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 983 results — Medium severity, no patch
MEDIUM

OpenClaw: sandbox path traversal leaks host temp files

CVE-2026-32026
6.5
EPSS 0.4%
Data ExtractionData LeakageAgentPlugin
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: symlink traversal leaks arbitrary local files

CVE-2026-32024
5.5
EPSS 0.3%
Data ExtractionPrivacy ViolationAgent
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: grep safeBins bypass enables arbitrary file read

CVE-2026-32022
6.5
EPSS 0.3%
Data ExtractionAuth BypassData LeakageAgentPlugin
OpenClaw35 ATLAS1 incident
MEDIUM

OpenClaw: auth bypass via Feishu display name spoofing

CVE-2026-32021
6.5
EPSS 0.2%
Auth BypassAgent
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: safeBins bypass enables arbitrary code execution

CVE-2026-32010
6.3
EPSS 0.3%
Auth BypassCode ExecutionAgentFramework
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: binary hijacking via safeBins path bypass

CVE-2026-32009
5.7
EPSS 0.1%
Code ExecutionAuth BypassSupply ChainAgentFrameworkPlugin
OpenClaw35 ATLAS1 incident
MEDIUMEXPLOIT AVAIL

OpenClaw: file:// bypass enables local file exfiltration

CVE-2026-32008
6.5
EPSS 0.4%
Data ExtractionPrivacy ViolationAgentPlugin
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: path traversal enables sandbox file escape

CVE-2026-32007
6.8
EPSS 0.4%
Supply ChainCode ExecutionData ExtractionAgentPlugin
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: auth bypass enables AI agent session poisoning

CVE-2026-32005
6.8
EPSS 0.3%
Auth BypassPrompt InjectionAgentFramework
OpenClaw35 ATLAS1 incident
MEDIUM

OpenClaw: auth bypass exposes protected channel API

CVE-2026-32004
6.5
EPSS 0.3%
Auth BypassSupply ChainAgentPlugin
OpenClaw35 ATLAS1 incident
MEDIUM

OpenClaw: env var injection enables RCE via allowlist bypass

CVE-2026-32003
6.6
EPSS 0.6%
Code ExecutionAuth BypassAgentPlugin
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: sandbox bypass exfiltrates files via vision API

CVE-2026-32002
5.3
EPSS 0.3%
Data ExtractionPrivacy ViolationAuth BypassAgentPlugin
OpenClaw35 ATLAS1 incident
MEDIUM

OpenClaw: auth bypass enables unauthorized node role injection

CVE-2026-32001
5.4
EPSS 0.3%
Auth BypassAgent
OpenClaw35 ATLAS1 incident
MEDIUM

libarchive: zisofs field validation flaw enables DoS

CVE-2026-4426
6.5
DoSFrameworkTraining Data
rhoai/odh-mlflow-operator-rhel96952 ATLAS
MEDIUM

OpenClaw: CWD injection enables integrity loss on Windows

CVE-2026-31999
6.3
EPSS 0.3%
Code ExecutionSupply ChainAgentFramework
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: post-approval PATH rebind enables arbitrary RCE

CVE-2026-31997
6.0
EPSS 0.1%
Code ExecutionAuth BypassAgentPlugin
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: safeBins bypass allows file read/write

CVE-2026-31996
4.4
EPSS 0.1%
Code ExecutionData ExtractionAuth BypassAgentPlugin
OpenClaw36 ATLAS1 incident
MEDIUM

OpenClaw: cmd injection via Windows shell fallback

CVE-2026-31995
5.3
EPSS 0.6%
Code ExecutionSupply ChainAgentPlugin
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: exec approval bypass allows RCE on macOS

CVE-2026-31993
4.8
EPSS 0.3%
Auth BypassCode ExecutionAgentPlugin
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: symlink traversal enables arbitrary file write

CVE-2026-31990
6.1
EPSS 0.1%
Code ExecutionSupply ChainAgent
OpenClaw33 ATLAS1 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