The AI Panic Comes to Washington

0:01Everyone wants health care to be simpler and easier. That's why UnitedHealth Group is fast-tracking more than 60 % of children's procedures that used to need pre-approval. So more kids get the care they need when they need it. And speeding up procedure approvals and payments for more than 1 ,500 rural hospitals. So they can focus on caring for their communities. And that's just the start. Learn more at uhg.com slash access. From the opinion pages of the Wall Street Journal, this is All Things with Kim Strassel. Controversy over artificial intelligence and data centers explodes on Washington with tech CEOs calling to slow the pace of development and politicians demanding regulation.

0:48But how worried should Americans really be about where all this is headed and how much of this is politics whipping up fear? Welcome to All Things with Kim Strassel, and here to provide some calming words is Andy Kessler, the author of the Wall Street Journal's Inside View column, here on this show live and for the first time today, and I am so thrilled. Thank you, Andy. Thanks for having me. Sure. Okay, so you had a really great column this week, which was a joy to read amid all the panic, entitled The AI Luddites Are Here. And before we actually talk about artificial intelligence itself, can you actually just give viewers a little history lesson on where we actually get that term, Luddite, the actual story of Ned Ludd?

1:32Because I think some people might have heard that word, but they don't really know where it comes from and how fitting it is for this discussion. Sure, sure. Well, Ned Ludd was a 18th century factory worker back when things were very manual. And he got fed up one day and broke a few things. But then years later, workers got even more annoyed when they started automating, making up socks and other things. And so they called him General Ludd. He turned into like a fabled person. And they went in and they started destroying these stocking frames, which are how you make socks, burn factories. And so the term Luddite really comes from those that think that technology is going to destroy jobs.

2:15Of course, the Industrial Revolution created way more jobs than it destroyed. But, you know, that's no one likes to think about that. Yeah. So, I mean, essentially it describes a reaction whenever there's something new, dramatic that threatens to, you know, overturn things. And that can be a little bit scary. One of the things I think is getting missed in this debate is there's actually a lot of different issues and fears that are colliding here. And you kind of go into some of those, isolate them in your column. I want to go through them. Let's just actually start with this. So tell people why at this sort of peak AI pandemonium, why you think this whole era is manageable and not necessarily worth losing our minds over, at least not yet.

2:59There is a hysteria, right? The sky is falling. It happens every time, right? The guy who left Anthropics said AI could kill us all. You know, that's about as extreme as you can get. I mean, it's a great issue. You've seen Democrats jump on it and say, you know, let's get rid of data centers. Let's put a halt on all this thing. But it's mostly thing rule to me. I mean, you know, every labor-saving device that has ever come along, we talked about the knitting frames, I'm sure dishwashers and vacuum cleaners, automobiles, nukes, Uber, CRISPR, right, which saves lives for people with psychosal anemia.

3:33That could kill us all. And probably even the wheel originally got protested as threatening to jobs. So one piece of this, and you were just talking about it, Ned Ludd, it's individuals or specific industries that are worried about losing their jobs. And I don't think people entirely understand what a huge motivating aspect that is in so many parts of this AI pushback. So you give some examples. You point out that some school districts are beginning to ban AI in public schools. You note New York, also LA. Obviously, this is not because there are concerns that AI is going to make the learning environment any worse.

4:15It's hard to think of it getting any worse, especially in some of those cities where kids are really struggling. But I mean, this is fundamentally, you use the magic words. It's about the teachers unions, right? Yes, yes, precisely. Because, you know, education, if it uses AI correctly, right now, there's a pushback on AI in schools because kids use it to cheat. Hey, write a 500-word paper, or I guess it's more in pages, right? Write a five-page paper on such and such, and they use open AI to cheat or for math problems. But that's not how AI is going to be used in classrooms. AI is going to be used in classrooms as like a self-paced tutor so that kids can go through and learn things and have a feedback loop.

