Meta, Nvidia, and Microsoft oppose proposed restrictions on open-weight artificial intelligence models, arguing that restricting access could slow innovation and reduce competition in the AI industry.
Tech companies spoke publicly to lawmakers on Friday in support of open-source AI models, sparking a new debate in business and policy circles over who controls the powerful technology.
A coalition of 25 major technology companies, venture capital firms, and foundations, including Meta, Microsoft, Nvidia, IBM, Dell Technologies, and Palantir, have signed an open letter urging U.S. policymakers not to restrict or impose premature restrictions on “open-weight” AI models.
In a letter posted on X and signed by 20 companies and organizations, including Meta Platforms and IBM, NVIDIA CEO Jensen Huang said lawmakers should avoid “premature restrictions on open models that stifle competition or drive innovation abroad.”
The joint statement, titled “Open Weight and American AI Leadership,” comes as the U.S. government debates national security measures against foreign competitors and sophisticated Chinese open models like Moonshot AI’s Kimi K3.
This letter directly challenges the idea that closed models are inherently secure. Supporters of the restrictions have raised concerns about potential risks associated with widely available AI models, including abuse, security threats and challenges in controlling advanced technology.
The debate comes as governments around the world consider how to regulate rapidly advancing AI systems, while balancing innovation and safety measures.
Industry leaders argue that the open weight system will help startups, researchers, and small businesses compete with big AI companies by lowering the barrier to developing AI-powered applications.
Open Weight also allows a broad global community of researchers to examine model behavior, identify vulnerabilities, and build safeguards to avoid the single points of failure inherent in closed source systems.
Meta, Nvidia, and Microsoft argue that the responsible development of openweight AI can strengthen the broader technology ecosystem and support advances in areas such as research, enterprise applications, and AI infrastructure.
The paper explicitly compares today’s open model discussion to the open source software movement of the 1980s, arguing that open ecosystems are what ensured America’s long-term technology advantage.
The letter follows recent warnings from US authorities about the “distillation” of secret models and theft of IP (intellectual property).
In response, the coalition argued that while open weights carry clear risks, banning them would be the wrong policy response and would only encourage innovation overseas.
Notably, closed model leaders OpenAI, Anthropic, and Google did not sign the open letter.
