On July 24, 2026, 25 leading US tech companies and institutions including Microsoft, NVIDIA, OpenAI, Meta, Hugging Face, and IBM issued a joint statement endorsing open-weight AI models and warning against restrictive policies [1, 2].

Open-weight AI models allow users to download, inspect, modify, and run AI models on their own infrastructure, making advanced AI more accessible and adaptable, the coalition said [1, 3, 2]. They argued that open weights enable wider innovation, increase competition, reduce risks through transparency, and help spread AI benefits broadly across sectors [1, 3, 2]. An unnamed coalition statement emphasized, "Open weight models make advanced AI more accessible, adaptable, and widely available" and stressed that US leadership depends on building a strong, open ecosystem rather than focusing on a single frontier AI model [3].

The coalition cautioned that banning open-weight models would be counterproductive. They urged comprehensive safeguards instead to manage risks. "Against prohibiting open-weight models, we advocate comprehensive protective measures to address possible risks," they stated [1]. They warned that relying solely on closed-source AI would create dangerous single points of failure and concentrate power in few hands [1, 2].

Open weights also allow more researchers and developers to identify vulnerabilities, improve safeguards, and continuously enhance AI models, the coalition pointed out [1, 2]. The statement noted that "distillation," using a model's output to train other models, is common and should not be confused with illegal misappropriation [1, 2].

Startups, universities, public institutions, and businesses can build on advanced AI without costly retraining thanks to open weights, expanding access and innovation [3, 2]. The coalition warned that premature restrictions could stifle US innovation or drive AI development overseas [2].

The coalition referenced the 1980s open-source software movement as a precedent where openness led to widespread innovation and institutional technology sovereignty [3]. A related blog post compared open-weight AI's growth to Kubernetes as a neutral, extensible platform enabling broad ecosystem development, calling for US engagement rather than isolation [4]. "Open-weight AI is becoming the foundation for the next AI ecosystem. The US should compete in it, not wall itself off," the author wrote [4].

The coalition’s July 24 statement marks a coordinated effort from major players to influence US policy on AI openness and regulation as debate over AI's future continues [1, 3, 2].