4:57And they will all come out educated as opposed to now where 30 children sit in a classroom, half of them are asleep or on their phones. And so it's going to restructure how education is done. It's not going to be a teacher sitting in front of the class and droning on. It's going to be this self-paced machine. That scares every teacher who says, well, I'm out of a job. I think it'll actually help the education system turn out smarter kids. They'll do even higher levels of education. Every kid will take and learn calculus and physics, not just 10 % or 15 % of them. But the teachers union loves to freeze things in place.

5:35They always like, you know, more kids per classroom. And this way, it's, you know, you have this machine that can be one-on-one and teachers can then be used to go beyond the screen and teach real life IRL, in real life education. I love that example. And it's something that people need to just bear in mind as they're reading some of the coverage of this. This is not necessarily over fear. It's kind of straight old job protectionism in a way. And you gave another great example of this, which made me giggle because you were talking about how now Uber and Lyft drivers are trying to take out Waymo, which are, of course, robo taxis operating in about a dozen systems.

6:14And you pointed out, though, that not so long ago, it was actually taxi drivers who were trying to take out Uber and Lyft because it was competition to them. But there's also, and you get into this, there's some nimbyism going on here, not in my backyard, in particular when it comes to data centers. I find this fascinating because usually most NIMBYism is paired with a little bit of a disconnect about how the world works. You know, people want all the stuff they want. They just don't want the nuts and bolts of where you get all the stuff and we're near them. So you pointed out that this former Philadelphia Eagles Center, Jason Kelsey, big time podcaster, is urging protests against data centers seeming not to care or know that everything that makes this podcast possible is internet and streaming and cloud services and cell phones.

7:02You also quote a poll showing that about 70 % of Americans oppose data centers. Do you think that most Americans actually understand what these are, how they fit into their life or what they even look like? No. I mean, I think most Americans, if they walk by a data center, they would think it was just a warehouse. I mean, they wouldn't know what it is. When you walk by an oil refinery, you know what it is or a aluminum smelter. They're big and ugly. And here, these are just buildings and they don't make a lot of noise. And contrary to popular belief, they don't use a lot of water. They do use a lot of power.

7:40And so people complain, oh, our bills are going up. But, you know, you can't even use natural gas anymore, right? You know, everyone wants to use wind and solar. And so, of course, when a power hungry customer comes in, they don't have enough power. And so there's a run on building power centers. Maybe they're going to be part of the data centers. But there's a huge misconception. But so 71 % said they don't want data centers. But then another poll said 71 % said, I'm worried about the pace of AI and what it's going to do to me. It's the same percentage. So people are not so much worried about the data centers, they're worried about their jobs.

8:17Or because they've been trained by these companies, they're worried about extinction, right? Oh, this is going to kill us all. And it's just the worst PR. Like, who runs the marketing departments at these companies? It's just bizarre. No, I don't understand why they're not out. Look, there's actually some opportunities here. You know, to the extent that you have some of these companies that are now engaging in building their own power facilities, which if they have additional electricity that can put onto the grid, you could actually reduce power prices around the country if this gets going. There's actually putting pressure on us to actually do some real energy construction in the country, something that we've been behind on.

8:57All of a sudden, nukes are OK, like just like that. Oh, we're building nuclear for the project. Yeah, OK. It's going to lower my bill. Yeah. Well, that's that's progress, right? We're going to take a quick break when we come back more with Andy Kessler, the author of The Wall Street Journal's Inside View column. Everyone wants health care to be simpler and easier. That's why UnitedHealth Group is fast-tracking more than 60 % of children's procedures that used to need pre-approval, so more kids get the care they need when they need it. And speeding up procedure approvals and payments for more than 1 ,500 rural hospitals, so they can focus on caring for their communities.

