OpenAI Frontier Governance Framework

This document describes how OpenAI governs the safety of its most advanced AI systems, covering serious risks like cyberattacks, biological and chemical weapons, harmful manipulation, and AI operating outside human control. It sets out that OpenAI has processes for incident response, model reporting, and security risk management. The framework is not fixed — OpenAI will update it as AI capabilities and regulatory requirements change.

The OpenAI Frontier Governance Framework establishes OpenAI's structured approach to identifying, assessing, and mitigating the most serious risks from advanced AI systems, built on the Preparedness Framework as its foundational instrument. The framework covers four primary risk categories — cyber offense, CBRN risks, harmful manipulation, and loss of control — and extends to security risk management, incident response, and model reporting. OpenAI commits to updating the framework as model capabilities, evaluations, and regulatory requirements develop. The document also positions OpenAI's safety and security practices in alignment with the California Transparency in Frontier AI Act and the EU AI Act's Code of Practice for General Purpose AI.

For an individual user, this document describes the internal governance commitments OpenAI has made around the most serious risks posed by its frontier AI systems. It establishes that OpenAI maintains structured processes for risk assessment, incident response, and model reporting across risk categories including harmful manipulation and loss of control. The framework is subject to change over time as capabilities and regulatory requirements develop, meaning the governing rules users can reference today may be revised.