Case mlflow-ssrf-cve-2026-64849AI supply chainWorldwide
MLflow SSRF flaw (CVE-2026-64849) actively exploited, exposing internal services
CISA says an MLflow server-side request forgery bug is actively exploited, letting attackers reach internal or cloud metadata services and read response status and body.
HighAttack on AIStatus: OngoingVerified case file
Open in the interactive mapCite this case
- Reported
- Oct 14, 2026
- Location
- Worldwide
- Target
- MLflow ML platform servers
- Severity
- High
- Triggered by
- Unknown
- Actor type
- Unknown
- Reported loss
- None reported
- CVE
- CVE-2026-64849
What happened
CISA says an MLflow server-side request forgery bug is actively exploited, letting attackers reach internal or cloud metadata services and read response status and body.
Impact: Unknown; attackers can read responses from internal and cloud metadata services.
Who triggered it
Actor: Unknown
Attribution: Not stated in the CISA KEV entry.
Motive: Unknown
How it happened
Attack vector: Server-side request forgery against the MLflow server to reach internal or cloud metadata endpoints.
Root cause: Server-side request forgery vulnerability in MLflow.
Timeline
- Reported by CISA KEV
Response
Added to CISA's Known Exploited Vulnerabilities catalog.
Current status: Ongoing
Sources (open in a new tab)
- CISA KEV cisa.gov
Cite this case
MLflow SSRF flaw (CVE-2026-64849) actively exploited, exposing internal services. AI Threat Watch, case mlflow-ssrf-cve-2026-64849, Oct 14, 2026. https://aithreat.io/cases/mlflow-ssrf-cve-2026-64849/
Permanent link: https://aithreat.io/cases/mlflow-ssrf-cve-2026-64849/