Amazon Threat Intelligence reports observing an exposed Cairn orchestration server used against East Asian universities in September 2026. Its records showed agents pursuing an operator-supplied access objective, including long unsuccessful attacks and a successful discovery of an upload-path traversal in a medical-imaging application. The failures matter: sustained automated activity did not reliably overcome hardened targets. AWS attributes the campaign to a Chinese-nexus actor with moderate confidence and assesses the vulnerability as a zero-day with high confidence; those are the investigator's judgments, not independently reproduced findings. The linked MONAI Label advisory supplies the immediate defensive action. It lists versions through 0.8.5 as affected, notes authentication is disabled by default, and says an authenticated annotator can also exploit the flaw. Although a source fix is merged, the advisory lists no patched package version and says the correction is due in the next release. Restricting exposure and filesystem privileges remains necessary while verifying which deployed build contains that fix.
Cairn intrusion report: patch MONAI Label and prepare for persistent agent attacks
Related research
More curated notes connected through Agent Security and AI Red Teaming.
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 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.
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.