AI Security Threat Feed

Latest CVEs affecting AI/ML systems, updated continuously. Tracked from NVD, GitHub Advisory, and CISA KEV.

1,604

AI/ML CVEs Tracked

225

Critical

New This Week

In CISA KEV

Sort: Date CVSS EPSS KEV first Most exploited

Filter: All Critical High Medium KEV only Active exploitation Has patch No patch

Latest AI Security Threats

Showing 20 of 1092 results — no patch

HIGH EXPLOIT AVAIL

Transformers: unsafe deserialization enables RCE on load

Supply Chain Code Execution Framework Model

transformers 7.9K 5 ATLAS

HIGH EXPLOIT AVAIL

HuggingFace Transformers: RCE via unsafe deserialization

Supply Chain Code Execution Framework Model

transformers CWE-502 7.9K 5 ATLAS

HIGH EXPLOIT AVAIL SCANNER

MLflow: path traversal exposes arbitrary files (no auth)

Data Extraction Framework

mlflow 624 4 ATLAS

HIGH EXPLOIT AVAIL SCANNER

MLflow: path traversal allows arbitrary file write

Supply Chain Code Execution Framework Training Data

mlflow CWE-22 624 4 ATLAS

HIGH EXPLOIT AVAIL

Gradio: command injection enables RCE on ML servers

Code Execution Data Extraction Supply Chain Framework Inference API

gradio 679 5 ATLAS

HIGH EXPLOIT AVAIL

MLflow: path traversal exposes arbitrary file read/write

Data Extraction Code Execution Supply Chain Framework Training Data Model

mlflow 624 5 ATLAS

HIGH EXPLOIT AVAIL

MLflow: SSTI enables RCE in ML experiment tracking

Code Execution Supply Chain Data Extraction Framework Training Data

mlflow 624 5 ATLAS

MEDIUM EXPLOIT AVAIL SCANNER

MLflow: reflected XSS via Content-Type header injection

Auth Bypass Data Extraction Framework

mlflow CWE-79 624 4 ATLAS

HIGH EXPLOIT AVAIL SCANNER

MLflow: unauth REST API leaks sensitive ML data

Data Extraction Data Leakage Auth Bypass Framework API Training Data

mlflow 624 5 ATLAS

CRITICAL KEV SCANNER

Ray: unauthenticated RCE via job submission API

Code Execution Auth Bypass Framework Training Data Inference

ray CWE-829 847 6 ATLAS

MEDIUM

TorchServe: ZipSlip arbitrary file write via model upload

Supply Chain Code Execution Framework Inference

torchserve 21.9K 4 ATLAS

CRITICAL EXPLOIT AVAIL

MLflow: auth bypass allows arbitrary account creation

Auth Bypass Data Extraction Supply Chain Framework Training Data Model

mlflow 624 6 ATLAS

CRITICAL EXPLOIT AVAIL SCANNER

MLflow: unauth file overwrite enables model poisoning

Code Execution Model Poisoning Supply Chain Framework Model Training Data

mlflow 624 5 ATLAS

HIGH EXPLOIT AVAIL

MLflow: unauthenticated arbitrary file write via PUT

Supply Chain Code Execution Model Poisoning Framework Model Training Data

mlflow 624 6 ATLAS

CRITICAL EXPLOIT AVAIL

MLeap: zip slip in model loading enables RCE

Supply Chain Code Execution Framework Model

3 ATLAS

HIGH

Infinite Image Browsing: path traversal leaks credentials

Data Extraction Auth Bypass Privacy Violation Framework Inference

5 ATLAS

HIGH EXPLOIT AVAIL

LangChain: prompt injection triggers SSRF via URL fetch

Prompt Injection Data Extraction Framework Agent

langchain CWE-74 2.6K 5 ATLAS

HIGH KEV

LangChain: SSRF in URL loader exposes internal network

Data Extraction Auth Bypass Framework RAG

langchain CWE-918 2.6K 4 ATLAS

CRITICAL KEV

LangChain: RCE bypass via __import__ in PAL chain

Code Execution Auth Bypass Framework Agent

langchain_experimental 2.6K 4 ATLAS

CRITICAL KEV SCANNER

TorchServe: SSRF + RCE via unrestricted model URL loading

Code Execution Supply Chain Data Extraction Framework Inference Model

torchserve 21.9K 5 ATLAS

Need deeper analysis?

Get ATLAS technique mappings, compliance reports (ISO 42001, EU AI Act), breaking alerts, and full CISO analysis with a Pro subscription.

Start 14-Day Free Trial

AI Security Threat Feed

Weekly CISO Take + top threats

Latest AI Security Threats

Need deeper analysis?