Kai Risdall

I think if you look at the polls in the U.S.,

Two out of three people see AI as a threat, not an opportunity.

In China, there's much more readiness to download even models that might be a bit dangerous, like the open claw one, which was a big thing about a month ago.

When I was in China, there were photographs in the newspapers of mom and pop consumers lining up outside tech companies to get help from the engineers to install this open claw thing on their laptops.

Just to see what it could do, just for the heck of it.

Yeah, because it's an agent, and you can tell the agent to organize all the information on your laptop, and it's fun to play around with, but it's also dangerous, and they didn't seem to mind.

Hmm.

You suggest in this piece in The Times from the other day that really what needs to happen now is I'm paraphrasing, but work with me here.

Enough of the arms race.

What needs to happen is that China and the United States have to get together and negotiate some kind of global safety pact on artificial intelligence.

Is that fair as a fair characterization?

Pretty much.

I mean, some element of arms race and competition is inevitable, but I think you can try and do both at the same time.

You know, in the Cold War, that was when the agreement for the IAEA, the International Atomic Energy Agency, was done.

Then you go forward to the 60s, and despite the Cuban Missile Crisis, the Nuclear Non-Proliferation Treaty was agreed.

So you can have what's known as adversarial cooperation, and I think that's what we need to think about now.

Which I appreciate.

And of course, Reagan and trust but verify does come to mind.

But but Donald Trump and Xi Jinping are not Leonid Brezhnev and Ronald Reagan.

That's true.