MLflow Vulnerability Exploited for Cloud Credential Theft

Threat actors are actively leveraging an unauthenticated server-side request forgery (SSRF) flaw in MLflow, tracked as CVE-2026-64849, to exfiltrate cloud credentials and secrets. The vulnerability stems from the MLflow Tracking Server exposing model-registry webhook APIs without proper authentication, allowing attackers to bypass SSRF protections introduced in version 3.10.0 and directly access cloud metadata services. With a CVSS score of 9.3, this defect affects all MLflow versions prior to 3.15.0 and has been added to the CISA Known Exploited Vulnerabilities catalog, prompting urgent remediation across affected systems.