Skip to content
Longterm Wiki
Back

RSP Evaluation Framework

web
Center for AI Safety·safe.ai/responsible-scaling-policies

Credibility Rating

4/5
High(4)

High quality. Established institution or organization with editorial oversight and accountability.

Rating inherited from publication venue: Center for AI Safety

This URL is a dead link (404) on the CAIS website; the intended content about Responsible Scaling Policies is not accessible here and should be sourced from original lab publications instead.

Metadata

Importance: 10/100organizational reportreference

Summary

This page on the Center for AI Safety website was intended to provide an overview of Responsible Scaling Policies (RSPs), frameworks that AI labs use to tie capability thresholds to safety commitments. However, the page currently returns a 404 error, indicating the content has been moved or removed.

Key Points

  • •Page is currently unavailable (404 error) on the CAIS safe.ai domain.
  • •RSPs are frameworks adopted by AI labs (e.g., Anthropic, OpenAI) linking capability evaluations to deployment decisions.
  • •The intended topic — RSPs — is a key governance mechanism for managing frontier AI risks.
  • •Users seeking RSP information should consult primary lab documentation (e.g., Anthropic's RSP directly).

Cited by 1 page

PageTypeQuality
AI Governance Coordination TechnologiesApproach91.0

Cached Content Preview

HTTP 200Fetched Mar 20, 20260 KB
![](https://cdn.prod.website-files.com/63fe96aeda6bea77ac7d3000/63fe9a72750bc67b129df210_CAIS%20Logo.svg)

## Page Not Found

The page you are looking for doesn't exist or has been moved

[Home page](https://safe.ai/)
Resource ID: 3daf7680ca3e48e3 | Stable ID: YTFkMDg4Zj