AI Threat Alert

AI Security Threat Feed

Latest CVEs affecting AI/ML systems — LLM frameworks, ML libraries, AI agents, vector databases, and inference servers. Vulnerabilities are tracked from NVD, GitHub Advisory, CISA KEV, MITRE ATLAS, and enriched with CVSS, EPSS, exploitation confidence, AI-component classification, and compliance mappings to ISO 42001, EU AI Act, NIST AI RMF, and OWASP LLM Top 10. Updated continuously as new CVEs are published.

Sources
Each CVE is enriched with
  • CVSS severity
  • EPSS exploit probability
  • Exploitation confidence
  • AI-component classification
  • Compliance mappings
3,019

AI/ML CVEs Tracked

424

Critical

113

New This Week

19

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 400 results — Medium severity, has patch
MEDIUM

OpenClaw: untrusted plugin RCE via workspace channel setup

GHSA-2qrv-rc5x-2g2h
--
Supply Chain Code Execution Auth Bypass Agent Plugin
openclaw Patch: 2026.4.2 CWE-829 4 4 ATLAS 1 incident
MEDIUM

OpenClaw: read-only scope bypass kills agent sessions

GHSA-5hff-46vh-rxmw
--
Auth Bypass DoS Agent
openclaw Patch: 2026.4.2 CWE-269 4 3 ATLAS
MEDIUM

openclaw: iOS bridge bypass enables unauthorized agent runs

GHSA-4p4f-fc8q-84m3
--
Auth Bypass DoS Agent
openclaw Patch: 2026.4.2 CWE-284 4 4 ATLAS
MEDIUM

OpenClaw: path traversal → host file exfiltration via QQ Bot

GHSA-846p-hgpv-vphc
--
Prompt Injection Data Extraction Agent Plugin
openclaw Patch: 2026.4.2 CWE-22 4 4 ATLAS 1 incident
MEDIUM

openclaw: path traversal enables remote dir overwrite

GHSA-m34q-h93w-vg5x
--
Supply Chain Code Execution Agent Framework
openclaw Patch: 2026.4.2 CWE-22 4 4 ATLAS 1 incident
MEDIUM

OpenClaw: pairing DoS blocks account onboarding

GHSA-wwfp-w96m-c6x8
--
DoS Agent
openclaw Patch: 2026.3.31 4 2 ATLAS 1 incident
MEDIUM

OpenClaw: pre-auth signature bypass enables pairing DoS

GHSA-h43v-27wg-5mf9
--
DoS Auth Bypass Agent Framework
openclaw Patch: 2026.3.31 CWE-347 4 3 ATLAS
MEDIUM

OpenClaw: exec allowlist bypass via shell init-file options

GHSA-wpc6-37g7-8q4w
--
Auth Bypass Code Execution Agent Plugin
openclaw Patch: 2026.3.31 CWE-184 4 4 ATLAS 1 incident
MEDIUM

openclaw: sandbox escape via mirror mode hook execution

GHSA-42mx-vp8m-j7qh
--
Code Execution Supply Chain Agent Framework
openclaw Patch: 2026.3.28 CWE-829 4 4 ATLAS 1 incident
MEDIUM

openclaw: auth bypass exposes agent session visibility

GHSA-fwjq-xwfj-gv75
--
Auth Bypass Data Extraction Agent Framework
openclaw Patch: 2026.3.31 CWE-863 4 4 ATLAS 1 incident
MEDIUM

openclaw: privilege escalation to admin voice config persistence

GHSA-3q42-xmxv-9vfr
--
Auth Bypass Supply Chain Agent Framework
openclaw Patch: 2026.3.28 CWE-269 4 3 ATLAS 1 incident
MEDIUM

openclaw: SSRF in marketplace plugin download

GHSA-vjx8-8p7h-82gr
--
Supply Chain Data Extraction Auth Bypass Agent Plugin
openclaw Patch: 2026.3.31 CWE-918 4 4 ATLAS 1 incident
MEDIUM

openclaw: media download bypass exhausts disk storage

GHSA-4g5x-2jfc-xm98
--
DoS Agent Plugin
openclaw Patch: 2026.3.31 CWE-434 4 3 ATLAS
MEDIUM

openclaw: operator scope bypass in phone arm/disarm cmds

GHSA-h2v7-xc88-xx8c
--
Auth Bypass Agent Plugin
openclaw Patch: 2026.3.28 CWE-285 4 3 ATLAS 1 incident
MEDIUM EXPLOIT AVAIL

MLflow: auth bypass exposes model artifacts across experiments

CVE-2026-33866
4.3
EPSS 0.4%
Auth Bypass Data Extraction Framework Model
mlflow Patch: 3.11.0rc0 CWE-862 680 4 ATLAS
MEDIUM EXPLOIT AVAIL

MLflow: stored XSS via MLmodel YAML artifact upload

CVE-2026-33865
--
EPSS 0.2%
Code Execution Auth Bypass Data Extraction Framework Model
mlflow Patch: 3.11.1 CWE-79 680 4 ATLAS
MEDIUM EXPLOIT AVAIL

HuggingFace Transformers: RCE via malicious checkpoint load

CVE-2026-1839
6.5
EPSS 0.3%
Code Execution Supply Chain Framework Training Data
transformers Patch: 5.0.0rc3 CWE-502 9.2K 3 ATLAS
MEDIUM

OpenClaw: script preflight bypass enables unsafe exec

CVE-2026-34425
--
EPSS 0.3%
Auth Bypass Code Execution Agent Plugin
openclaw Patch: 2026.4.2 CWE-184 4 4 ATLAS 1 incident
MEDIUM

kedro-datasets: path traversal enables arbitrary file write

CVE-2026-35492
6.5
EPSS 0.4%
Supply Chain Model Poisoning Code Execution Framework Training Data
kedro-datasets Patch: 9.3.0 CWE-22 2.9K 3 ATLAS
MEDIUM

vLLM: OOM DoS via unbounded video frame decoding

CVE-2026-34755
6.5
EPSS 0.4%
DoS Framework Inference API
vllm Patch: 0.19.0 CWE-770 130 3 ATLAS

Frequently asked questions

What is an AI security threat feed?

An AI security threat feed is a continuously updated stream of vulnerabilities (CVEs) affecting AI and machine-learning systems — LLM frameworks, ML libraries, AI agents, vector databases, and inference servers — filtered out of the broader CVE firehose and enriched for relevance.

Which sources are the AI CVEs tracked from?

CVEs are tracked from NVD, GitHub Advisory, CISA KEV, and MITRE ATLAS, then enriched with CVSS, EPSS, exploitation confidence, AI-component classification, and compliance mappings.

What AI systems do these vulnerabilities affect?

Coverage spans LLM frameworks, ML libraries, AI agents, vector databases, and inference servers — the components most security teams now run in production.

How often is the AI threat feed updated?

The feed updates continuously as new CVEs are published and enriched, so the most recent AI/ML vulnerabilities appear at the top.

Is the AI security feed free?

Yes — the public feed is free to browse. A Pro subscription adds breaking alerts, MITRE ATLAS mappings, compliance reports (ISO 42001, EU AI Act), and full CISO analysis.

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