Chinese company Moonshot AI released a new version of its Kimi model this week, and the reaction quickly spilled into the usual fight over who gets to control the future of AI. The announcement came as Wall Street flinched, with the Nasdaq dropping about 1% on Friday as investors sold off stocks in chip companies like Nvidia. The people at the top of the machine saw a threat. The people who’ll pay for the scramble are everyone else.
Moonshot said Kimi K3 "still trails the most powerful proprietary models, Claude Fable 5 and GPT 5.6 Sol," but that the new open source model "demonstrated frontier-level performance across our evaluation suite, consistently outperforming other tested models." Independent analyses from Arena.ai and Vals AI also suggested that Kimi is competitive with flagship frontier models. That’s the core tension here: a Chinese company pushing an open source model into a field dominated by proprietary systems, while the gatekeepers of capital and state power rush to decide what the public is allowed to use.
Who Gets to Decide
The announcement coincided with a speech from Chinese president Xi Jinping at the World AI Conference in Shanghai. The article says the news seemed to spook Wall Street, and the market reaction landed fast. Investors sold off chip stocks. The Nasdaq slid. The apparatus of finance doesn’t need much to panic when a new model threatens the profit streams built around scarcity, control, and locked-down systems.
The debate echoed the uproar that followed another Chinese company, DeepSeek, when it released its open source R1 model in January 2025. The article says the discussion now feels heightened after the Trump administration’s tariff war with China, repeated fights over the national security threat supposedly posed by Anthropic, and as major AI companies prepare to finally go public. That’s the same old script: state rivalry, corporate fear, and public policy shaped around protecting power, not people.
David Sacks, the Trump administration’s former AI czar and now co-chair of the President’s Council of Advisors on Science and Technology, framed the issue as a race the United States can’t afford to lose. He said the U.S. is "tying itself in knots: politicians and bureaucrats are banning new data centers, piling on state regulations, and pushing for new federal agencies to pre-approve frontier models. This is how you lose the AI race." He also took a dig at Anthropic, calling Claude an example of "woke lobotomized models" that are "the enemy American competitiveness."
The Gatekeepers Talk Control
Travis Kalanick, the former Uber CEO, echoed complaints that Chinese are "distilling off" American AI models, meaning they are being trained on the outputs of those models. Kalanick wrote, "If distillation isn’t enforced against, then everyone should be able to distill from everyone else.. otherwise one arm [would be] tied behind American models’ backs." The article adds that American models have also been built on top of Chinese ones, specifically Kimi. So the hand-wringing about theft sits right beside the reality that these systems already borrow from each other across borders, while the bosses argue over who gets the bigger slice.
OpenAI’s head of strategic futures Dean Ball said Kimi is "a very good model" whose performance probably can’t be "explained away by distillation or anything like that," and added that he’s "personally surprised the Chinese state continues to allow the open sourcing of models this good, given potential risks." Ball suggested that the "probable outcome of an open-weight-model-dominant world is full AI communism," where AI is treated as "a 'public good' which will ultimately be provided by the state as a kind of 'digital public infrastructure.'"
Ball said, "This future strikes me as a dystopian hellscape, but I’ve never met an open-weight models advocate who doesn’t ultimately concede this is where things end." He also suggested that the Trump administration, which he used to work for, will eventually realize it needs to "create large amounts of regulatory risk around the use of open-weight Chinese models."
That’s the playbook in plain language. Not open debate. Not democratic control. Just manufactured fear, soft law, and pressure until regulated enterprises back off.
What They Call Protection
Ball said, "You don’t need to 'ban open source' (one of the dumber motifs of AI policy discussion). You just need to direct every agency to issue soft law that creates FUD [fear, uncertainty, and doubt]. 'A Federal Reserve Advisory Bulletin found that there may be backdoors in Chinese AI models.' It needn’t be that well justified. You just create enough regulatory risk that every regulated enterprise backs off."
That quote lays out the machinery with unusual honesty. The state doesn’t always need a clean ban. It can lean on agencies, advisory bulletins, and vague warnings until the market does the dirty work itself. Control without the headline. Repression with paperwork.
Shakeel Hashim, editor of the AI-focused publication Transformer, pushed back and argued that much of the worry is overblown because Kimi "likely does not have dangerous cyber capabilities," and because the Chinese government will face "extremely similar incentives" to restrict open Chinese models once they develop those capabilities. Even that response stays trapped inside the same framework, where governments decide what gets opened, what gets closed, and what risks the public is supposed to absorb.
The whole fight circles back to the same hierarchy. Corporate executives, former officials, and state-linked policy voices argue over regulation, competitiveness, and national security while ordinary people get the bill, the surveillance, and the fallout. The model is open. The power structure isn’t.