Anthropic and OpenAI seek tighter control over external AI evaluator publications
Sources reported that Anthropic and OpenAI have asked for substantial control over what outside evaluators can publish about their systems, fearing that detailed findings might reveal trade secrets. The companies argue that unrestricted disclosure could expose proprietary techniques and give competitors an advantage.
Key points
- Anthropic and OpenAI are reportedly seeking greater control over external AI evaluator publications
- The firms cite risk that evaluator findings could expose trade secrets
- The move could limit public benchmark data and reduce transparency for researchers
The request comes as external audits and benchmark studies become more common in the AI field. Analysts say tighter publication rules could limit the amount of performance data that reaches the public, potentially reducing transparency for researchers, investors and regulators.
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Policy & Regulation
All →- Graylark’s Raven identifies homes from interior photos, per demo · 1 src
- Microsoft CEO Nadella urges treating AI models as compromised by default · 2 src
- Google expands SynthID detector to check all partner AI content globally · 1 src
- Trump orders federal agencies to rename AI as 'super intelligence' · 1 src
- AFL orders Sportsbet to halt AI-generated player content without consent · 1 src
Comments
via GitHub Discussions