Unit 42 AI Security · November 12, 2026

Cairn intrusion report: harden identities and exposed applications

Why it matters

Unit 42 reports attackers using the Cairn agent framework against semiconductor, defense and civil-society organizations. Observed access relied on familiar weaknesses, including authentication bypasses, exposed APIs and weak credentials; attackers sometimes supplied prior target intelligence. The report’s China-nexus attribution is moderate-confidence analysis. Its operational account does not establish a general autonomous attack-success rate.

My takeaway: Review public registration and administrative endpoints, remove default credentials, and test object-level authorization and cloud-storage write permissions. Alert on unexpected administrator creation and persistent API tokens. Correlate identity and application logs across repeated probes; treat the report’s infrastructure indicators as dated leads, not exclusive proof of attribution.
Keep exploring

More curated notes connected through Agent Security 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.

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.