haunt them. But the real question to me is, we know that the public is concerned. And we have seen, at least at the local level, elected officials protect consumers from things like paying for the costs of the energy needs of data centers. But when it comes to the larger issue of catastrophic risk, states like New York and California have done some regulation around frontier models.

13:19Bradley Tusk:But some of this really needs to be done at a federal level. And right now, we're seeing the opposite from this White House. We saw this White House issue an executive order in December telling states you're not allowed to regulate AI. And luckily, governors just from both parties roundly ignored that. But there are areas where you're going to see Washington need to step up. And I think whether or not they do so may dictate how this whole thing plays out. Stay tuned for more of this panel right after the break. And if you're enjoying the show, please follow our new Prof G Markets YouTube channel.

13:50The link is in the description.

13:58This is advertiser content brought to you by Virgin Atlantic Ed. A couple of weeks back, I got you a birthday gift Not to pat myself on the back, but it was a pretty good one. It was indeed. You surprised me with Virgin Atlantic upper-class tickets to London. So tell us all about it. It was pretty incredible. From the moment I entered that upper-class cabin, I have to tell you, I felt like a VIP. Anything I needed, a drink, snack, assistance with the seat. Flat seats. Flat seats. That's the key. Flat seats, exactly. Had the four-course meal, got my champagne, very delicious, enjoyed the food. And the journey home?

14:33The journey home was great. I went to the Virgin Atlantic LHR Clubhouse. That's the Heathrow Clubhouse. Heathrow Clubhouse was awesome. Got myself a coffee, headed over to the meditation pod that they call the Soma Dome. Kind of felt like a sort of spaceship where you relax and think nice thoughts. So I did that for a little bit. Then we went over to the wing, which are these acoustically sealed booths where you could do some work. You could even record a podcast. I didn't do that, but maybe I should have. It was a very enjoyable experience. So, Ed, the real question here is, what are you planning to get me for my birthday?

15:12See the world differently with Virgin Atlantic. Flying should be more than just transport. It is part of the adventure. Go to virginatlantic.com to learn more. Tickets and lounge access provided by Virgin Atlantic.

15:28Bradley Tusk:This episode is brought to you by Indeed. Stop waiting around for the perfect candidate. Instead, use Indeed Sponsored Jobs to find the right people with the right skills fast. It's a simple way to make sure your listing is the first candidate C. According to Indeed data, Sponsored Jobs have four times more applicants than non-sponsored jobs. So go build your dream team today with Indeed. Get a$75 sponsored job credit at indeed.com slash podcast. Terms and conditions apply.

16:20Bradley Tusk:Respect in this battle. Pick a meal to pick a side. And participate in McDonald's while supplies last.

16:30We're back with Prof G Markets. It's a very difficult time in a lot of ways to be an AI executive because, I mean, on the one hand, as you say, there is an economic incentive or maybe a fundraising incentive would be the right way to put it. to say that this stuff is going to be very damaging and it's going to just structurally completely upend the entire economy as we know it. But at the same time, I also wonder if they also actually believe that. And that seems to be something that you also have to kind of reckon with, especially in the context of a government which seems pretty unwilling in general to promote any form of policy, any form of regulation.

17:12And if you're building in the AI space in that environment, and you seem to recognize, you know, this administration doesn't really want to do anything in terms of regulation, then maybe you do feel you need to sound the alarm and say, hey, this is going to be, this is actually a big deal. This is actually going to be a problem. And then on our end, it becomes very difficult to understand what's true and what's marketing and what's hype. So I guess, I mean, Brian, just turning it to you, which parts of the story do you think are real? I mean, when Sam Altman goes out and says, yes, this is going to be massively destructive in a lot of ways, or when Dario Amadei says that, I mean, to what extent should we take that seriously versus write it off as, you know, marketing?

18:02Yeah, I mean, I think you're absolutely right that it's, you know, both of those tendencies are kind of bound up in this same trajectory. and part of this is necessity, right? Like the tech landscape is such that if somebody wants to release a product that can compete with one of the giants like Meta or Amazon or Google, then you need just truly an immense amount of capital. If you want to compete rather than angle to get bought up or something. So you need a story that can command the kind of capital that can compete with one of, you know, three or four of the tech oligopolies that are out there, right?