9:36And that's just the start. Learn more at uhg.com slash access. This is Matthew Hennessey, the editor of Free Expression. I want to personally invite you to come check out what we're doing. Signing up for the newsletter couldn't be easier. Just go to our homepage. Directly below the Wall Street Journal logo, you'll find the magic words. Free Expression, wedged between opinion and arts. There's an option to sign up for the newsletter right at the top of our section. Just hit that button and you'll start getting the world's hardest working newsletter delivered directly to your inbox Monday through Friday.

10:13If you aren't a journal subscriber, the newsletter is free, but we are going to need your email address. There's a lot of fun stuff in the newsletter that you can only read in the newsletter, and I don't want you to miss any of it. Sign up for free expression today. We'll see you in your inbox.

10:31Welcome back to All Things with Kim Strossel. I'm here this week with Andy Kessler talking about AI. So a huge part of this, though, I believe, is simply politics in various different forms. And I really want to dig into this. So you mentioned it, what kicked this off, it was this latest round of drama, was this anthropic employee who quit partly on the grounds that he believes everyone in the world in AI knows that this new technology is going to kill us all. And there's some interesting evidence that this individual is actually allied by, supported by this kind of quasi-political movement known as effective altruism or rationalism.

11:10And, you know, the AI industry is absolutely stocked with these folks. Can you give us a little bit more about what that philosophy is and how scary it is in a way? Well, I'll just go back a few years to the Sam Bankman-Fried of FTX, right, who scammed everyone. And he was a big investor in Anthropik. I mean, you wait long enough and his investments, even though he stole them from his customers, worked out. But he was a big funder of effective altruism. And a lot of these companies are stocked with similar devotees to effective altruism. What it means is, you know, everything we do should save the world.

11:48So, you know, if we could do something today that could save a child 25 years from now or 50 years from now, we should do it. And they're worried about extinction-level threats, nukes or biohazards or runaway AI. I mean, it's just you can read the papers turned out by these effective altruism. They're a little kooky. And not realizing that technology in progress gets us to a point where we solve these problems over time anyway. But they like to throw money thinking about these problems today. Remember, OpenAI was founded as a nonprofit worried about AI safety. until Sam Altman, the CEO, said, you know, this is going to take a lot of money to do this right.

12:31And so we can't be a nonprofit. We need to be a for-profit so we can raise billions. Anthropic, the founders, said OpenAI is not doing enough based on safety. We're going to spin out and do our own company. And so they're focused on safety until it came along that it's really good at writing code and it's really good at designing chips and it's really good at customer service things. And they need to raise billions. And all of a sudden, the effective altruism stuff, you know, capitalism is OK, but they can't stop talking about saving the world. And I think that's one of their problems. So you mentioned the word kooky, and I mentioned the word scary.

13:11I mean, the kookiness comes from my understanding from reading about all this is that this is kind of this utilitarian philosophy that sort of argues that you use evidence and reason, and you rank everything in the world through these competing needs. You end up with very weird things, for instance, like where some of these papers suggest, like, why should a single country like the United States take precedence? Or why should a single species take precedence? You know, there's a lot of people out there worried about the future of shrimp farming, for instance. But I think the thing and the reason I call it scary is that the ultimate goal of effective altruists or AI safety individuals seems to be to limit average Americans' access to this new technology achievement for their own good.

14:03And that, to me, sounds like elitists essentially trying to tell me what I should or should not be able to have access to. Well, I mean, remember, Anthropic kept some of their technology away from the Pentagon because they were worried, what if there's an incoming missile? Do we get to decide whether to shoot it down rather than the Pentagon? It's just, well, it's scary kooky to coin a phrase between us. So, yes, in fact, even I remember there was a conversation, I think it was Sam Altman, who said, you know, we need to give this technology to the Chinese and the Russians to balance ours. Right.

14:40The U.S. And you go, well, who are you working for here? What are your goals? But I think there's a somewhat simpler explanation of some of the rhetoric. You know, Anthropics founder of Modi came out with a paper recently about safety and stuff asking to be regulated. I'm always suspect of anyone asking to be regulated because it's the worst thing in the world. And it's funny that the top three companies all quickly agreed. It was Anthropic and then Altman's OpenAI and Elon Musk with XAI. They said, yeah, yeah, yeah, we should be regulated. So what is it? Is it regulatory capture? Do they want to freeze the competition?

