← Back to the board · Anthropic · Table

Anthropic evaluations Partial ⚠ contested ⚠ under review

Capability re-assessment cadence in the Responsible Scaling Policy

The RSP set a regular re-assessment cadence; a 2026-04-02 update extended a three-month evaluation interval to six months, citing rushed elicitation.

  1. Committed 2023-09-19
  2. Due regular re-assessment cadence per the RSP
  3. Evaluated 2026-04-02
  4. Ruling Partial

Why this ruling

The interval was extended from three to six months; current policy frames Risk Reports as every 3–6 months. Whether this is a relaxation is debated.

Sources
Anthropic ↗ · archived
Committed text
“regular re-assessment of model capabilities”
Committed
2023-09-19
Reviewed
2026-06-18
Contested
This ruling is genuinely disputable.

Cite this commitment

copy-paste
Overdue. "Capability re-assessment cadence in the Responsible Scaling Policy." Overdue, 2026. https://overduetracker.org/c/anthropic-eval-cadence (retrieved 2026-08-05).