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 · October 20, 2026

Monitoring evaluations: test the deployed detection and response path

analysis Model EvaluationAgent SecurityAI Red Teaming
Why it matters

METR’s limited-access review of Anthropic’s August risk report questions whether isolated covert-capability benchmarks represent evasion of deployed monitoring. It recommends testing the complete monitoring system and considering influence on safety decisions. METR’s low-risk judgment for the covered models rests mainly on limited capabilities; it is not evidence that monitoring would detect higher risk.

My takeaway: Use an isolated replica of the deployment to test monitor coverage, escalation, and containment together. Include harmful actions framed as ordinary work and gaps between monitored components. Save traces and independently verify whether alerts arrived and actions stopped. Treat missing detection as a test failure, and keep model self-reviews separate from independent evidence.
Keep exploring

More curated notes connected through Model Evaluation and Agent Security.

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.

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.

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.