CISA Issues Urgent Warning on Actively Exploited MLflow Vulnerability

The Cybersecurity and Infrastructure Security Agency (CISA) has added a critical vulnerability in the MLflow open-source AI engineering platform to its Known Exploited Vulnerabilities (KEV) catalog, signaling that threat actors are actively exploiting the flaw. In an urgent advisory directed at federal civilian agencies, CISA mandated remediation of the vulnerability by August 27, 2026.

The vulnerability affects MLflow, an open-source platform used for managing the machine learning lifecycle, including experimentation, deployment, and model tracking. While the advisory confirms active exploitation in the wild, specific technical details regarding the nature of the flaw, affected software versions, and available patches were not provided in the initial reporting.

The inclusion in CISA's KEV catalog requires federal agencies to address the vulnerability within the mandated timeframe. Organizations using MLflow are advised to monitor CISA advisories and official MLflow security updates for patch information and mitigation guidance. The incident underscores the growing security risks associated with open-source AI/ML tools as their adoption expands across government and enterprise environments.