- What changed
- OpenAI, Anthropic, and xAI cosigned the AEF-1 standard, committing to provide embedded third party evaluators ongoing access to verify safety practices.
- Why you should care
- Standardized third party evaluation creates structured pathways for external verification of frontier model training and safety processes.
- Your move
- Watch. Monitor how labs implement embedded evaluator access in practice.
- What to watch next
- Subsequent disclosures detailing the operational agreements and access terms between labs and embedded evaluation teams.
- Event
- policy
- Event date
- Sep 15, 2026
- Relevant to
- General AI readers
