OpenAI, Anthropic, and Google DeepMind have been in discussions since July 2026 to establish an industry-led organization for AI safety standards. The initiative aims to create a framework for shared technical evaluations, pre-release audits, independent testing, and standardized safety protocols.
Demis Hassabis, CEO of Google DeepMind, proposed the structure on July 14, 2026, suggesting a model similar to the Financial Industry Regulatory Authority (FINRA). This organization would be industry-funded and governed, with the authority to enforce safety requirements across the sector.
Jakub Pachocki, chief scientist at OpenAI, supported this approach in an essay on September 6, 2026, advocating for uniform safety requirements. By mid-September 2026, the companies began regular working-group meetings to define benchmarks and auditor access to models.
This collaboration builds upon the 2024 Frontier Model Forum, which previously included Microsoft. Notably, OpenAI, Anthropic, and Google DeepMind chose not to join the Nvidia-led Open Secure AI Alliance, which launched in July 2026 with over 30 participating firms.
The effort is led by Dario Amodei of Anthropic, Sam Altman of OpenAI, and Demis Hassabis of Google DeepMind. The three companies currently lead the AI Safety Index published by the Future of Life Institute, which monitors industry safety commitments.
