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
2,385

AI/ML CVEs Tracked

340

Critical

390

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 672 results — High severity, no patch
HIGH EXPLOIT AVAIL

Claude Code CLI: OS command injection via TERMINAL env

CVE-2026-35020
8.4
EPSS 0.1%
Code Execution Supply Chain Agent API
claude-code CWE-78 5 ATLAS
HIGH EXPLOIT AVAIL

KubeAI: RCE via shell injection in Ollama startup probe

CVE-2026-34940
8.8
EPSS 0.4%
Code Execution Inference Framework
CWE-78 3 ATLAS
HIGH

oauthenticator: auth bypass enables JupyterHub account takeover

CVE-2026-33175
8.8
EPSS 0.4%
Auth Bypass Framework
CWE-287 4 ATLAS
HIGH

Ajenti: missing authz lets any user install packages

CVE-2026-35175
--
EPSS 0.3%
Auth Bypass Supply Chain Code Execution Framework Plugin
CWE-862 4 ATLAS
HIGH

vLLM: audio downmix mismatch enables adversarial input

CVE-2026-34760
7.1
EPSS 0.3%
Adversarial Examples Supply Chain Inference Framework
CWE-20 4 ATLAS
HIGH

Claude Setup: DLL search-order hijacking LPE

CVE-2026-22561
7.8
EPSS 0.2%
Code Execution Supply Chain API Framework
3 ATLAS
HIGH

OpenClaw before 2026.3.28 contains a server-side request forgery vulnerability in the fal provider image-generation-provider.ts component that allows attackers...

CVE-2026-34504
8.3
EPSS 0.2%
OpenClaw 4
HIGH

OpenClaw before 2026.3.28 fails to disconnect active WebSocket sessions when devices are removed or tokens are revoked. Attackers with revoked credentials can...

CVE-2026-34503
8.1
EPSS 0.3%
OpenClaw 4
HIGH

OpenClaw before 2026.3.28 contains an insufficient scope validation vulnerability in the node pairing approval path that allows low-privilege operators to...

CVE-2026-33577
8.1
EPSS 0.4%
OpenClaw 4
HIGH

OpenClaw before 2026.3.11 contains a sandbox boundary bypass vulnerability in fs-bridge staged writes where temporary file creation and population are not...

CVE-2026-32988
7.5
EPSS 0.1%
OpenClaw 4
HIGH

OpenClaw before 2026.3.13 contains an information disclosure vulnerability in the fetchRemoteMedia function that exposes Telegram bot tokens in error messages....

CVE-2026-32982
7.5
EPSS 0.4%
OpenClaw 4
HIGH

OpenClaw before 2026.3.11 contains an approval-integrity vulnerability in node-host system.run approvals that displays extracted shell payloads instead of the...

CVE-2026-32971
7.1
EPSS 0.3%
OpenClaw 4
HIGH

OpenClaw before 2026.3.12 automatically discovers and loads plugins from .OpenClaw/extensions/ without explicit trust verification, allowing arbitrary code...

CVE-2026-32920
8.4
EPSS 0.3%
OpenClaw 4
HIGH

1millionbot Millie: Boolean prompt injection bypasses restrictions

CVE-2026-4399
7.5
EPSS 0.3%
Prompt Injection Jailbreak API Agent
6 ATLAS 3 incidents
HIGH

gdk-pixbuf: JPEG heap overflow crashes vLLM inference

CVE-2026-5201
7.5
EPSS 1.1%
DoS Inference
rhaiis/vllm-cuda-rhel9 130 2 ATLAS
HIGH EXPLOIT AVAIL

awesome-llm-apps MCP Agent: cross-session credential theft

CVE-2026-29872
8.2
EPSS 0.3%
Data Leakage Auth Bypass Privacy Violation Agent API Plugin
7 ATLAS
HIGH

CrewAI: arbitrary file read via JSON loader tool

CVE-2026-2285
7.5
EPSS 0.6%
Data Extraction Data Leakage Prompt Injection Framework Agent Plugin
6 ATLAS
HIGH

libarchive: integer overflow in zisofs hits vllm containers

CVE-2026-5121
7.5
EPSS 1.1%
Code Execution Data Extraction Inference
rhaiis/vllm-cuda-rhel9 130 3 ATLAS
HIGH

OpenClaw before 2026.3.12 embeds long-lived shared gateway credentials directly in pairing setup codes generated by /pair endpoint and OpenClaw qr command....

CVE-2026-33575
7.5
EPSS 0.2%
OpenClaw 4
HIGH

OpenClaw before 2026.3.11 contains an authorization bypass vulnerability in the gateway agent RPC that allows authenticated operators with operator.write...

CVE-2026-33573
8.8
EPSS 0.3%
OpenClaw 4

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