AI Agents Hacked a University Server in 11 Minutes, Amazon Warns
In a chilling demonstration of AI-powered cyberattacks, Amazon Chief Information Security Officer CJ Moses revealed that a suspected Chinese-linked hacking group used autonomous AI agents to discover and exploit a previously unknown software vulnerability. The agents took full control of a university server in just 11 minutes.
According to Amazon Threat Intelligence, the attackers used the open-source Cairn AI agent framework to target MONAI Label, an open-source medical imaging AI application. Without human direction, the agents:
- Identified the application’s framework and mapped its API endpoints.
- Found an upload endpoint vulnerable to path traversal.
- Wrote arbitrary files to the server and ultimately gained root access.
Amazon assessed with high confidence that the flaw was a genuine zero-day: no prior fix, advisory, or vulnerability database entry existed at the time. After Amazon notified the MONAI Label project maintainers, they released a critical security fix and advisory on November 18, 2026.
Security experts warn that the incident signals a new era of fully autonomous, AI-driven cyberattacks.
Sources: The Register via LAVX | WindowsForum