15:20Do they want to control the regulations? And then you have the politicians jump on it. You know, AOC, Alexandria Ocasio-Cortez came out yesterday saying, you know, these things should be owned and controlled by the government. I don't even want to hear that. I don't know how messed up it would be if these things were used for political reasons rather than the productivity. And yes, jobs are going to be destroyed, but even more jobs are going to be created using these productivity tools. So you said it before, we're sort of in the peak hysteria. And until there are more users of this, like, you know, programmers, that's all they do now.

16:00No one writes code anymore. where they use Claude at Anthropic and Code at OpenAI to tell it what they want. This is an enormous productivity enhancement. Oh, healthcare wrote something about a company called Viz.ai and they use Anthropic actually, but they have a system that's set up to flag in hospital patients, whether they are at risk of sepsis. And it saves lives from the day they put it in, it started saving lives. That's the kind of things that these companies should be talking about. And when he talks, oh, we're going to cure diseases in the next five to ten years. Well, get a little more specific.

16:39Like, what kind of things are you working on? Right now everyone worries about biohazards and some rogue agent is going to, you know, come up with the next COVID virus. That's not the use of these things. I don't even think that's possible. But to use computational and pattern recognition to cure disease, it's going to be the best thing that happened to mankind. It really is. Worst marketing departments in the whole world, in any industry, maybe ever. Yes. We're going to take one more break when we get back. More on the politics of the pushback against artificial intelligence. Everyone wants health care to be simpler and easier.

17:21That's why UnitedHealth Group is fast-tracking more than 60 % of children's procedures that used to need pre-approval, so more kids get the care they need when they need it. And speeding up procedure approvals and payments for more than 1 ,500 rural hospitals, so they can focus on caring for their communities. And that's just the start. Learn more at uhg.com slash access.

17:48From the opinion pages of the Wall Street Journal, this is All Things with Kim Strassel. Welcome back. AOC, I think, kind of gives the game away a little bit, right? In that it happens to be an industry. It happens to be a new technology, but she's just slotted it into her world framework anyway. Is there really any industry that AOC doesn't think should be run by the government and it would be better. So let's talk about the politics just a little bit. I think politicians are using this for different reasons in different ways. I think some Republicans at the moment are simply reacting to it because they're scared about getting on the wrong side of public opinion.

18:27I would put Texas Governor Greg Abbott in that category. He recently jumped on the anti-data center bandwagon. But I think there's a lot of Democrats who see political benefit in driving some of this fears? You know, Bernie Sanders has been a ringleader. He's out there with the Ban Artificial Super Intelligence Act. What are they hoping to get out of this? Like, what goals are they pursuing by doing this? Well, if 71 percent of Americans are against it, it's votes, first of all. Right. And there's a midterm coming up. And, you know, I see Chuck Schumer all over the place. You know, Trump is for this.

19:00We need to be against it. And you know, well, OK, how about you think it through a little better? I think the only good news to come out of this is no one's talking about climate change anymore. They're talking about data centers. So you go, okay, because, you know, that's an easier issue to argue against than the polar bears dying and icebergs melting and glaciers melting or whatever. Politicians will, as you know better than me, will jump on any issue that people agree with and use it to their advantage. And it is going to be a vote getter. I've heard from a lot of people since writing this piece and, oh, you don't understand, you know, they're putting data centers within 500 feet of neighborhoods and they have diesel fuel backup generators that are noisy and everyone, you know, so no more data centers, we can't have it.

