Over the weekend, several current and former advisors to President Donald Trump on AI publicly exchanged insults at the country's leading AI companies. David Sacks, the president's AI and crypto 'czar' until March, branded Anthropic's models as 'lobotomized' and 'woke.' Emil Michael, a top Pentagon official, called OpenAI's new head of strategic futures a 'supreme village idiot.'
The trigger was the launch of Kimi, a free open-source model from Chinese company Moonshot. Kimi appears to rival the intelligence of models from OpenAI and Anthropic, which are far from free. This situation is creating a deep rift among Trump's AI strategists.
Kimi disrupts the US AI market
Every time a smart, free Chinese model like Kimi is released, US companies see less reason to pay for access to Anthropic or OpenAI models. Given that enthusiasm for these AI companies drives an outsized share of economic growth, Chinese models create both economic and political problems for the president. Anton Leicht, a fellow at the Carnegie Endowment, wrote on X that they are 'a threat for an administration that really doesn't want more economic bad news.' US stocks have already been rattled.
Sponsored Protocol
This tension comes just a week after New York imposed the country's first state ban on new data centers. There is growing distrust of AI companies, and many Americans may have little sympathy for OpenAI or Anthropic as they fend off cheaper competitors, saying it's not the government's job to protect their interests.
Factions clash over government intervention
David Sacks, now in a non-formal role, criticized top AI companies that 'want the government to eliminate their open source competition.' He also argued that Chinese AI models are popular because they come with fewer restrictions on use (aside from state censorship). However, his position has been replaced by one that sees a larger role for government intervention. This view fuels the new White House review process to vet AI models for security before release.
Sponsored Protocol
Dean Ball, a former Trump advisor now at OpenAI, criticized the process as a 'de facto licensing regime for frontier AI.' Ball predicted Trump might solve the problem with soft power, making US companies afraid to use models like Kimi. This drew a response from Michael, who called Ball the 'supreme village idiot,' bristling at the suggestion of a clandestine scheme.
The role of chips and distillation
How Kimi became so good remains unclear. Export controls on chips to China have loosened, with Trump allowing Nvidia to sell more chips in exchange for government cuts. But China has limited computing power, and it's not known which chips Moonshot used. Possibly distillation, a practice where AI models train on outputs of existing models, was involved. OpenAI and Anthropic have long complained about this, and in April the Trump administration announced efforts to curb it.
Sponsored Protocol
Kimi is now available for free and is nearly as good as the Anthropic model that the US government considered so powerful it was briefly shut down for national security threats. The weekend's sparring suggests many in Trump's orbit see this as a wake-up call, but no one agrees on what to do.
For more on the evolving AI coding landscape, check out the article on Beyond grep: the new paradigm for context-rich AI coding harnesses. The current debate is also covered in the original piece from MIT Technology Review.
