Define ASL-4 safeguards before reaching ASL-3

Anthropic’s Responsible Scaling Policy v1.0 stated its commitment was to write the ASL-4 measures before any model reaches ASL-3 capabilities.

Committed 2023-09-19
Due before any model reaches ASL-3
Evaluated —
Ruling —

Why this ruling

The original v1.0 trigger is documented. The trigger has since elapsed — Anthropic activated ASL-3 with Claude Opus 4 on 2025-05-22 — and v3.0 (Feb 2026) restructured away from the ASL-4 framing; whether the loosely-specified ASL-4 commitment was satisfied is genuinely disputed.

Sources: Anthropic (RSP v1.0) ↗ · archived
Anthropic (context) ↗ · archived
Committed text: “define ASL-4 evaluations before we first train ASL-3 models”
Committed: 2023-09-19
Reviewed: 2026-06-18
Contested: This ruling is genuinely disputable.

Cite this commitment

copy-paste

Overdue. "Define ASL-4 safeguards before reaching ASL-3." Overdue, 2023. https://overduetracker.org/c/anthropic-asl4-before-asl3 (retrieved 2026-08-05).