Menu
Sign In Search Podcasts Libraries Charts People & Topics Add Podcast API Blog Pricing
Podcast Image

TechStuff

The Man Who Wrote the AI Textbook Says We're Heading For Extinction - The Story

24 Jun 2026

Transcription

Transcript generated automatically by AI and may contain errors.

Chapter 1: What is the main topic discussed in this episode?

0.031 - 19.761 Brendan Boyle

This is an iHeart podcast. Guaranteed human. Run a business and not thinking about podcasting? Think again. More Americans listen to podcasts than add supported streaming music from Spotify and Pandora. And as the number one podcaster, iHeart's twice as large as the next two combined. Learn how podcasting can help your business. Call 844-844-iHeart.

0

35.772 - 59.967 Oz Voloshin

Welcome to Tech Stuff, I'm Oz Veloshin. Our guest today is a world-renowned expert in the field of AI. In fact, he literally wrote the book on it. Stuart Russell, together with co-author Peter Norvig, released Artificial Intelligence, A Modern Approach, back in the mid-1990s. Since then, it's been translated into 14 languages and is used in 1,500 universities in 135 countries.

0

60.548 - 81.157 Oz Voloshin

But just because he wrote the book doesn't mean he can't criticize the technology, or more specifically, the way it's being rolled out, often without regard to consequence. Stuart has spent the past few years sounding the alarm about the lack of AI safety protocols and what he sees as the very real possibility that our current path could lead to human extinction.

0

81.698 - 99.186 Oz Voloshin

The conversation comes at a timely moment as the US government is forcefully cracking down on who can and can't use the leading models coming from Anthropic. Today, Stuart is a professor of computer science at Berkeley and the president of the International Association for Safe and Ethical AI.

0

99.807 - 113.333 Oz Voloshin

He also recently took the stand as the only expert witness on AI for Elon Musk in his lawsuit against Sam Altman and OpenAI. Stuart, welcome to Tech Stuff. It's a pleasure to be with you.

113.971 - 143.615 Stuart Russell

Well, it was more what I was not allowed to say. I was not allowed to talk about anything related to existential risk, which was surprising because OpenAI was set up for that reason. Musk and others were concerned that if AI were in the hands of for-profit companies, that they would disregard safety and put humanity at risk. So OpenAI was created to counter that.

144.357 - 172.273 Stuart Russell

The judge said, no, you can't talk about that. which was disappointing. Why did the judge say you couldn't talk about existential risk? I was not privy to those discussions either. Maybe the defense lawyers thought it would be prejudicial, that it was in some sense speculative, because we haven't yet gone extinct. And this is a strange argument I hear from a lot of people.

173.154 - 182.024 Stuart Russell

It's just science fiction. Well, what do you mean by that? Well, it hasn't happened yet. You know, everything that's ever happened, there was a time before it happened. Before it happened.

Chapter 2: What prompted Stuart Russell to change his views on AI?

182.064 - 205.892 Stuart Russell

And by your argument, nothing could ever happen because everything was at some point science fiction. But, you know, when you look at science fiction, you know, they talked about nuclear weapons in 1912, H.G. Wells. They talked about space travel in the 19th century. And in fact, they talked about AI in the 19th century. Samuel Butler wrote a book that described...

0

205.872 - 217.35 Stuart Russell

society where there had been an enormous conflict between those who were in favor of the machines and those who predicted that the machines would be the ruin of the human race.

0

217.364 - 246.409 Oz Voloshin

Now, is OpenAI particularly bad? I mean, I know you were asked by their counsel in Cross, you know, if you believe that the for-profit sort of motivation to recklessly develop AI by definition endangers, you know, humanity, surely that also applies to Elon and XAI and SpaceX. And you said, if that hypothesis is correct, then yes. In other words, that the same critique could apply to

0

246.389 - 250.293 Oz Voloshin

OpenAI or Anthropic or Google or SpaceX?

0

250.914 - 274.337 Stuart Russell

Yeah, it was specifically not my job to compare the safety records of different companies or their safety positions. I think, if I understand it correctly, Anthropic is a public benefit corporation, which is one way of allowing considerations other than profit to affect the decisions made by management and the board.

