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
2,388

AI/ML CVEs Tracked

342

Critical

358

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 352 results — High severity, has patch
HIGH

OpenClaw: stale bearer token survives SecretRef rotation

GHSA-xmxx-7p24-h892
--
Auth Bypass Agent API
openclaw Patch: 2026.4.15 CWE-324 4 3 ATLAS 1 incident
HIGH

PraisonAI: SQL injection across 9 DB backends

GHSA-rg3h-x3jw-7jm5
8.1
Data Extraction Code Execution Data Leakage Framework Agent
praisonaiagents Patch: 1.6.8 CWE-89 11 4 ATLAS
HIGH

openclaw: path traversal exposes host files via media tags

GHSA-66r7-m7xm-v49h
--
Data Extraction Data Leakage Privacy Violation Agent
openclaw Patch: 2026.4.10 CWE-22 4 4 ATLAS 1 incident
HIGH

openclaw: exec approval bypass via opaque multi-call binaries

GHSA-2cq5-mf3v-mx44
--
Auth Bypass Code Execution Supply Chain Agent Plugin
openclaw Patch: 2026.4.12 CWE-863 4 5 ATLAS 1 incident
HIGH

openclaw: auth bypass lets write-scope callers mutate admin config

GHSA-7jp6-r74r-995q
--
Auth Bypass Data Leakage Agent Framework
openclaw Patch: 2026.4.10 CWE-266 4 3 ATLAS 1 incident
HIGH

openclaw: sandbox escape via host=node exec routing bypass

GHSA-736r-jwj6-4w23
--
Auth Bypass Code Execution Agent Framework
openclaw Patch: 2026.4.10 CWE-863 4 5 ATLAS 1 incident
HIGH

openclaw: untrusted plugin auto-enabled during onboarding

GHSA-939r-rj45-g2rj
--
Supply Chain Auth Bypass Agent Plugin
openclaw Patch: 2026.4.9 CWE-829 4 4 ATLAS 1 incident
HIGH

openclaw: CDP relay exposes browser DevTools on 0.0.0.0

GHSA-525j-hqq2-66r4
--
Auth Bypass Data Extraction Agent
openclaw Patch: 2026.4.10 CWE-284 4 4 ATLAS 1 incident
HIGH

openclaw: trust bypass loads untrusted workspace plugins

GHSA-82qx-6vj7-p8m2
--
Supply Chain Auth Bypass Code Execution Agent Plugin Framework
openclaw Patch: 2026.4.10 CWE-862 4 4 ATLAS 1 incident
HIGH

openclaw: env denylist bypass enables code exec in agents

GHSA-vfp4-8x56-j7c5
--
Code Execution Auth Bypass Agent Plugin
openclaw Patch: 2026.4.10 CWE-184 4 4 ATLAS 1 incident
HIGH

openclaw: WebSocket DoS via oversized frame ingestion

GHSA-vw3h-q6xq-jjm5
--
DoS Agent
openclaw Patch: 2026.4.10 CWE-400 4 4 ATLAS 1 incident
HIGH

openclaw: config redaction bypass exposes provider API keys

GHSA-8372-7vhw-cm6q
--
Data Extraction Data Leakage Auth Bypass Agent API
openclaw Patch: 2026.4.14 CWE-212 4 4 ATLAS 1 incident
HIGH

Flowise: unauth TTS endpoint exposes stored AI API keys

GHSA-5fw2-mwhh-9947
--
Auth Bypass Data Extraction Agent API
flowise Patch: 3.1.0 CWE-639 8 ATLAS
HIGH

Flowise: credential exposure via public chatflow API

GHSA-w47f-j8rh-wx87
--
Data Extraction Auth Bypass Data Leakage Agent Framework API
flowise Patch: 3.1.0 7 ATLAS
HIGH

Flowise: Mass assignment enables cross-tenant store takeover

GHSA-3prp-9gf7-4rxx
--
Auth Bypass Data Extraction Model Poisoning Agent RAG Framework
flowise Patch: 3.1.0 CWE-284 7 ATLAS
HIGH

Paperclip: RCE via workspace runtime command injection

GHSA-w8hx-hqjv-vjcq
7.3
Code Execution Data Extraction Supply Chain Agent Plugin Framework
@paperclipai/server Patch: 2026.416.0 CWE-77 4.2K 8 ATLAS
HIGH

Flowise: auth bypass enables account takeover via null token

GHSA-f6hc-c5jr-878p
--
Auth Bypass Data Extraction Agent Framework
flowise Patch: 3.1.0 CWE-287 4 ATLAS
HIGH

Flowise: Cypher injection allows full Neo4j DB wipe

GHSA-28g4-38q8-3cwc
--
Data Extraction Code Execution Data Leakage Agent Framework Plugin
flowise-components Patch: 3.1.0 CWE-943 5 ATLAS
HIGH

Flowise: HTTP reset link exposes tokens to MITM takeover

GHSA-x5w6-38gp-mrqh
--
Auth Bypass Data Extraction Agent Framework
flowise Patch: 3.1.0 CWE-319 4 ATLAS
HIGH

Flowise: OAuth token theft via unauthenticated endpoint

GHSA-6f7g-v4pp-r667
--
Auth Bypass Data Extraction Privacy Violation Agent Framework API
flowise Patch: 3.1.0 CWE-306 6 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