18:42The tech monopolies that have sort of over the last 20 years sort of concentrated their power. And so that story then becomes not just, hey, here's a cool product. That's not going to get you there. You need a story that is on the magnitude of we are creating the software that can automate every meaningful job. And, you know, that language is right there in OpenAI's charter still to this day. you can look at that is intrinsic to the pitch to investors. And so I think there are a number of different factors there. I think if you look at the last 10 years of the history of sort of this latest AI boom, then you really see it beginning in earnest around at least expressed fears about X-Risk and the possibility, you know, as sort of presented by Nick Bostrom and others, that AI could become super intelligent and become this danger.

19:34I think one of Sam Altman's key intuitions was that, you know, early on when he was, you know, just sort of heading up, you know, just quote unquote, you know, heading up Y Combinator, he sensed that there was a lot of energy here in this space that he could tap into one way or the other. And so he reached out to Elon Musk and kind of mimicked this language and was able to sort of use that concern just as a lightning rod to get some interest and power and momentum into AI in general. And then from there, it's hard to walk away from that narrative. You see that the more you talk about it, that it does affect investors.

20:08It does sort of compel people to pay attention. It does get headlines. And so I think it does sort of balloon on and on and out. So some of these guys, I think like Dario Amadei, I am sure he's legitimately concerned about all of this stuff. Is his marketing department aware that he can win a round of headlines by expressing that concern in the release of Mythos? Of course they are. So they present every sort of white paper, every released or unreleased model, you know, with the same sort of level of gravity as though it were sort of, you know, a new set of promotional materials. And so, it becomes difficult to distinguish between the two.

20:48But I would say it is yes and both. And now we're in this pickle where the AI industry can't really walk away from its promise that has attracted so much investment in the first place. They can't say, you know what, we're not going to automate all the jobs. Then SoftBank might say, well, then what was that$30 billion for, right? So it really – we're sort of up on the brink and the precipice here. And I think Bradley was absolutely right. It's not just the politicians. It's also the AI industry. Meta and OpenAI and all these guys are bankrolling PACs right now to the tune of$100 million to sort of influence elections.

21:25They supported the moratorium to ban state-level AI lawmaking. So the very least they could do if, you know, they want to de-escalate the rhetoric, as Sam Altman says, is, you know, is stop interfering in the democratic process, right, is to let voters feel empowered and feel some sway over this technology that is, you know, being integrated into every pore of society. Yeah, it's a great point. I mean, if there's one thing that's going to make you dislike AI even more, it's to read a headline that Marc Andreessen is bankrolling millions of dollars into these pro-AI super PACs that we are continually starting to read more and more about.

22:04I mean, Bradley, what is the right policy response here? I mean, what we've kind of identified is that we don't seem to have much regulation at all. Americans are very scared. They're getting increasingly angry about it to the point where we are seeing literal violence against these tech CEOs. Like, what are we supposed to do this from a policy perspective?

22:25Bradley Tusk:Yeah, I mean, I think you almost have to think about it from a taxonomy of how to regulate AI. because I've never, you know, I've been working around politics for over 30 years, and there's never been anything quite like this. So there's, in my mind, kind of four different categories. The first is consumer protection, and that typically tends to be the province of state and local government. So that's things like regulating chatbots, especially around things like mental health, regulating data centers and the negative externalities that can post on others, regulating the use of AI in hiring decisions, things like that.

22:56Bradley Tusk:The second would be catastrophic harm. Like we've said, California and New York have tried to pass regulations or have passed regulations around frontier models, but that's two of 50 states. And this is the kind of thing that really should be done by the U.S. government. The EU has a framework that covers 22 countries. We have two states. So that's number two. Number three would be jobs. And I don't think there is any plan whatsoever for how to deal with the fact that we could be seeing 10, 20 percent unemployment at some point because of AI. And look, I do believe that at some point in 20 years or whenever it is, all kinds of new industries that we can't conceive of today will be created that will have a lot of jobs thanks to AI.