274.52 - 301.388 Stuart Russell

Because for a regular for-profit company, there is a legal obligation to maximize shareholder return. And what's happening here is that the risks imposed on the rest of humanity are externalities, as economists call it, which means that someone is making a decision and there's a bad consequence that is being loaded onto somebody else.

301.368 - 331.696 Stuart Russell

So you think about chemical companies who skimp on safety, and as it stands, according to the companies, these are all externalities, meaning they don't accept responsibility for these consequences. Those are harms that don't figure into their balance sheet. And so the same would be true, in a sense, for human extinction. And even liability would not really be a deterrent for that, right?

331.736 - 348.585 Stuart Russell

Because obviously you wouldn't be around to pay the compensation. So they just sort of factor that out of their decision-making. And that's exactly what OpenAI was set up to avoid, but now it's with the transition to a for-profit entity. It's part of that calculus.

349.506 - 368.972 Oz Voloshin

So take us back in time. In 1995, you wrote this textbook that became the defining textbook on AI. And you came up with a concept called the standard model. Can you explain what that is and how it relates to the conversation which you're now so engaged in today as to the potential extinction of the human race because of AI?

Chapter 3: How does Stuart Russell define the standard model of AI?

1963.775 - 1985.968 Stuart Russell

I'm a professor at Berkeley, so I've been doing this in my little research center with a few grad students. And I've come to the conclusion that, yeah, probably if I had a few billion and a few hundred of the absolute top engineers and vast amounts of compute, I could probably make more progress on this. So that's one idea. But basically, that's the research track.

0

1986.927 - 2016.748 Stuart Russell

The other track is the regulatory track, right, which is government policy. And you mentioned what recently happened with Anthropic's Mythos and Fable models. Let's just roll back a few weeks to when Mythos was first made public. And Mythos is the latest version of Anthropic's large language models, and it's able to carry out end-to-end cyber attacks without human assistance.

0

2017.808 - 2046.516 Stuart Russell

And either it or its soon-to-come successors would basically be, I think I just wrote an article in The Garden where I said it's a weapon of mass cyber destruction. And you're putting those weapons of mass cyber destruction in the hands of a billion people. What could possibly go wrong, right? And all of a sudden, the U.S. government, which had been

0

2046.648 - 2074.645 Stuart Russell

on a deregulatory binge, basically trying to crush anyone who talked about regulation or talked about AI safety, suddenly said, well, why did nobody warn us about these AI systems? He said, well, we sort of have been warning you, but anyway, they got the message. And then they followed a sort of very messy process that eventually led to an executive order, which was pretty weak.

0

2074.745 - 2099.065 Stuart Russell

It basically said, companies can voluntarily submit their systems to the government for testing 30 days beforehand. public release. And the executive order says explicitly, this is absolutely not a licensing regime. It's absolutely not putting any obligatory hurdles in the way of American innovation, blah, blah, blah.

2099.827 - 2132.713 Stuart Russell

But then a couple of days later, Amazon tells the government, oh, we found some ways of jailbreaking mythos. or Fable, which I guess is Fable is the sort of defanged version of Mythos. And they said, oh, look, you know, we can jailbreak Fable and make it do some cybersecurity things. And the government shuts down both Fable and Mythos. So they put in effectively de facto licensing

2132.693 - 2157.643 Stuart Russell

architecture where they said, look, if it doesn't meet these standards of being safe, then we're shutting it down, right? That's exactly what a licensing architecture is. And just to be clear, licensing architectures exist for buildings, for food, for hairdressers, uh, for, uh, airplanes, right? We, you don't get in an airplane until it gets certified by the FAA.

2157.723 - 2163.612 Stuart Russell

You don't go in a building until it's been inspected for the building code, et cetera, et cetera. So this is normal.

2163.632 - 2177.213 Oz Voloshin

So is this the moment you've been waiting for? Is this, I mean, is this the culmination of what you've been advocating for? Can you, can you go back to, to, to research rather than regulation as your full-time profession or, or is this a hint of a, of a changing of the guard or what do you make of this moment?

Comments

There are no comments yet.

Please log in to write the first comment.