OpenAI just dropped its Frontier Governance Framework, a public document that explains how its safety protocols line up with new laws like California’s AI Transparency Act and the EU AI Act. It’s basically a transparent version of their internal Preparedness Framework, which already tackles the worst-case scenarios from advanced AI—think cyberattacks, bioweapons, manipulation, and even losing control of the system. The framework covers risk assessments, model reporting, security, incident response, and getting outside experts to weigh in. OpenAI says this is a living document that will evolve as AI capabilities and regulations change. Why it matters: This is a rare, detailed look at how a leading AI lab plans to stay compliant with emerging global regulations, setting a precedent for the whole industry.