Critical SSRF Vulnerability in MLflow Actively Exploited, CISA Issues Warning

Critical SSRF Vulnerability in MLflow Actively Exploited, CISA Issues Warning

First seen 21 Aug 2026, 16:48 UTC •Ciberseguridadlatam •80% similarity •69.9

Article Content

Browse articles
ThreatCluster

A critical SSRF vulnerability (CVE-2026-64849) in MLflow, the world's most downloaded ML platform, allows unauthorized access to metadata services on AWS, Azure, and GCP. The U.S. Cybersecurity and Infrastructure Security Agency (CISA) added this vulnerability to its catalog of actively exploited threats on August 19, 2026. The flaw, which lacks authentication, poses a significant risk to cloud environments. As of August 21, 2026, the vulnerability is under active exploitation, with confirmed incidents of credential theft reported. Organizations using MLflow are advised to take immediate action to mitigate this risk.

Key Points: • CVE-2026-64849 is a critical SSRF vulnerability in MLflow affecting major cloud platforms. • CISA added this vulnerability to its active threats list on August 19, 2026. • Credential theft incidents have been confirmed due to this vulnerability.

ThreatCluster AI How this analysis works

Timeline

2026-08-17
CVE-2026-64849 published
The SSRF vulnerability in MLflow was officially published, detailing its impact on cloud services.
Ciberseguridadlatam
2026-08-19
CISA adds CVE-2026-64849 to active threats
CISA confirmed the active exploitation of the SSRF vulnerability in MLflow, prompting urgent warnings.
Ciberseguridadlatam
2026-08-21
Credential theft confirmed
Reports indicate that the vulnerability has led to credential theft incidents in cloud environments.
Ciberseguridadlatam

Extracted Entities

1 / 2

Community

Browse all →

Tracked Entities in This Story

AWS Azure MLflow GCP