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
2,404

AI/ML CVEs Tracked

347

Critical

310

New This Week

18

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 1577 results — no patch
MEDIUM

OpenClaw: safeBins bypass allows file read/write

CVE-2026-31996
4.4
EPSS 0.1%
Code Execution Data Extraction Auth Bypass Agent Plugin
OpenClaw 4 6 ATLAS 1 incident
MEDIUM

OpenClaw: cmd injection via Windows shell fallback

CVE-2026-31995
5.3
EPSS 0.5%
Code Execution Supply Chain Agent Plugin
OpenClaw 4 4 ATLAS 1 incident
HIGH

OpenClaw: cmd injection via scheduled task generation

CVE-2026-31994
7.1
EPSS 0.6%
Code Execution Auth Bypass Agent
OpenClaw 4 3 ATLAS 1 incident
MEDIUM

OpenClaw: exec approval bypass allows RCE on macOS

CVE-2026-31993
4.8
EPSS 0.3%
Auth Bypass Code Execution Agent Plugin
OpenClaw 4 4 ATLAS 1 incident
HIGH

OpenClaw: guardrail allowlist bypass enables arbitrary command execution

CVE-2026-31992
7.1
EPSS 0.3%
Auth Bypass Code Execution Agent Framework
OpenClaw 4 5 ATLAS 1 incident
LOW

OpenClaw: auth bypass via DM pairing grants group access

CVE-2026-31991
3.7
EPSS 0.2%
Auth Bypass Agent
OpenClaw 4 3 ATLAS 1 incident
MEDIUM

OpenClaw: symlink traversal enables arbitrary file write

CVE-2026-31990
6.1
EPSS 0.1%
Code Execution Supply Chain Agent
OpenClaw 4 3 ATLAS 1 incident
HIGH

OpenClaw: SSRF in citation redirect exposes internal network

CVE-2026-31989
7.4
EPSS 0.2%
Data Extraction Auth Bypass Agent Plugin
OpenClaw 4 5 ATLAS 1 incident
MEDIUM

OpenClaw: argv rewrite bypasses approval, enables RCE

CVE-2026-29608
6.7
EPSS 0.1%
Code Execution Auth Bypass Agent Plugin
OpenClaw 4 5 ATLAS 1 incident
MEDIUM

OpenClaw: auth bypass enables unapproved RCE via wrapper

CVE-2026-29607
6.8
EPSS 0.4%
Auth Bypass Code Execution Agent Plugin
OpenClaw 4 5 ATLAS 1 incident
HIGH

OpenClaw: webhook DoS via memory exhaustion

CVE-2026-28461
7.5
EPSS 0.4%
DoS Agent
OpenClaw 4 3 ATLAS
HIGH

OpenClaw: allowlist bypass enables OS command execution

CVE-2026-28460
7.1
EPSS 0.4%
Auth Bypass Code Execution Agent
OpenClaw 4 4 ATLAS 1 incident
MEDIUM

OpenClaw: webhook replay triggers duplicate agent actions

CVE-2026-28449
6.5
EPSS 0.3%
Auth Bypass DoS Agent Plugin
OpenClaw 4 5 ATLAS
MEDIUM

OpenClaw: TOCTOU race enables arbitrary file write in agents

CVE-2026-27670
5.3
EPSS 0.1%
Supply Chain Code Execution Agent Framework
OpenClaw 4 3 ATLAS 1 incident
HIGH

OpenClaw: allowlist bypass enables arbitrary OS command exec

CVE-2026-27566
7.1
EPSS 0.4%
Auth Bypass Code Execution Agent Plugin
OpenClaw 4 5 ATLAS 1 incident
MEDIUM

OpenClaw: cmd injection in Windows task script generation

CVE-2026-22176
6.1
EPSS 0.6%
Code Execution Supply Chain Agent Framework
OpenClaw 4 4 ATLAS 1 incident
HIGH EXPLOIT AVAIL

deepdiff: DoS causes service disruption

CVE-2026-33155
--
EPSS 0.5%
DoS Supply Chain Framework
CWE-400 4 ATLAS
CRITICAL EXPLOIT AVAIL

onnx: Integrity Verification bypass enables tampering

CVE-2026-28500
9.1
EPSS 0.3%
Supply Chain Model Poisoning Code Execution Framework RAG API
onnx CWE-345 1.2K 10 ATLAS
MEDIUM

OpenClaw versions prior to 2026.2.26 contain an approval bypass vulnerability in system.run execution that allows attackers to execute commands from unintended...

CVE-2026-27545
6.1
EPSS 0.1%
OpenClaw 4
MEDIUM

OpenClaw: prototype pollution bypasses command gate

CVE-2026-27524
4.3
EPSS 0.2%
Auth Bypass Code Execution Agent Framework
OpenClaw 4 4 ATLAS 1 incident

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