Understanding the panic surrounding China’s AI

The announcement of the latest AI model by a Chinese company (Kimi Moonshot AI) has reignited the debate over U.S. competitiveness and open vs. proprietary AI.

While there has been much discussion on social media, this discussion also appears to be happening behind the scenes in Washington, D.C., where OpenAI and Anthropic have reportedly approached regulators with concerns about China’s open model.

On the latest episode of TechCrunch’s Equity podcast, Kirsten Kolosek, Sean O’Kane, and I discussed why this is a hot topic. In addition to suggesting that certain people should “touch grass” rather than spend their weekends debating about

And Kirsten pointed out that imposing strict limits on China’s AI models could primarily benefit a small number of companies.

Continue reading excerpts of the conversation, edited for length and clarity.

Anthony Ha: For people who have been following the discourse around AI in China, this is probably very familiar since the launch of DeepSeek, which is basically a Chinese model. It appears to perform similarly in some benchmarks, or at least compete with some Frontier models. And some in the tech industry are losing their minds.

Part of this (discussion) came under even more scrutiny because one of the people who posted about it was an (executive) at OpenAI. But in general, there is (is) a recurring question: “Can Chinese companies beat American companies on at least some fronts, and can they do it cheaper and in a more open way?”

Sean O’Kane: Yeah, there are a lot of elements in this that feel like we’re watching a repeat of previous anomalies. One of my favorites is, “Everyone is so ready and expecting something to arrive and blow everything else away.” I think my favorite example from last week was when people were bragging, “Oh my god, you made a perfect replica of macOS in 30 minutes.” Yes, it’s a very impressive graphical recreation of what macOS looks like, but it’s not an OS.

We keep seeing things like this happen over and over again because everyone is so excited in the tech industry. And especially with some of the Chinese models coming out, there’s this expectation, and I think that goes to the heart of why people reacted the way they did last weekend. (By the way, go out and touch grass, it’s the weekend. Everyone in the industry was exchanging rants on Twitter all weekend.) But this jump was very interesting to me. Because another week has passed and I don’t think anyone feels as close to the end as they did a week ago.

Kirsten Kolosek: There’s a really great article by one of our reporters, Tim Fernholz. He’s trying to figure out the psychosis on this here in the United States. He cites several reasons for this. And the bottom line, I don’t want to give him any conclusions, but I think there are some who come out on top more than others.

There are concerns that these Chinese indiscriminate weight models have implicit bias against China, and there are other concerns about safety risks and guardrails. But there’s also a pretty big idea here. That’s protectionism. Who quotes or quotes “beat the competition”? Will it be America or China? And that seems to be causing a lot of the fear.

I don’t know, Anthony, do you agree with that?

Anthony: I completely agree. I think the Chinese aspect always adds a certain kind of hysteria. And that’s not to say people don’t have to worry about how the United States stands up against China in various industries. But it is greatly amplified.

Another thing this reminded me of was the discussion around TikTok from a few years ago. Again, it’s not that I thought the concerns about TikTok were completely fabricated. The problem was how panicked people were. As soon as you add the word China to any discussion, things seem to accelerate dramatically. And in this case, it links to this discussion about openness and this idea that AI is so powerful and so dangerous that the only way it can be controlled is by using the proprietary models that these American frontier companies provide.

Obviously, most people who say this have a reason for saying so. David Sachs, who was the Trump administration’s AI czar (and now has a different role in the Trump administration), exclaimed on X, “I can’t believe people are against data centers. We’re locking ourselves in. There’s too much regulation.” So this is a way of asserting the position they already had on AI. “Well, if China beats us, that’s unthinkable, so I just have to do what I want.”

Kirsten: Yeah, and I’m not saying there aren’t real concerns if you were to ban the Chinese open-weight model altogether, but let’s give it a shot. If we did that, it would benefit models created by OpenAI, for example, and force companies to use those models instead of using models like Kim’s.

So you really need to ask questions. Are we accelerating and ensuring that Americans win the AI ​​race, or are we ensuring that certain frontier laboratories do better than others?

Sean: At this point, I have to say that a lot of this discussion actually started with Dean Ball, Head of Strategic Futures at OpenAI. He was the first to publish this very long post addressing some of these concerns.

Part of me thinks the reaction to this was because people didn’t agree with what Dean wrote. Part of me thinks the reaction may have been caused by the fact that he just said that out loud. He basically said that the US should create regulatory FUD (fear, uncertainty, and doubt) and cripple the ability of these open-ended models to compete with the US. (Ball later withdrew from this discussion.)

And I think, for me, you can read some of people’s reactions like, “Dean, you shouldn’t say that out loud.”