Anthropic grants independent evaluators employee-level access to AI models
Anthropic has announced it is granting independent evaluators from the non-profit organization METR employee-level access to its artificial intelligence models and internal safety systems. This initiative, reported across 14 outlets following Anthropic's own update, aims to enhance AI safety through external scrutiny. The evaluators will be able to "look under the hood" to monitor potential risks and assess the effectiveness of Anthropic's safety measures. While the core fact of granting this unprecedented access is widely reported, the specific operational details and the full scope of the evaluators' powers are primarily based on Anthropic's own statements. This move is presented as a significant step towards greater transparency and accountability in AI development.
13 articles from 13 outlets covered this story. No difference in framing or figures was found between them. The underlying claim is sourced from a press release.
What do all outlets agree on?
13 outlets covered “Anthropic grants independent evaluators employee-level access to AI models”. All of them report the following:
- Anthropic granted employee-level access to its AI models and safety systems
- Access provided to independent evaluators from non-profit METR
- Purpose is to enhance AI safety and monitor risks
Did outlets disagree about this?
No. All 13 outlets covering “Anthropic grants independent evaluators employee-level access to AI models” reported it the same way — no difference in framing or figures was found between them.
Which outlets covered this?
All 13 articles found on this story, grouped by the stance of the piece. Every link goes to the original publisher.