A critical security vulnerability identified as CVE-2026-133968 has been reported in the MONAI Label software, an open-source framework used for medical imaging annotation. The flaw, categorized as a path traversal vulnerability, allows unauthorized actors to access files outside of the intended directory structure. Security researchers have highlighted the severity of the issue, noting that it was actively exploited in the wild on the same day it was officially disclosed.
The vulnerability was formally reserved and published on November 20, 2026. According to industry tracking data, the flaw was added to a commercial vulnerability catalog on that same date, marking it as an active target for exploitation. While the vulnerability has been identified as being under active attack, it currently does not appear on the official Known Exploited Vulnerabilities (KEV) lists maintained by the U.S. Cybersecurity and Infrastructure Security Agency (CISA) or the European Union Agency for Cybersecurity (ENISA).
Evidence of the exploitation has been linked to sophisticated threat actors. Reports indicate that China-linked hacking groups have utilized the Cairn AI framework to automate the exploitation of this specific medical software zero-day. By leveraging autonomous AI agents, these actors are reportedly able to identify and target vulnerable instances of MONAI Label, potentially gaining unauthorized access to sensitive medical data or system configurations.
The classification of this event as an active exploitation rests primarily on findings from commercial security research organizations. While these reports provide a timeline for the vulnerability's emergence and subsequent misuse, the claims of widespread exploitation remain unverified by broader government-led security bodies. The discrepancy between commercial research findings and official federal KEV lists suggests that while the threat is real, the scope of the exploitation may still be in the early stages of documentation.
The technical nature of the path traversal flaw in MONAI Label presents a significant risk to healthcare institutions and research facilities that rely on the platform for AI-assisted image analysis. By manipulating input paths, an attacker can bypass security restrictions to read or potentially write files on the host system. This capability is particularly concerning given the sensitive nature of the medical imaging data processed by the software.
Organizations utilizing MONAI Label are encouraged to review their security configurations and monitor for unauthorized access attempts. As this vulnerability was exploited immediately upon disclosure, the window for remediation is extremely narrow. Users should prioritize applying any available patches or security updates provided by the project maintainers to mitigate the risk of remote file access.
The involvement of AI-driven exploitation tools marks a notable shift in the tactics used by nation-state actors. By automating the discovery and exploitation of zero-day vulnerabilities in specialized software like MONAI Label, these groups can scale their operations and target high-value infrastructure with increased efficiency. This incident serves as a reminder of the evolving threat landscape facing specialized scientific and medical software.
Further investigation into the extent of the compromise is ongoing. Security teams are advised to consult official project documentation and security advisories for specific guidance on hardening their MONAI Label deployments against path traversal attacks. As of the current reporting, no secondary independent catalogs have corroborated the full extent of the exploitation, though the initial findings remain a primary concern for the cybersecurity community.






