Why the panic over Chinese AI is overblown

The latest AI model from Moonshot AI, called Kimi, has once again triggered heated debate about whether the United States is losing its edge to Chinese competitors, and whether open-source or proprietary AI is the better path forward.

Social media exploded with commentary, but the conversation appears to be just as intense behind closed doors in Washington, D. C. Reports indicate that OpenAI and Anthropic have been lobbying regulators, warning about the risks posed by open Chinese AI models.

On a recent episode of TechCrunch’s Equity podcast, my colleagues Kirsten Korosec, Sean O’Kane, and I explored why this topic keeps igniting such strong reactions. Sean suggested that some people might benefit from stepping away from their keyboards rather than spending entire weekends arguing on X. He observed that, in many ways, “this feels like we’re seeing repeats of prior freakouts,” with Silicon Valley constantly “expecting that something is going to arrive and blow everything else away.”

Kirsten raised a pointed question about the real beneficiaries of heavy restrictions on Chinese AI: “Are we accelerating and ensuring that Americans win the AI race, or are we ensuring that certain frontier labs do better than others?”

Below is an excerpt from our discussion, edited for clarity and length.

Anthony Ha: For anyone following the discourse around Chinese AI, this pattern is extremely familiar. It happened with DeepSeek. A Chinese model launches, performs well on some benchmarks, appears competitive with frontier models, and a significant portion of the tech industry reacts with alarm.

This latest round received extra attention because an OpenAI executive was among those posting about it. But the recurring question remains: Can Chinese companies outperform U. S. companies, at least in certain areas, while doing so more cheaply and more openly?

Sean O’Kane: There are many elements here that feel like repeats of earlier panics. One of my favorites is how ready everyone is for something to arrive that will blow everything else away. This past week, people were showing off that Kimi supposedly replicated macOS in 30 minutes. And yes, it produced an impressive graphical reproduction of what macOS looks like. But it is not an operating system.

We keep seeing this pattern. Everyone in the tech industry is so jumpy. With these Chinese models in particular, there is this expectation that drives the reaction we saw last weekend. (Also, by the way: Go outside, touch grass. It was the weekend, and everyone was trading barbs on Twitter all weekend.) But this jumpiness is fascinating because now we are a week out, and I do not think anyone feels the end is nigh like they did a week ago.

Kirsten Korosec: We have a great story by our reporter Tim Fernholz that tries to unpack this psychosis in the United States. He points to several reasons. I do not want to conclude for him, but I think one rises to the top.

There are concerns that Chinese open-weight models might have an implicit bias toward China. There are worries about security risks and guardrails. But a big idea driving this fear is protectionism and who will “win the race” , the U. S. or China. That seems to be the core driver.

Anthony, do you agree?

Anthony: Completely. The China aspect always adds a certain level of hysteria. That is not to say people should not be concerned about how the U. S. stacks up against China across different industries. But the reaction gets so amped up.

This reminds me of the discussion around TikTok a few years ago. I did not think those concerns were entirely made up, but the level of panic was extreme. As soon as you add the word “China” to any discussion, things ramp up dramatically. In this case, it is linked to the debate about open weights and the idea that AI is so powerful and dangerous that only proprietary models from American frontier companies can control it.

Obviously, most people saying this have reasons for saying it. David Sacks, who was the AI czar for the Trump administration and now has a different role, was shouting on X about how he cannot believe people oppose data centers, that there is too much regulation. It is a way to argue for positions they already held. “My gosh, if China beats us, that is unthinkable, so you have to do what I want to do anyway.”

Kirsten: Right. If you put across-the-board bans on Chinese open-weight models , and I am not saying there are no real concerns , but let us play that out. Doing so would benefit models created by OpenAI, for instance, and would force enterprises to use those instead of models like Kimi.

So you really have to ask: Are we accelerating and ensuring that Americans win the AI race, or are we ensuring that certain frontier labs do better than others?

Sean: At this point, we should note that much of this discussion was kicked off by Dean Ball, the head of strategic futures at OpenAI. He was the first to publish a long post mentioning these concerns.

Part of me thinks the reaction came because people disagreed with what Dean wrote. Part of me also thinks the reaction was driven by the fact that he kind of just said the thing out loud. He basically said the U. S. should create regulatory FUD , fear, uncertainty, and doubt , to muck up the ability for these open-weight models to compete with the U. S. (Ball later backed away from this argument.)

To me, you could read in some of the responses: “You are not supposed to say that out loud, Dean.”