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
3,591

AI/ML CVEs Tracked

481

Critical

96

New This Week

24

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 1245 results — has patch
MEDIUM

OpenClaw: token rotation bypasses role approval

GHSA-whf9-3hcx-gq54
--
Auth Bypass Agent
openclaw Patch: 2026.4.8 CWE-863 3 4 ATLAS 1 incident
MEDIUM

openclaw: local file exfiltration via trusted MEDIA refs

GHSA-qqq7-4hxc-x63c
--
Data Extraction Privacy Violation Agent Plugin
openclaw Patch: 2026.4.8 CWE-668 3 4 ATLAS 1 incident
MEDIUM

OpenClaw: eval approval bypass enables unintended code exec

GHSA-q2gc-xjqw-qp89
--
Auth Bypass Code Execution Agent Framework
openclaw Patch: 2026.4.8 CWE-20 3 5 ATLAS 1 incident
HIGH

OpenClaw: env var injection enables host RCE

GHSA-7437-7hg8-frrw
--
Code Execution Supply Chain Agent Plugin
openclaw Patch: 2026.4.8 CWE-184 3 5 ATLAS 1 incident
HIGH

OpenClaw: wake hook trust violation elevates to System prompt

GHSA-jf56-mccx-5f3f
--
Prompt Injection Auth Bypass Agent Plugin
openclaw Patch: 2026.4.8 CWE-501 3 6 ATLAS 1 incident
HIGH

openclaw: trust boundary bypass enables prompt injection

GHSA-gfmx-pph7-g46x
--
Prompt Injection Auth Bypass Agent Plugin
openclaw Patch: 2026.4.8 CWE-501 3 5 ATLAS 1 incident
CRITICAL

PraisonAI: RCE via shell injection in agent workflows

GHSA-2763-cj5r-c79m
9.7
Code Execution Prompt Injection Supply Chain Agent Framework
PraisonAI Patch: 4.5.121 CWE-78 1 6 ATLAS
MEDIUM

LangChain: f-string template injection exposes object internals

GHSA-926x-3r5x-gfhw
5.3
Data Extraction Prompt Injection Framework
langchain-core Patch: 0.3.84 CWE-20 5.9K 4 ATLAS
CRITICAL

Marimo: pre-auth RCE via terminal WebSocket

GHSA-2679-6mx9-h9xc
--
Auth Bypass Code Execution Framework
marimo Patch: 0.23.0 CWE-306 3.1K 5 ATLAS
HIGH

n8n-mcp: authenticated SSRF leaks cloud metadata

GHSA-4ggg-h7ph-26qr
8.5
Data Extraction Auth Bypass Privacy Violation Agent Framework Plugin
n8n-mcp Patch: 2.47.4 CWE-918 5 ATLAS
MEDIUM

praisonaiagents: agent context leak + path traversal

GHSA-766v-q9x3-g744
6.5
Data Leakage Data Extraction Agent Framework
praisonaiagents Patch: 1.5.115 CWE-22 20 4 ATLAS
HIGH EXPLOIT AVAIL

praisonai: SSTI enables RCE via agent instructions

CVE-2026-39891
8.8
EPSS 0.6%
Code Execution Prompt Injection Data Extraction Agent Framework Plugin
praisonai Patch: 4.5.115 CWE-94 1 5 ATLAS
HIGH

PraisonAI: unauth A2U stream leaks all agent activity

CVE-2026-39889
7.5
EPSS 0.4%
Auth Bypass Data Leakage Data Extraction Agent Framework
praisonai Patch: 4.5.115 CWE-200 1 5 ATLAS
CRITICAL EXPLOIT AVAIL

praisonaiagents: sandbox escape enables host RCE

CVE-2026-39888
10.0
EPSS 0.5%
Code Execution Auth Bypass Agent Framework
praisonaiagents Patch: 1.5.115 CWE-657 20 5 ATLAS
CRITICAL EXPLOIT AVAIL

PraisonAI: YAML deserialization enables unauthenticated RCE

CVE-2026-39890
9.8
EPSS 0.6%
Code Execution Supply Chain Agent Framework
praisonai Patch: 4.5.115 CWE-502 1 5 ATLAS
MEDIUM EXPLOIT AVAIL

LobeChat: auth bypass via forged XOR obfuscated header

CVE-2026-39411
5.0
EPSS 0.1%
Auth Bypass Data Extraction DoS API Inference
@lobehub/lobehub Patch: 2.1.48 CWE-287 1.2K 5 ATLAS
HIGH

LiteLLM: auth bypass chain enables full privilege escalation

GHSA-69x8-hrgq-fjj8
--
Auth Bypass Data Extraction Privacy Violation API Inference Framework
litellm Patch: 1.83.0 CWE-200 6 5 ATLAS
HIGH

MONAI: pickle deserialization RCE in Auto3DSeg

GHSA-89gg-p5r5-q6r4
7.7
Code Execution Supply Chain Framework Training Data
monai Patch: 1.6.0 CWE-502 125 4 ATLAS
MEDIUM

OpenClaw: cleartext WebSocket exposes gateway credentials

GHSA-83f3-hh45-vfw9
--
Data Leakage Auth Bypass Agent
openclaw Patch: 2026.4.2 CWE-200 3 3 ATLAS 1 incident
MEDIUM

openclaw: timing side-channel leaks shared-secret length

GHSA-jj6q-rrrf-h66h
--
Auth Bypass Data Extraction Agent Framework
openclaw Patch: 2026.4.2 CWE-208 3 3 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