Back to Feed
VulnerabilitiesNov 21, 2026

Attackers' AI Agents Found a Zero-Day in Medical-Imaging Software and Gained Root Access

AI agents found a zero-day in MONAI Label, granting root access to a university server.

Suspected China-linked hackers utilized autonomous AI agents to discover a zero-day vulnerability in the open-source medical imaging server MONAI Label. This flaw allowed them to gain root access to a university server by writing files outside the designated storage folder. While a fix has been merged into the source code, no patched version is currently available.

AI summary of the linked source. Not reviewed by an editor. How this works

Vendor Watch

Run Amazon Threat Intelligence?

Get an email when a new story names Amazon Threat Intelligence, usually within the hour.

Free. Your list stays private and never appears in a subject line. One click stops it. How Vendor Watch worksPrivacy

The lesson from this incident

Full lesson

Suspected China-linked threat actors deployed autonomous AI agents to discover and exploit a previously unknown path-traversal vulnerability in MONAI Label, an open-source medical imaging server, ultimately achieving root access on a university system. The flaw allowed attackers to write files outside the designated storage directory, a classic directory traversal weakness that enabled privilege escalation. What makes this incident particularly alarming is the use of AI to accelerate zero-day discovery, compressing the window between vulnerability existence and active exploitation. Even though a code fix has been merged, no official patched release is yet available, leaving all deployed instances exposed with no straightforward remediation path. This underscores the critical risk of running open-source software in sensitive healthcare environments without compensating controls when patches are unavailable.

What helps against this

How we list vendors

Indicators of Compromise

Addresses and file hashes are copied from the publisher's page by pattern matching. Names are picked by AI. On stories from before 17 November 2026, some addresses and hashes were picked by AI too. Not reviewed by an editor. How this works

  • cve: CVE-2026-133968

Entities

MONAI Label (product)AI agents (technology)Amazon Threat Intelligence (vendor)China-linked hackers (threat_actor)Cairn (product)