Ultralytics YOLO AI model compromised with cryptomining malware in supply chain attack
Versions 8.3.41 and 8.3.42 of the Ultralytics YOLO11 AI model were compromised in a supply chain attack that injected XMRig cryptomining malware into users' systems. The company's founder confirmed malicious code was injected into the PyPI deployment workflow and traced the activity to a compromised GitHub account in Hong Kong. Ultralytics paused deployments, removed the compromised versions, and released version 8.3.43 to resolve the issue.
What happened
Versions 8.3.41 and 8.3.42 of the Ultralytics YOLO11 AI model were compromised in a supply chain attack. A developer discovered the compromised code by comparing the Ultralytics PyPI package with the GitHub repository.
Malicious code was injected into the PyPI deployment workflow, installing XMRig cryptomining software on users' systems. Ultralytics founder Glenn Jocher confirmed the malicious code injection and traced the activity to a GitHub user in Hong Kong, whose account was subsequently blocked.
Ultralytics paused automatic deployments and released version 8.3.43 on Thursday to resolve the issue. Versions 8.3.41 and 8.3.42 were removed from PyPI, and version 8.3.40 and earlier were confirmed to be safe.
Reported impact
Users who installed versions 8.3.41 or 8.3.42 unknowingly ran XMRig cryptomining software on their systems. Google Colab blocked access to systems running the compromised YOLO versions.
Governance controls
Ultralytics paused automatic deployments to investigate the supply chain attack. The company has not published a public advisory on the attack as of the sources' publication date.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.