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
4,331

AI/ML CVEs Tracked

535

Critical

108

New This Week

25

In CISA KEV

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

Latest AI Security Threats

Showing 20 of 1295 results — High severity, no patch
HIGH

OpenClaw: workspace plugin auto-load enables RCE

CVE-2026-32920
8.4
EPSS 0.3%
Supply ChainCode ExecutionAgentPlugin
OpenClaw36 ATLAS1 incident
HIGH

1millionbot Millie: Boolean prompt injection bypasses restrictions

CVE-2026-4399
7.5
EPSS 0.3%
Prompt InjectionJailbreakAPIAgent
6 ATLAS3 incidents
HIGH

gdk-pixbuf: JPEG heap overflow crashes vLLM inference

CVE-2026-5201
7.5
EPSS 1.1%
DoSInference
rhaiis/vllm-cuda-rhel9942 ATLAS
HIGHEXPLOIT AVAIL

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

CVE-2026-29872
8.2
EPSS 0.3%
Data LeakageAuth BypassPrivacy ViolationAgentAPIPlugin
7 ATLAS
HIGH

CrewAI: arbitrary file read via JSON loader tool

CVE-2026-2285
7.5
EPSS 0.6%
Data ExtractionData LeakagePrompt InjectionFrameworkAgentPlugin
6 ATLAS
HIGHEXPLOIT AVAIL

libarchive: integer overflow in zisofs hits vllm containers

CVE-2026-5121
7.5
EPSS 1.1%
Code ExecutionData ExtractionInference
rhaiis/vllm-cuda-rhel9943 ATLAS
HIGH

OpenClaw: gateway credential exposed in pairing setup codes

CVE-2026-33575
7.5
EPSS 0.3%
Auth BypassData LeakageAgentAPI
OpenClaw35 ATLAS1 incident
HIGH

OpenClaw: workspace boundary bypass, arbitrary exec

CVE-2026-33573
8.8
EPSS 0.3%
Auth BypassCode ExecutionData ExtractionAgentFramework
OpenClaw37 ATLAS1 incident
HIGH

OpenClaw: insecure transcript files expose agent secrets

CVE-2026-33572
8.4
EPSS 0.1%
Data ExtractionPrivacy ViolationAgentFramework
OpenClaw34 ATLAS1 incident
HIGH

OpenClaw: unauthenticated webhook DoS via body buffering

CVE-2026-32980
7.5
EPSS 0.5%
DoSAgent
OpenClaw33 ATLAS
HIGH

OpenClaw: TOCTOU race enables local code execution

CVE-2026-32979
7.3
EPSS 0.1%
Code ExecutionSupply ChainAgent
OpenClaw35 ATLAS1 incident
HIGH

OpenClaw: script approval bypass allows RCE

CVE-2026-32978
8.0
EPSS 0.2%
Auth BypassCode ExecutionSupply ChainAgentPlugin
OpenClaw34 ATLAS1 incident
HIGH

OpenClaw: auth bypass triggers forged webhook tool execution

CVE-2026-32974
8.6
EPSS 0.3%
Auth BypassCode ExecutionAgentPlugin
OpenClaw34 ATLAS1 incident
HIGH

OpenClaw: auth bypass enables persistent CDP backdoor

CVE-2026-32972
7.1
EPSS 0.3%
Auth BypassCode ExecutionAgentPlugin
OpenClaw34 ATLAS1 incident
HIGH

OpenClaw: session sandbox escape exposes cross-agent state

CVE-2026-32918
8.4
EPSS 0.1%
Auth BypassData ExtractionPrivacy ViolationAgentFramework
OpenClaw36 ATLAS1 incident
HIGH

OpenClaw: sandbox bypass enables sibling agent hijack

CVE-2026-32915
8.8
EPSS 0.1%
Auth BypassCode ExecutionAgentFramework
OpenClaw35 ATLAS1 incident
HIGH

OpenClaw: access control bypass in config/debug handlers

CVE-2026-32914
8.8
EPSS 0.3%
Auth BypassData ExtractionData LeakageAgentFramework
OpenClaw35 ATLAS2 incidents
HIGHEXPLOIT AVAIL

MLflow: broken access control exposes experiment traces

CVE-2025-15381
8.1
EPSS 0.3%
Auth BypassData ExtractionData LeakageFrameworkTraining Data
mlflowCWE-2006955 ATLAS
HIGHEXPLOIT AVAIL

OpenClaw: path traversal in media parsing leaks secrets

CVE-2026-32846
7.5
EPSS 0.7%
Data LeakageData ExtractionAgent
OpenClawCWE-2233 ATLAS1 incident
HIGH

libtiff: integer overflow causes heap OOB write

CVE-2026-4775
7.8
EPSS 0.6%
Code ExecutionDoSInferenceModel
rhaiis/vllm-spyre-rhel9943 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