Tech Giants Push Back Against Proposed Restrictions on Open-Source AI Models

A coalition of 25 major technology firms, led by Nvidia and Microsoft, has publicly opposed government restrictions on open-source artificial intelligence models, arguing that such limitations would harm innovation and American competitiveness. The letter, signed by executives including Jensen Huang of Nvidia and Satya Nadella of Microsoft, draws a historical parallel to the open-source software movement of the 1980s, suggesting that premature regulation would repeat past policy mistakes.

The Coalition's Core Argument

The signatories—including Palantir, Hugging Face, Perplexity, IBM, and others—contend that open-weight models strengthen safety, cybersecurity, and innovation by allowing external researchers to audit code and identify vulnerabilities rather than relying on proprietary systems controlled by a small number of firms. The letter emphasizes that open models expand economic access for startups and universities, preventing artificial intelligence capabilities from becoming concentrated among a handful of dominant companies.

The statement also defends distillation—the practice of training one model using another's outputs—as a legitimate research technique that should not be conflated with intellectual property theft, even as this remains contested among policymakers.

The Chinese Competition Factor

The letter arrives amid growing U.S. government scrutiny of Chinese open-source models. Moonshot AI, a Chinese startup, released a model called Kimi K3 on July 16 that demonstrated superior performance across certain industry benchmarks compared to leading American offerings from proprietary developers like OpenAI and Anthropic. U.S. Treasury Secretary Scott Bessent indicated that the Trump administration would investigate whether Chinese companies were misappropriating American intellectual property, stating the government possesses "the ability to sanction them because of this theft."

The tech coalition's letter directly addresses this concern, warning that overly restrictive policies would "stifle competition or drive innovation overseas," potentially undermining American technological leadership rather than protecting it.

Notable Absences

Notably, OpenAI and Anthropic did not sign the letter, despite their prominence in the artificial intelligence sector. Both companies are preparing for potential initial public offerings valued at nearly one trillion dollars each, and both have confidentially filed prospectus documents with the Securities and Exchange Commission. Elon Musk, who operates an artificial intelligence business under his rocket company SpaceX, amplified the letter on social media with his "full support," though SpaceX did not formally sign the statement.