Top AI Firms Oppose Ban on Chinese Open-Source Models

Top AI companies are speaking out against a potential ban on Chinese open-source AI models. Nvidia, Meta, OpenAI, Microsoft, and others have signed an open letter titled "Open Weights and American AI Leadership." Nvidia CEO Jensen Huang posted on X, stating that open models strengthen safety, cybersecurity, accelerate innovation, and enable sovereignty. OpenAI's president, Greg Brockman, also expressed his belief in access and that "having more models is a good thing." The Trump administration is reportedly considering a ban to prevent cybersecurity incidents and maintain the U.S.'s leading edge in AI. This concern grew after Beijing-based Moonshot AI released its Kimi K3 model, which is widely available and rivals American models. Michele Catasta, president of Replit, a signatory, believes a ban would set a bad precedent and that Chinese models could introduce price pressure, which he sees as beneficial. Interestingly, Anthropic did not sign the letter. This discussion highlights a key tension between national security concerns and the AI industry's push for open access and innovation.