OpenAI and Anthropic are urging the US government to review the risks associated with powerful open-weight AI models. The companies have collaborated in this effort, highlighting their concerns to policymakers and emphasizing the potential dangers these technologies pose. Previously, they participated in voluntary pre-release reviews of their advanced models with the Trump administration, which focused on addressing cybersecurity concerns.

OpenAI: OpenAI develops and deploys frontier AI systems including the GPT family of models. In the reported development, the company is actively advocating for expanded US government review processes of advanced AI models in coordination with policymakers. Recent engagement includes collaboration on cybersecurity vetting frameworks amid administration directives.
Anthropic: Anthropic builds AI systems with a focus on safety and reliability, centered on its Claude models. It is partnering with OpenAI to urge greater US oversight and pre-release reviews of powerful AI models. This push aligns with its recent compliance and input on voluntary government security assessments for new releases.

Policy Alignment: OpenAI and Anthropic are jointly warning US policymakers about risks associated with powerful open-weight AI models.
Government Review: The companies have participated in voluntary pre-release reviews of their advanced models with the Trump administration for cybersecurity concerns.