Frontier AI Needs an Aerospace Flight Halt | Rep. Whitesides
MTS
November 3, 2026
Representative George Whitesides discusses the reality of recursive self-improvement inside frontier labs, his push to monitor automated AI R&D, and his call for a mandatory one-month pause on frontier model development modeled on aerospace flight halts.
George Whitesides
In that context, when we have a big incident or accident, we stop flying, we try to understand what happens, so we fully understand it, we write down how we're going to operate safely going forward, and only then do we return to flight. And to me, it's very obvious that the AI development sector could use more lessons from industries where human health and safety has always been a crucial part of the industry, you know, whether that's airlines or aerospace or the nuclear sector or whatever.
To me, there's a lot that can be learned from those other sectors.
Theo
And we are back, we are live with Congressman George Whitesides, who represents California's 27th Congressional District and is vice ranking member of the House Science, Space and Technology Committee. He's introduced several bills around AI, which we'll talk about today. Congressman, welcome.
George Whitesides
Hey, thanks for having me on.
Theo
Absolutely. So we'll start with the non AI bill, which is the No Bets on Wildfires Act, which you introduced recently, a bipartisan bill to ban betting on wildfires on prediction markets. There's some people who would say that, you know, the prediction markets perform an important function here by allowing people to better understand and price the risk of future wildfires or purchase hedges on the prediction markets to protect themselves in the event of wildfire damage, kind of like how insurance works. So what would you say to the people who would criticize your bill on those grounds?
George Whitesides
Well, thanks.
It's nice to be on your show. I represent a district on the north side of LA. County, where we have a ton of wildfires. And three days after I got sworn in in 2025, we had the devastating Eaton Canyon and Palisades fire in Los Angeles County, where we lost 16,000 homes. I've been working ever since to help those folks rebuild and to protect our communities more. Look, here's the problem. People can start wildfires, you know? And so they can bet on something, and then they can actually affect that bet by, you know, setting a new fire or whatever. And that's just like totally wrong. Furthermore, I have a fundamental belief that, you know, people should not be betting on the misfortune and suffering of others.
Wildfire is one of the most interesting, is unique in natural disasters, because a single human being can cause a massive wildfire by starting it on a day where there is high winds, low humidity, and dry brush. And so we felt it was just absolutely crucial to make clear that this was prohibited activities and would be punished severely if anybody does it. Keep in mind, you know, that over a million dollars was gambled, was bet on those devastating LA wildfires. And I just think that's totally wrong.
Theo
Right. So tell us about also your understanding AI in the Classroom Act. What specifically do you think government should be learning about how AI works in schools?
George Whitesides
Well, thanks a lot. I mean, I have a couple of kids. They're 14-year-olds and 16-year-olds. I'm living through it. All the different challenges of having kids growing up in this new era of AI. And, you know, for all the parents out there, they're seeing what we're seeing, which is that teachers are no longer assigning as much homework because, particularly written assignments, because kids are using the AIs to write their stuff. And then they'll use different AIs to make their stuff less AI-like.
And so more and more work is being done in the classroom. That's sort of one way to fix it, but it does not, you know, fix the problem that in order to develop our brains, we need to be thinking about an essay or a problem set for, you know, sometimes for hours. And that's, you know, what you need to do as adults. So I think we need to understand what's happening first. So this thing, this basically just says we need to study it.
It says that the NSF can look at attention span and learning independence, cognitive, social and emotional behavioral development, whether there's over-reliance by students on artificial intelligence, and see how different types of students respond to AI. I think the first step is to understand how people are using these tools, how kids are using these tools, and how it's affecting them as they grow up and develop into independent thinkers. So that's what the, that's what the bill would mainly do.
Theo
Right. What do you think some preliminary results might be?
George Whitesides
Well, I don't know. I mean, I don't know if you probably know that people have been measuring IQ for over a century, and for most of the 20th century, global IQs were going up.
Read the other 13 minutes of this episode here.
$15 gets you this episode and 99 more. Credits never expire. After checkout, one link brings you back here to read it.
Not what you expected? Full refund within 14 days.
Thousands of transcripts fetched by people building searchable podcast archives
Fetch transcripts with code
The same credits work from a script or an AI agent. The demo key returns a sample episode in full, no card needed:
curl -H "x-api-key: pt_demo" \
https://spoken.md/transcripts/1000651996090Markdown with the speakers named, for your notes, your knowledge base, or anything that makes HTTP calls.
From $0.10 per transcript. No subscription. Credits never expire. Prices exclude VAT, added at checkout for EU customers. Not what you expected? Email us within 14 days with 20 or fewer credits used and we refund the pack in full.
Using your own key:
curl -H "x-api-key: YOUR_KEY" \
https://spoken.md/transcripts/1000799501336