AI Agents Exploit Zero-Day in Medical Imaging Software for Root Access
Original source
Attackers' AI Agents Found a Zero-Day in Medical-Imaging Software and Gained Root Access
The Hacker News →Suspected China-linked hackers used autonomous AI agents to identify a zero-day path-traversal vulnerability in MONAI Label, an open-source medical imaging server, enabling root access to a university server. The flaw allows unauthorized file writes outside the server’s storage folder, affecting versions up to 0.8.5. Amazon Threat Intelligence linked the attack to an exposed Cairn control panel, which revealed the agents’ task settings and methods. The AI agents mapped the server’s interface and exploited the vulnerability within 11 minutes, though root access required a separate session leveraging a Python startup file. The maintainers merged a fix in the source code, but no patched version has been released, leaving users to apply mitigations like restricting network access and enabling authentication. The vulnerability is now listed as CVE-2026-133968 with a critical 9.8/10 severity rating. No evidence of data exfiltration or broader impact beyond the compromised server has been confirmed.
Read the full article
Continue reading at The Hacker News →This is an AI-generated summary. Read the original for the full story.