Incident library · AI software · United States

    Hugging Face model users: About 100 malicious models found

    Researchers found about 100 AI models on a popular public model hub that ran hidden code when loaded, some opening a remote connection to the user's machine.
    BleepingComputerFeb 2024

    Malicious AI models on Hugging Face backdoor users' machines

    About 100 malicious models found

    Identified in an independent review

    Downloaded AI models were loaded and run like trusted files, without checking where they came from or what code they contained, researchers found.

    Identified in an independent review

    The reported contributing factor

    Downloaded AI models were loaded and run like trusted files, without checking where they came from or what code they contained, researchers found.

    Incidents usually have several contributing factors. This is the one public reporting names; it is not presented as the sole cause.

    Which StackUp assessment checks for this gap

    Find out if your organisation has this gap

    Sources

    Published Feb 2024. Sources last checked Nov 2026. Spotted an error or an update? Tell us and we will correct it.

    • The RegisterJul 2025

      Destructive AI prompt published in Amazon Q extension

      Compromised AI tool shipped to users

      Stated by the organisation

      An outside contributor's change reached a public release through an over-permissioned build token, and the AI agent could run commands without human confirmation, AWS said.

    • Infosecurity MagazineJun 2025

      M365 Copilot: New Zero-Click AI Flaw Allows Corporate Data Theft

      Critical flaw rated 9.3 out of 10

      Identified in an independent review

      Untrusted content in an email could steer the AI assistant's access to company data, researchers found. Microsoft said no customer data was taken.

    • Silicon UKJan 2024

      DPD Disables AI Chatbot After It Swears At Customer

      AI chatbot switched off

      Stated by the organisation

      After a system update, the chatbot could be talked into ignoring its rules, and the change was not caught before customers saw it, the company said.

    Find out which of these gaps you have

    The free Readiness Check takes ten questions and shows the incidents tied to your weakest answers.