CISA warns of hackers exploiting critical MLflow vulnerability

A critical Server-Side Request Forgery (SSRF) vulnerability in the MLflow AI/ML platform is being actively exploited by threat actors, allowing unauthenticated attackers to pivot into internal services and harvest cloud credentials. The root issue lies in insufficient input validation and a lack of authentication controls on sensitive endpoints, compounding the risk for organizations running MLflow in cloud environments. This matters because stolen cloud credentials can lead to full cloud account takeover, data exfiltration, and lateral movement across infrastructure. CISA's inclusion of this CVE in the Known Exploited Vulnerabilities catalog signals that exploitation is widespread and not merely theoretical, making immediate remediation non-negotiable.