The AI Evaluator Forum has published AEF-1, a proposed baseline for independent third-party AI evaluations covering access, conflicts of interest, funding relationships, recusal, and transparency. This standard emerges alongside Dario Amodei's proposal for frontier AI companies to commit to ongoing, employee-like access for embedded third-party evaluators.
- AEF-1 establishes requirements for independence, including rules on conflicts of interest and funding.
- Anthropic is unilaterally committing to providing unparalleled access to its safety teams, including office desks and internal tools.
- The broader debate includes calls for pacing progress versus focusing on control and containment.
- GitHub added auto model selection tiers to Copilot/Codex workflows.
- Cline launched a native desktop app for working with open-weight models.
The publication of AEF-1 signals a move toward formalized self-regulation and independent verification within the industry.