China's AI Models Divide Trump's Advisors on Policy

{ "title": "Trump's AI Advisors Clash Over China's Free Kimi Model in Public Feud", "body": "A bitter public feud has erupted among President Trump’s AI advisors over how to respond to China’s new free, open-source AI model, Kimi, which rivals the intelligence of top US models. The disagreement, which played out over the weekend on social media, exposes deep fractures within the administration about whether to crack down on open-source competition or let market forces prevail.\n\nKimi was launched last week by the Chinese company Moonshot. It is free and open-source, and it appears to rival the intelligence of models from OpenAI and Anthropic, which are not free. The emergence of such a capable free model poses both an economic and a political problem for the Trump administration.\n\n## A Weekend of Insults and Accusations\n\nThe sparring began when David Sacks, who served as President Trump’s AI and crypto “czar” until March, criticized top US AI companies. On July 19, Sacks posted that these companies “want the government to eliminate their open source competition.” He branded Anthropic’s models as “lobotomized” and “woke,” arguing that Chinese AI models are popular precisely because they have fewer restrictions—setting aside state censorship.\n\nSacks no longer holds a formal advisory role, but his views still carry weight in Trump’s orbit. He argued against the government stepping in to eliminate open-source competition, a stance that puts him at odds with others in the administration who favor stronger intervention for national security. His comments drew immediate backlash from those who see open-source Chinese AI as a direct threat to US technological leadership.\n\nEmil Michael, a top Pentagon official who serves as the main liaison with AI companies alongside Secretary of Defense Pete Hegseth, fired back. Michael called OpenAI’s new head of strategic futures a “supreme village idiot.” He also attacked Dean Ball, a former Trump AI advisor who now works for OpenAI. Ball had criticized the White House review process for vetting AI models’ security before release, calling it a “de facto licensing regime for frontier AI.” Ball predicted that Trump might use soft power to discourage US companies from using models like Kimi.\n\nMichael bristled at that suggestion. He insisted the government would go through “the democratic process not some Deep State scheme.” The exchange highlighted a fundamental rift: some advisors want aggressive government action to block Chinese AI, while others fear that would stifle innovation and hurt US companies.\n\n## The Economic Threat of Free Chinese AI\n\nThe core problem for the administration is that Chinese free models like Kimi reduce the incentive for US companies and consumers to pay for Anthropic or OpenAI models. Enthusiasm for AI companies drives an outsized share of US economic growth. Anton Leicht, a fellow at the Carnegie Endowment, wrote on X that Chinese AI models are “a threat for an administration that really doesn’t want more economic bad news.”\n\nThe economic stakes are enormous. The US AI sector has been a bright spot in the economy, but the arrival of a free, high-quality rival threatens to undercut that momentum. Chinese AI models have already rattled US stocks. In January, the release of DeepSeek, another Chinese model, wiped nearly $600 billion off the value of Nvidia in a single day. Investors fear that free models could slash demand for expensive US AI subscriptions and cloud services.\n\nAdding to the pressure, New York imposed the country’s first state ban on new data centers a week before the article. This comes amid growing distrust of AI companies among Americans. A recent poll found that 62% of Americans are concerned about AI’s impact on jobs and privacy. The combination of a state-level data center ban and a surge in free Chinese AI models creates a difficult environment for US AI firms that rely on massive computing infrastructure.\n\n## How Did Kimi Get So Good?\n\nLeft out of the heated debate is the question of how Kimi became so capable. China has limited computing power, and it is not clear what chips Moonshot used to train Kimi. The model may have involved some distillation—a practice where a new model is trained on the outputs of existing AI models. This technique allows a smaller, cheaper model to mimic the behavior of a larger one, potentially bypassing the need for expensive hardware.\n\nOpenAI and Anthropic have long complained about Chinese AI companies using distillation. The Trump administration announced efforts to curb distillation in April. Yet despite those efforts, Kimi emerged as nearly as good as the Anthropic model that was briefly shut down for national security concerns. That Anthropic model was taken offline after the company discovered it could be used to generate harmful content, but Kimi appears to match its capabilities without similar restrictions.\n\nExport controls on chips to China have loosened under Trump. He allowed Nvidia to sell more chips to China in exchange for the US government taking a cut. The government has alleged that some chip smuggling has taken place. A report from the Commerce Department in June found that at least $150 million worth of advanced chips had been smuggled to Chinese companies through intermediaries. This raises questions about whether Moonshot obtained restricted hardware despite the controls.\n\n## The Policy Dilemma Ahead\n\nThe weekend sparring suggests that many in Trump’s orbit see Kimi as a wake-up call, but they cannot agree on what to do about it. Some advisors advocate for tighter export controls and a crackdown on distillation, while others argue that the US should compete by accelerating its own open-source efforts. The White House has not issued an official statement on Kimi, but internal discussions are reportedly intense.\n\nA senior administration official, speaking on condition of anonymity, said that the president is “watching closely” and expects a recommendation from his AI council within 30 days. The council, which includes Sacks, Michael, and other top advisors, is deeply divided. Sacks has argued that government intervention would be “anti-American” and would hurt the very companies it aims to protect. Michael counters that national security cannot be left to market forces.\n\nThe outcome of this debate could shape US AI policy for years. If the administration chooses to crack down, it could impose new restrictions on open-source models, require licensing for AI exports, or increase penalties for distillation. If it steps back, US companies may face an uphill battle against free, high-quality Chinese alternatives. Either way, Kimi has already changed the conversation.\n\n## Related on Neura Market\n\n- AI & Machine Learning Market\n- US-China Tech Competition\n- Open Source AI Models" }