← Back to the board · Anthropic · Table

Anthropic transparency Met

Publish sabotage risk reports for future frontier models

Anthropic committed at the Claude Opus 4.5 launch to publish sabotage risk reports for future frontier models; the first such report (covering Opus 4.6) was published on 2026-02-10.

  1. Committed 2025-11-24
  2. Due for each future frontier model clearly exceeding Opus 4.5
  3. Evaluated 2026-02-10
  4. Ruling Met

Why this ruling

Commitment made at the Opus 4.5 launch; first report fulfilling it covered Opus 4.6 (2026-02-10).

Sources
Anthropic ↗ · archived
Committed text
“publish sabotage risk reports for future frontier models”
Committed
2025-11-24
Reviewed
2026-06-18

Cite this commitment

copy-paste
Overdue. "Publish sabotage risk reports for future frontier models." Overdue, 2026. https://overduetracker.org/c/anthropic-sabotage-reports (retrieved 2026-08-05).