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
4,307

AI/ML CVEs Tracked

533

Critical

95

New This Week

25

In CISA KEV

Sort:DateCVSSEPSSKEV firstMost exploited
Filter:AllCriticalHighMediumKEV onlyActive exploitationHas patchNo patch

Latest AI Security Threats

Showing 20 of 1527 results — Medium severity
MEDIUM

OpenClaw: empty allowlist bypasses BlueBubbles DM auth

CVE-2026-22170
6.5
EPSS 0.3%
Auth BypassPrompt InjectionAgentPlugin
OpenClaw33 ATLAS
MEDIUM

OpenClaw: safeBins allowlist bypass enables command exec

CVE-2026-22169
6.7
EPSS 0.2%
Code ExecutionAuth BypassAgentPlugin
OpenClaw33 ATLAS
MEDIUM

OpenClaw: cmd.exe argument smuggling evades approval log

CVE-2026-22168
6.5
EPSS 0.4%
Code ExecutionAuth BypassAgentPlugin
OpenClaw34 ATLAS1 incident
MEDIUM

fickling: Allowlist Bypass evades input filtering

GHSA-5cxw-w2xg-2m8h
--
Supply ChainData ExtractionCode ExecutionFrameworkModelTraining Data
ficklingPatch: 0.1.10CWE-184625 ATLAS
MEDIUM

fickling: Allowlist Bypass evades input filtering

GHSA-r48f-3986-4f9c
--
Data ExtractionCode ExecutionAuth BypassFramework
ficklingPatch: 0.1.10CWE-184625 ATLAS
MEDIUM

systemd-machined: access control flaw allows local root

CVE-2026-4105
6.7
Code ExecutionFrameworkInference
rhoai/odh-mlflow-operator-rhel96952 ATLAS
MEDIUM

OpenClaw: $include path traversal exposes secrets

CVE-2026-32061
4.4
EPSS 0.2%
Data LeakagePrivacy ViolationAgentFramework
openclaw34 ATLAS1 incident
MEDIUM

Greenshift: Info Disclosure leaks sensitive data

CVE-2026-2589
5.3
EPSS 0.3%
Data ExtractionData LeakageSupply ChainAPIRAGPlugin
CWE-2009 ATLAS
MEDIUM

OpenClaw voice-call: webhook auth bypass via headers

CVE-2026-28465
5.9
EPSS 0.4%
Auth BypassSocial EngineeringPluginAgentAPI
voice-call33 ATLAS
MEDIUM

langgraph: Deserialization enables RCE

CVE-2026-28277
6.8
EPSS 5.5%
Code ExecutionData LeakageSupply ChainFrameworkAgent
langgraphPatch: 1.0.10CWE-5024.1K5 ATLAS
MEDIUM

gradio: Info Disclosure leaks sensitive data

CVE-2026-28415
4.7
EPSS 0.3%
Data ExtractionCode ExecutionAuth BypassFrameworkRAGAPI
gradioCWE-2007357 ATLAS
MEDIUMEXPLOIT AVAIL

gradio: Weak Credentials allow account compromise

CVE-2026-27167
5.9
EPSS 0.5%
Supply ChainModel PoisoningCode ExecutionFrameworkAgentAPI
gradioCWE-5227358 ATLAS
MEDIUM

uv: ZIP parsing differential allows code execution

CVE-2025-13327
6.3
EPSS 0.2%
Supply ChainCode ExecutionFrameworkInference
rhaiis-preview/vllm-cuda-rhel9943 ATLAS
MEDIUM

n8n: XSS enables session hijacking

CVE-2026-27578
5.4
EPSS 0.2%
Prompt InjectionData ExtractionCode ExecutionAgentRAGAPI
n8nCWE-791610 ATLAS
MEDIUM

langgraph-checkpoint: Deserialization enables RCE

CVE-2026-27794
6.6
EPSS 0.8%
Code ExecutionSupply ChainFrameworkAgent
langgraph-checkpointPatch: 4.0.0CWE-5024.1K4 ATLAS
MEDIUM

fickling: Allowlist Bypass evades input filtering

GHSA-mhc9-48gj-9gp3
--
Supply ChainCode ExecutionFrameworkModel
ficklingPatch: 0.1.8CWE-184627 ATLAS
MEDIUMEXPLOIT AVAIL

ray: Missing Auth allows unauthenticated access

CVE-2026-27482
5.9
EPSS 0.3%
Auth BypassDoSFrameworkInference
rayPatch: 2.54.0CWE-3066424 ATLAS
MEDIUM

OpenClaw: path traversal allows arbitrary file write

CVE-2026-26972
6.7
EPSS 0.2%
Code ExecutionData LeakageAgentPlugin
openclawCWE-2233 ATLAS1 incident
MEDIUM

OpenClaw: UI deception enables arbitrary command execution

CVE-2026-26320
6.5
EPSS 0.5%
Social EngineeringCode ExecutionAgent
openclawCWE-45135 ATLAS1 incident
MEDIUM

ffmpeg: security flaw enables exploitation

CVE-2025-12343
5.5
EPSS 0.2%
Code ExecutionAuth BypassDoSFrameworkRAGModel
5 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