Chinese company Moonshot AI has released a new version of its Kimi model, and the internet is once again pretending to be shocked that a foreign entity can do math. The Kimi K3, according to Moonshot, "still trails the most powerful proprietary models, Claude Fable 5 and GPT 5.6 Sol" - but hey, it's open source and "demonstrated frontier-level performance" on their tests. Independent analyses from Arena.ai and Vals AI confirm it's competitive with the big boys.

The announcement, timed conveniently with Chinese President Xi Jinping's speech at the World AI Conference in Shanghai, sent Wall Street into a tizzy. The Nasdaq dropped about 1% on Friday as investors frantically sold chip stocks like Nvidia, presumably fearing that AI competition is now a two-horse race and one horse is state-backed.

Tech figures recycled their greatest hits from the DeepSeek drama in January 2025, but now with extra seasoning from Trump's tariff war and the ongoing saga of whether Anthropic is a national security threat. David Sacks, the former AI czar turned science advisor, warned that the US is "tying itself in knots" with regulations while China races ahead. He also took a gratuitous swipe at Anthropic, calling Claude an example of "woke lobotomized models" - because nothing says serious policy analysis like a culture war jab.

Travis Kalanick, former Uber CEO and professional grudge-haver, complained that Chinese companies are "distilling off" American models - i.e., training on their outputs. He conveniently ignored that American models have also been built on Chinese ones, specifically Kimi. Fair play, apparently, is only fair when it's one-sided.

OpenAI's head of strategic futures, Dean Ball, offered a more nuanced take: Kimi is "a very good model" whose performance can't be explained away by distillation. He then predicted that open-weight models lead to "full AI communism," where AI becomes state-provided digital infrastructure. "This future strikes me as a dystopian hellscape," Ball said, presumably while sipping artisanal coffee in a San Francisco office funded by venture capital. He suggested the Trump administration could create "regulatory risk" around Chinese models without banning open source - just enough FUD to scare regulated enterprises.

But Shakeel Hashim, editor of Transformer, argued the panic is overblown. Kimi "likely does not have dangerous cyber capabilities," and the Chinese government will eventually face the same incentives to restrict open models once they get too powerful. So, in other words: panic now, or panic later - your choice.