19:49I mean, well, one, I'm not sure that's 100 % true. And second, get better politicians in your town, right, to fight these things and to create regulations and to push it further out of town. But in general, on a federal level, I think politicians are using this as an issue, as a wedge issue and 71 % agree with them. I'm not sure that that's going to change how people pull the lever in a month, but it's going to sway some percentage of voters. Well, you just made an excellent point because you were talking about climate change. And this has occurred to me too. I mean, people need to understand as well, at least in the political realm, that some of this is being driven by activists who've got this new issue, right?

20:30And issues are money-raising ventures. Climate was the big thing for a long time. You mentioned a name in your column that was such a blast from the past. I was like, really? She's still out there doing this. But you mentioned that Erin Brockovich, the famous Erin Brockovich, now has an AI data center reporting website. OK, you cannot tell me that this isn't now a thing out there, especially all of these kind of left-wing groups. This is like a new way to go out there and say, we're here, we're the group that's going to fight this. By the way, we really need your donation. I can't take credit for this, but someone pointed out that when Greta Thunberg comes out against data centers, then we know it's over, right?

21:13It was climate change, then it was taking a ship and headed towards Gaza. And so as soon as she flips to data centers, then the data center backlash will have peaked. Okay, so two, three years from now, where do you see all of this being? I mean, and I mean, do you think Americans come to understand the benefits of this? Do you think that the United States is still on? Or do you think we're in this kind of really scary moment where the public views of this could actually clamp down on it with all kinds of ramifications in terms of our ability to compete with China and successes that AI could bring to the country?

21:49No, it's a great question. And, you know, my view is that you can't hold technology back. that, you know, if you put up a roadblock, it just goes around the roadblock and squeezes it, or it takes a detour and goes around and will always end up being used. I mean, I started on Wall Street, and there were people in giant rooms with Mylar paper and X-Acto knives doing what they called spreadsheets. They were spread on tables. And then it was VisiCalc and then Lotus 1-2-3, and it destroyed those jobs, but it allowed millions to do financial models when in the past, it was just the lords of, you know, the numbers at the top of Wall Street.

22:30And so it's the same thing here. It's the democratization of an intelligent assistant. And that's what this is. Yes, there are going to be some cases where things get just automated and there are going to be jobs that are lost. And so for the next two to three to five years, we're always, because the press is going to amplify it, we're always going to hear about the job losses. Oh, this poor person was out of a job, but not quite the benefits of smart, not just white-collar workers, but blue-collar workers. In fact, we're using AI for productivity and enhancing their jobs and doing more with less.

23:08That's the productivity powerhouse that this thing is going to be. And so it's going to be a divide. I think there'll be more people on the positive side than there are today touting the benefits of AI, but the negatives are still going to be out there. The catastrophic, it's going to kill us all, I think will be muted over time. I mean, you can put a kill switch on these things. You know, it's still a machine that needs electricity. You can pull the plug. So I think the existential extinction threat, you know, that has peaked, but the turnover, it happens every cycle, the turnover of job types.

23:50Yes, it's going to destroy a lot of jobs, but we don't even know the new types of jobs that are going to be invented. Because if we did, you know, you just say, okay, I'm just going to invest in these and the hell with doing anything else. But those will become clear over time. But that's where the mood changes and the public perception of AI changes. And it'll happen over time. And I'm bullish of the uses. I think they're overbuilding. I don't think they need trillions of worth of data centers. That's a whole other discussion. But the negativity will ease. Well, if only the Luddites had known how many people would one day be working for Google and Apple, right?

24:28Andy Kessler, thank you for providing us a little bit of common sense and rationality in the middle of all the drama. We appreciate your coming on. We want to thank our viewers and listeners. We are here every week. If you'd like the show, please do hit the subscribe button and if you'd like to write to us you can at atkim at wsj.com

24:53everyone wants health care to be simpler and easier that's why united health group is fast tracking more than 60 percent of children's procedures that used to need pre-approval so more kids get the care they need when they need it and speeding up procedure approvals and payments for more than 1 ,500 rural hospitals so they can focus on caring for their communities. And that's just the start. Learn more at uhg.com slash access.