OpenAI and Anthropic Lobby US Regulators for Stricter Open-Source AI Model Restrictions

OpenAI and Anthropic urge US regulators to tighten open-source AI controls, citing security risks and competition from Chinese models lowering deployment barriers.

Woofun AI reports that OpenAI and Anthropic have engaged in closed-door communications with US regulatory authorities to advocate for stricter restrictions on open-source AI models. The companies argue that fully open-sourced model weights increase misuse risks as capabilities approach human levels, while also highlighting competitive pressure from Chinese models that allow global developers to deploy advanced AI without high API costs. OpenAI seeks a government-led safety assessment system with mandatory tests, while Anthropic warns that open-source proliferation lowers the threshold for dangerous capabilities, necessitating tighter management measures.