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
Microsoft Security Blog · October 16, 2026

Open-weight model assessments: separate provenance, safeguards, and deployment risk

news Model EvaluationAgent SecurityAI Compliance
Why it matters

Microsoft and NVIDIA propose mapping deployment risks to owners, acceptance criteria, evaluations, and mitigations. Their paper separates artifact provenance from behavior and a failure rate under test from likelihood in actual use. It calls for reevaluation after changes such as quantization or fine-tuning. Its worked example is a plain inference endpoint; model-layer evidence does not establish the safety of an agent and its tools.

My takeaway: Record the exact model artifact, configuration, users, and permitted actions. Define risk-specific acceptance criteria before testing, then compare the deployed configuration with and without controls. After a model change, rerun task-quality and safety checks separately. Keep provenance gaps and untested tool permissions visible, with named owners and response plans.
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.