The world’s most valuable chipmaker endorsed the idea of American companies chatting with Chinese-made artificial intelligence models.
Nvidia CEO Jensen Huang brushed aside concerns about open-source Chinese AI being used by Americans and potentially stealing intellectual property in a Tuesday interview for Axios’ “Behind the Curtain” series. Beijing-based Moonshot AI released Kimi K3 on July 16, triggering market panic as chip and semiconductor stocks sold off amid fears that cheaper Chinese AI models could undercut demand for chips and data centers, according to Forklog.
“These Chinese models are excellent,” Huang, the Tawianese-born CEO of the American AI giant, told Axios. “Open-source models that are excellent should be used.”
“If everything just becomes one single model, one single point of attack, one single source of failure, I think the world is much, much more vulnerable,” Huang told the outlet, arguing that AI “openness” makes it more secure than restricting it.
Nvidia’s stock dropped from $212.50 to $207.40 when Kimi K3 launched, according to StatMuse.
Nvidia and the White House each did not immediately respond to the Daily Caller News Foundation’s requests for comment.
“The market’s misunderstood the impact of DeepSeek the first time. It’s misunderstood the impact of Kimi again, this time,” Huang said. “If there’s great AI, even if it’s open, wherever it comes from, there will be more use … You will have to sell a lot more Nvidia computers, … build more data centers, … have more services.”
Anthropic, which announced a multi-billion-dollar partnership with Microsoft and Nvidia in November 2025, has been outspoken in its opposition to China-based AI models.
“We have identified industrial-scale campaigns by three AI laboratories — DeepSeek, Moonshot, and MiniMax —to illicitly extract Claude’s capabilities to improve their own models,” Anthropic said in a February announcement, naming three Chinese AI companies.
“[Claude] Mythos should be available as a service … Just because Mythos is not available, open models are available anyhow. So I think: Let Anthropic run,” Huang said about Anthropic’s restricted model, Axios reported.
The Trump administration is investigating claims that Chinese AI companies used distillation to commit “IP theft” and train their models on stronger U.S. models, CNBC reported.
“We have information that Moonshot AI distilled Anthropic’s Fable for the development of its K3 model,” Michael Kratsios, director of the White House Office of Science and Technology Policy said on X. Moonshot created a “sophisticated internal platform” to distill U.S. AI models, Kratsios alleged.
“If we see, especially that overseas models are stealing from our great companies, we have the ability to sanction them because of this theft,” Treasury Secretary Scott Bessent told Fox Business Tuesday, according to CNBC. Bessent said U.S. AI models’ “watermarks” were found inside Chinese models.
“Distillation, learning from AI, learning from other sources of knowledge, is fundamental to intelligence,” Huang argued, according to Axios. He said misconduct should be targeted instead of the AI models themselves, and companies should face consequences for breaking the law.
Kimi K3 is a free, open-weight AI model that can be downloaded and modified by the user. The top six AI models, based on token usage, are Chinese-made, according to OpenRouter’s usage rankings.
“There’s no scenario where China runs U.S. companies off the road,” Huang told Axios. “Zero possibility.”
(DCNF)
