OpenAI Preparedness Framework (v1 beta 2023; v2 2025)
OpenAI · 2023-12-18 (v1 beta); 2025-04-15 (v2)
OpenAI's internal system for measuring dangerous capabilities before release. Its 2025 update gave the company explicit room to relax safeguards if competitors do, which critics read as a loophole.
What it calls for
- Safety evaluations
- Transparency
- Human oversight
Scope
Single company; tracked categories: bio/chem, cyber, AI self-improvement (persuasion category dropped in v2)
What actually happened
The April 2025 rewrite added a clause allowing OpenAI to 'adjust' safeguards if a rival ships a high-risk system without comparable protections, removed persuasion as a tracked category, and shifted toward automated evaluations; the Financial Times reported testers were given under a week for some models (https://techcrunch.com/2025/04/15/openai-says-it-may-adjust-its-safety-requirements-if-a-rival-lab-releases-high-risk-ai). OpenAI has published system cards with Preparedness scorecards for each frontier release and signed the EU GPAI Code of Practice (Aug 2025). The Midas Project's Watchtower logs framework changes (https://www.themidasproject.com/projects). Looked for: any instance of OpenAI delaying a launch under the framework - none publicly documented.