clm_0C963ABC8B39valid
OpenAI and Anthropic share findings from a joint safety evaluation
- valid_from
- 2025-08-27
- valid_to
- null (still holds)
- published
- 2025-08-27
- confidence
- 0.50
- superseded_by
- null
- recorded
- 2026-10-27 (first held by AsOf)
- checked
- 2026-10-29 · the publisher's own feed
- last rechecked
- not yet rechecked against a source
Sources
/claim/clm_0C963ABC8B39.md/v1/claims/clm_0C963ABC8B39timelinesubmit a correction