Anthropic's Responsible Scaling Policy version 2.0, effective October 15, 2024, and its 2.x revisions list cyber operations among capabilities requiring ongoing assessment rather than as a formal capability threshold with required safeguards. The policy says Anthropic would consult cyber experts, consider tiered access controls or phased deployment for models with advanced cyber capabilities, and run pre- or post-deployment testing.
Why it matters
It shows how one major lab treated autonomous exploit development as a monitored risk without committing to a cyber-specific ASL trigger during 2024-2025.
Key facts
As stated in the sources, with where to find them.
- Cyber Operations is defined as the ability to significantly enhance or automate sophisticated destructive cyber attacks, including discovering novel zero-day exploit chains, developing complex malware, or orchestrating extensive hard-to-detect network intrusions.RSP v2.2, capabilities under Ongoing Assessment table
- The ongoing-assessment entry mentions possible tiered access controls or phased deployments for models with advanced cyber capabilities.RSP v2.2, Ongoing Assessment column
- Version history: v2.0 effective Oct 15, 2024; v2.1 Mar 31, 2025; v2.2 May 14, 2025.RSP updates page, version table
Findings that cite this record
No tracked finding cites this record yet.
Key questions this bears on
- How are attackers using AI agents in real operations?To run large parts of intrusions under human direction, including espionage, credential theft and extortion; reported autonomy and success vary.
Sources
Related records
May 22, 2025
Sep 22, 2026
Feb 13, 2026
Oct 6, 2026
Oct 26, 2026
Jun 2, 2026