Skip to main content
Mallory
← Back to News
actively-exploited-vulnerabilityai-enabled-threat-activityeducation-sector-threatinternet-facing-service-vulnerability

AI Agents Found and Exploited MONAI Label Zero-Day in University Attacks, Amazon Says

Active Exploitation · 9 sources · Updated 5h ago · First seen Nov 20, 2026

AI Agents Found and Exploited MONAI Label Zero-Day in University Attacks, Amazon Says

Suspected China-linked attackers used Cairn, an open-source autonomous AI penetration-testing framework, to discover and exploit a zero-day in MONAI Label, gaining root access to a university server, Amazon Threat Intelligence reported. The September 8–10 campaign targeted universities in East Asia outside mainland China. Given target IP addresses and objectives such as obtaining shell access, the agents independently mapped APIs and discovered a path traversal flaw allowing arbitrary file writes in approximately 11 minutes. In a subsequent session, they exploited the flaw to deploy a Python reverse shell with root privileges. Amazon assessed the zero-day finding with high confidence and the attackers’ Chinese nexus with moderate confidence.

Amazon uncovered the operation through an unauthenticated, publicly accessible Cairn instance. The observed activity also included prolonged unsuccessful exploitation and credential-guessing attempts; a later pivot into another university network required direct operator assistance, distinguishing autonomous exploitation from human-directed follow-on activity. Amazon notified the application’s maintainers, who released a security update and advisory on November 18, and coordinated victim notification through a government partner. Organizations running MONAI Label should prioritize the update and investigate exposed deployments for unauthorized file writes and reverse-shell activity.

This is being exploited nowSee which actors are running it and whether you're in range.
Check your exposure
Event timeline

How this story unfolded

Events from the most recent confirmed update back to the earliest known activity.

11 events · Sep 2026 – Nov 18, 2026
  1. Reported Nov 20, 202614h ago

    Amazon discloses campaign details and assesses a Chinese nexus

    Amazon disclosed the autonomous exploitation campaign and assessed a Chinese nexus with moderate confidence based on Simplified Chinese prompts, China Standard Time timestamps, and target selection. Its report also identified 142[.]4.98.145 and 142[.]4.107.34 in campaign records without assigning them specific operational roles.

    Amazon Says AI Agents Found and Exploited a Zero-Day | Frontier
  2. Reported Nov 20, 202614h ago

    AWS coordinates victim notification through a government partner

    AWS worked with a government partner to notify affected organizations. The victims were not AWS customers, leaving AWS without a direct notification path.

    Chinese-nexus actor’s Cairn AI agents found a zero-day on their own | AWS Security Blog
  3. Reported Nov 20, 202614h ago

    Amazon reports the vulnerability to MONAI Label maintainers

    Amazon notified the affected medical imaging project's maintainers about the vulnerability. The notification preceded the maintainers' release of a fix and security advisory.

    Hackers use autonomous AI tool Cairn to find zero-day, take over server
  4. Nov 18, 20263d ago

    MONAI Label maintainers release a security update

    MONAI Label maintainers released a security update addressing the exploited path traversal vulnerability on November 18, 2026. They also issued a GitHub Security Advisory following Amazon's notification.

    Hackers use autonomous AI tool Cairn to find zero-day, take over server
  5. Sep 10, 20262mo ago

    Amazon discovers attackers' publicly exposed Cairn instance

    Amazon Threat Intelligence discovered an unauthenticated, publicly accessible Cairn instance exposing project configurations, operator prompts, AI output, and operational history. The exposure enabled investigators to reconstruct the campaign.

    Hackers use autonomous AI tool Cairn to find zero-day, take over server
  6. Sep 8, 20262mo ago

    Cairn campaign begins targeting East Asian universities

    Attackers used Cairn's autonomous AI agents to target universities in East Asia outside mainland China during September 8–10, 2026. Operators supplied target IP addresses and objectives such as obtaining shell access, leaving agents to plan and adapt their attacks.

    Hackers use autonomous AI tool Cairn to find zero-day, take over server
  7. Sep 2026

    Operator provides agents a tunnel into a second university network

    After the zero-day discovery, the operator supplied a proxy into a second university network and directed agents to target additional machines. Amazon inferred that the operator had exploited the same vulnerability on the proxy-connected server to establish the tunnel.

    Hackers use autonomous AI tool Cairn to find zero-day, take over server
  8. Sep 2026

    Agents exploit MONAI Label flaw to obtain a root reverse shell

    In a later session, agents used the vulnerability to write a Python reverse-shell payload into sitecustomize.py. Triggering a training API endpoint launched a Python process that imported the payload and established a reverse shell running as root.

    Amazon Says AI Agents Found and Exploited a Zero-Day | Frontier
  9. Sep 2026

    Cairn agents discover and exploit a MONAI Label zero-day

    Agents independently discovered a path traversal flaw in an upload endpoint's filename parameter in MONAI Label, an open-source medical imaging application. They progressed from API mapping to arbitrary file writes in approximately 11 minutes; Amazon assessed the flaw as a zero-day with high confidence.

    Amazon Says AI Agents Found and Exploited a Zero-Day | Frontier
  10. Sep 2026

    Agents test over 30,000 credential combinations without gaining access

    A separate operation targeting a university GPU server lasted more than six hours. Agents assembled approximately 80 real usernames from public sources and tested more than 30,000 credential combinations across university services without finding a working login.

    Hackers use autonomous AI tool Cairn to find zero-day, take over server
  11. Sep 2026

    Autonomous Redis intrusion attempt fails after 148 actions

    During the September campaign, agents spent approximately 16 hours attempting Redis exploitation, Log4Shell injection, and credential spraying without achieving a successful compromise. They also exploited an SSRF vulnerability in a co-located Qdrant instance to map internal Docker networks.

    Hackers use autonomous AI tool Cairn to find zero-day, take over server
Linked entities

Related entities

Vulnerabilities, threat actors, malware, products, organizations, breaches, and observables Mallory has linked to this story. Indicator values are masked here and available in full in the app.

Open in app →

Malware

1 linked

Observables

3 linked

Indicator values are masked on this page. See the values in Mallory Domains, IPs, hashes, and URLs are exportable to your SIEM.

Do you run this?Correlate live exploitation activity against the software you actually run, and see where you're exposed.
See Who's Exploiting This
Mallory platform

See the full picture, correlated to your attack surface.

This page covers what’s public. Mallory adds the parts that aren’t: which of your assets are affected, which threat actors are using it right now, and what to deploy next.

Exposure mappingMap indicators from this story to your assets and identify affected systems in minutes.
Threat actor evidenceEvery observed campaign, victim, and pivot linked to actors named in this story.
Associated malwareMalware, exploits, and IOCs connected to the activity described here.
Detection signaturesYARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Scheduled alertsGet matching new stories delivered to your team as they break, not the next morning.
AI threadsAsk questions about this story and take action on the answers.