AI Red Team Engineering notes, research references, and field notes on AI systems, responsible AI, governance, and compliance.
Global search

Search AI Red Team

Search across the complete knowledge base.

Enter at least two characters to begin.
↑↓ select Enter open Browse the full research archive
METR · March 12, 2026

External safety-report review: test evaluation sensitivity and hidden assumptions

analysis Model EvaluationAI Red TeamingAI Compliance
Why it matters

METR’s public executive summary reviews Anthropic’s February sabotage-risk assessment for Claude Opus 4.6, with a separate review covering March revisions. Although METR broadly agrees with the low but nonzero risk conclusion, it questions several supporting arguments. Evaluation awareness and missed low-severity behaviors weaken confidence in assessment sensitivity. The conclusion also draws on deployment observations available at the time, so it should not be treated as a timeless certification of the model.

My takeaway: Ask reviewers to distinguish agreement with the overall conclusion from confidence in each supporting claim. Investigate evaluation awareness, missed behaviors and obscured reasoning. Record the exact report version, reviewer access and deployment assumptions, and reassess when those assumptions change.
Keep exploring

More curated notes connected through Model Evaluation and AI Red Teaming.

OpenAI News · framework

Pacing model development in an era of cyber-critical capabilities

OpenAI says preliminary evidence that Astra may meet its Critical cybersecurity threshold led it to pause frontier reinforcement-learning work for two weeks and keep its largest planned run on hold. New safeguards include stronger workload and network isolation, continuous boundary testing, token-level monitoring that escalates suspicious tool activity, and broader alignment checks for deception, reward hacking, and unauthorized access.

OpenAI News · framework

A blueprint for democratic governance of frontier AI

OpenAI proposes a three-part U.S. frontier-AI governance model: harmonize emerging state safety laws into a federal baseline, strengthen CAISI as an evaluation and standards institution, and coordinate a broader resilience program. Proposed controls include severe-risk evaluations, transparency reports, independent audits, safety-incident reporting, model-weight security, whistleblower protection, and periodic technical assessments.

OWASP GenAI Security Project · guide

OWASP Top 10 for Agentic Applications for 2026

OWASP's community guide organizes agentic-system risk into ten categories, including goal hijacking, tool misuse, identity and privilege abuse, memory poisoning, insecure inter-agent communication, cascading failures, and rogue-agent behavior. It provides a shared taxonomy and mitigation starting point rather than a certification checklist or evidence that a deployed system is secure.