BETA RELEASE

Summary

Voluntary safety commitments from leading AI organizations to identify, assess, and manage severe risks associated with frontier AI models.

Key quotes

In the extreme, organisations commit not to develop or deploy a model or system at all, if mitigations cannot be applied to keep risks below the thresholds.

The document lists participating organizations including OpenAI, Google, Meta, and Microsoft. It outlines three primary outcomes: risk management, accountability, and transparency.