Huang: US Should Use Chinese Open-Source AI
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Jensen Huang told Axios that Chinese open-source AI models are "excellent" and should be used by American companies, calling it a direct challenge to Trump officials and US labs lobbying Washington to shut them out.
- Huang dismissed fears that Chinese AI could displace US companies, saying "There's no scenario where China runs U.S. companies off the road" — "Zero possibility."
- Kimi K3, released by Beijing-based Moonshot AI, triggered a sell-off of Nvidia and other chip stocks because it combined near-frontier performance, dramatically lower prices and downloadable open weights.
- Huang argued cheaper open models expand the market for AI hardware, stating "Free AI should be great for hardware. Free AI should be great for chips. Free AI should be great for data centers."
- Huang rejected the "backdoor" security argument against downloaded Chinese models, saying companies can sandbox them and that openness makes AI more secure by letting outside researchers expose weaknesses.
- Huang called for Anthropic to release its restricted cyber model Claude Mythos to "everyone" as a service, arguing "Holding Anthropic back is not in the benefit of the United States."
- Treasury Secretary Scott Bessent told Fox Business the administration is examining Chinese AI models for stolen US intellectual property and considering sanctions, citing "watermarks" of US models inside Chinese ones.
Why it matters: Huang's defense puts Nvidia publicly at odds with two of its largest customers — OpenAI and Anthropic — who are actively lobbying Washington for restrictions. The conflict lands just as Treasury's Bessent signals potential sanctions, meaning Nvidia's "free AI equals more chip demand" thesis now depends on regulators siding with Huang rather than the labs paying his bills.
