Anthropic's RSP v3 quietly drops deployment-restriction conditions from earlier versions
dfrsrchtwts · x · 2026-09-18
A researcher digging into Anthropic's Responsible Scaling Policy found that RSP v2.0–v2.2 explicitly described conditions under which the company would restrict deployment of a model — reportedly the first time Anthropic said it might un-deploy a model in response to safety risks. In the latest RSP v3, however, those conditions no longer appear, a change worth noting for anyone tracking Anthropic's governance framework.
Related event: Anthropic's Updated RSP Drops Model Deployment Restriction Clauses(2 posts)→
More from Safety
- Security experts clash: two frontier models can now autonomously execute the full cyber kill chain — AccBalanced · 2026-09-18
- Stanford philosopher defends p(doom): subjective probabilities are perfectly legitimate — sethlazar · 2026-09-18
- Hugging Face hit by AI-led cyberattack; CEO says existing cyber laws may suffice — whurley · 2026-09-18
- Targeted attacks on prominent Rust developers use fake video calls to deploy malware — Simon Willison · 2026-09-18
- Steve Eisman: AI firms have no moats and are manufacturing a crisis to shape regulation — GaryMarcus · 2026-09-18
- AI text watermarking can make models more vulnerable to adversarial prompts — luisdans · 2026-09-18