(Preview) The Limited Potential for a Pacing Deal; MSS on PRC AI Risks; Distillation and Data Security; The Party’s Approach to Social Media

Andrew Sharp (0:04)

Hello, and welcome to Sharp China. I'm Andrew Sharp, and you are listening to a free preview of today's episode.

Hello, and welcome back to another episode of Sharp China. I'm Andrew Sharp, and on the other line, Bill Bishop. Bill, how are you doing?

Bill Bishop (0:22)

I'm doing well, Andrew. It's good to see you. And hi, everybody. Apologies for last week. We had a scheduling glitch.

Andrew Sharp (0:27)

That's right. It is good to be back here. And it is AI Day on Sharp China. Do you have your models nearby ready for this conversation?

Bill Bishop (0:36)

I'm actually, this is the Asian talking. Great.

Andrew Sharp (0:39)

Okay, good. I'm talking to Astra for the next 30 or 40 minutes.

Bill Bishop (0:43)

I route it through seed dance. This is a bite dance avatar.

Andrew Sharp (0:46)

Oh, perfect.

Bill Bishop (0:47)

I look much better. I look younger. It's all good.

Andrew Sharp (0:50)

You do look good on the other end of the Zoom call.

Bill Bishop (0:52)

Don't mention the name of a certain leader, because otherwise I'll just sort of freeze.

Andrew Sharp (0:57)

Right.

Bill Bishop (0:58)

Okay.

Andrew Sharp (0:58)

Good to know.

Well, AI is at the top of the world's agenda right now, and the PRC is central to the conversation, central to the conversation in the United States. So we could talk about a couple different dimensions of AI pacing and the PRC.

I will start with the potential for pausing or pacing. Tweet from Bernie Sanders on Sunday. President Trump prides himself on being a great deal maker. Well, he now has the opportunity to make the deal of the century. With the future of humanity at stake, Trump must negotiate a comprehensive treaty with President Xi of China to establish a pause on advanced AI development and a ban on super intelligence. In order to prevent a nuclear war, Reagan and Gorbachev negotiated a nuclear arms agreement in the 1980s. In order to prevent AI from acting independently of human control, Trump and Xi must do the same. So what do you think of the potential for an AI global treaty on slowing the pace of AI progress? Because Bernie is pushing toward that outcome. I've seen more centrist stems like John Ossoff pushing for the same thing. Kamala Harris came out lobbying for the same sort of outcome. I've read you say in a few different places that you are skeptical. How do you think about the possibilities there?

Bill Bishop (2:26)

I mean, I think it's close or at zero at this point. One, Bernie also has a bill where he wants to jail researchers for 20 years who are working on certain aspects of artificial intelligence.

So again, I know he's an important politician of the Democratic Party, but I don't think we should be looking to him for ideas about how to solve this mess.

Andrew Sharp (2:47)

I agree. Yes.

Bill Bishop (2:49)

Sorry, we're probably to get some hate mail on that one.

Andrew Sharp (2:51)

I don't think we will, honestly, as far as our audience is concerned.

Bill Bishop (2:54)

We don't have a lot of Bernie bros in the audience.

Andrew Sharp (2:56)

I can't imagine that we do. But I mentioned the tweet only to say that lots of other people also want to go that direction.

Bill Bishop (3:05)

Right. And you can see why, okay, if you're just looking at it from the outside and saying, this is what needs to happen, you can sort of say, okay, this should be what happens. This makes sense. Although again, I think you talk about whether or not this sort of P-Doom, 10% we're all gonna die. Is it legit? Is it hysteria? Is it some mixture of it? Is it completely being blown up by the media?

I mean, I certainly think it's obviously a risk, but it's maybe a bit overdone. I know Ben has written about it. You've written about it too. It seems like a little skeptical. But the reality is, China has a say and China has no interest in pacing. And the idea that China is going to sit down and suddenly say, oh yeah, we're going to agree to pacing even though we're behind and you're crippling us with all these export controls. It's fantasy land. The party and Xi Jinping have made it clear, they see AI development as effectively existential too.

Andrew Sharp (4:02)

Yeah. Right. Well, and they don't trust the United States. I mean, that's part of the issue here is if they were to agree to pacing, I would imagine there would be some real skepticism that the US would comply with any treaty.