Steven Adler

Certainly, there are a lot of things that individuals can still do to have control over their tech lives, but a lot of the solutions to the problems that we're talking about now are these collective solutions that require people to vote or to organize.

I'm thinking about Our conversation just a few weeks ago with the former OpenAI researcher, Steven Adler, who was talking about the need to pay attention to what your elected representatives are saying about AI.

Ultimately, I think this needs to be an international coordination action.

And the types of things that a normal person can do, you could call your representative and tell them you're really concerned about this.

I feel like during that episode, I was slacking you and saying, like, I'm crying right now listening to this.

It might have been the closest I've come to crying listening in on one of your interviews.

Basically, he was talking about how researchers are using artificial intelligence to try to decode how animals communicate.

And not so much how we could talk to animals, but really the other way around of trying to understand what they're saying and sort of tune into their world.

MPs on the business committee have been hearing from former researchers who've left the big AI firms, such as OpenAI and Anthropic, over concerns about safety.

Steven Adler used to work at OpenAI, and he explained that researchers just couldn't stop pressing ahead despite knowing the risks.

Overall, I think this effort that people are racing toward to build super intelligence, these systems much more capable than humans, it's clear to me that we don't know how to do it safely, and yet the industry continues, and that concerns me.

AI bosses have themselves issued warnings about the risks of AI, and indeed admitted to instances where it had acted in unexpected and unasked for ways.

And we'll actually have a lot more on this story with the New York City's Council Speaker Julie Menin joining us right here on The Exchange tomorrow in a first on CNBC interview.

Next hour we'll also be speaking with former OpenAI safety researcher Steven Adler, and that'll be 2:00 PM Eastern on Power Lunch today.

Coming up, bet the net, on the names that is.

Jeff Kilburg buying cybersecurity stocks at the start of Q4, including this one, our mystery chart today.

OpenAI has not issued a separate public statement disputing the resignation as of today.

What's striking to me about Robinson's departure versus the long list of safety exits that preceded it, Jan Leike in May 2024, Daniel Kokotajlo that April, William Saunders, Miles Brundage, Steven Adler, is the specificity of his proposed fix.

He's not just saying slow down.

He's saying the fix has to be cultural and structural, and that new science is needed before these systems get any more capable.