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: identity spoofing bypasses agent safety checks

CVE-2026-35655
5.7
EPSS 0.2%
Auth BypassCode ExecutionAgentPlugin
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: auth bypass on Teams feedback invoke

CVE-2026-35654
5.3
EPSS 0.2%
Auth BypassAgentPlugin
OpenClaw33 ATLAS1 incident
MEDIUM

OpenClaw: auth bypass enables unauthorized action execution

CVE-2026-35652
6.5
EPSS 0.4%
Auth BypassCode ExecutionAgent
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: access control bypass via allowlist reconciliation

CVE-2026-35649
6.5
EPSS 0.3%
Auth BypassSupply ChainAgentPlugin
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: access control bypass allows DM policy evasion

CVE-2026-35647
5.3
EPSS 0.3%
Auth BypassSocial EngineeringAgent
OpenClaw33 ATLAS1 incident
MEDIUM

OpenClaw: privilege escalation via scope bypass in allowlist

CVE-2026-35621
6.5
EPSS 0.3%
Auth BypassAgentAPI
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: missing authz enables session policy hijack

CVE-2026-35620
5.4
EPSS 0.4%
Auth BypassData ExtractionAgent
OpenClaw34 ATLAS1 incident
MEDIUMEXPLOIT AVAIL

OpenClaw: auth bypass exposes AI gateway model metadata

CVE-2026-35619
4.3
EPSS 0.3%
Auth BypassData ExtractionAgentAPI
OpenClaw34 ATLAS1 incident
MEDIUM

Log4j: non-finite floats break JSON log ingestion

CVE-2026-34481
--
EPSS 0.7%
DoSFramework
MEDIUMEXPLOIT AVAIL

OpenClaw: SSRF via web-fetch enables internal network pivot

CVE-2026-6011
5.6
EPSS 0.4%
Data ExtractionPrivacy ViolationAgentPlugin
openclawPatch: 2026.1.29CWE-91834 ATLAS1 incident
MEDIUMEXPLOIT AVAIL

PraisonAI: arbitrary file read via unguarded skill tool

CVE-2026-40117
6.2
EPSS 0.3%
Prompt InjectionData ExtractionData LeakageAgentPlugin
praisonaiagentsPatch: 1.5.128CWE-862205 ATLAS
MEDIUMEXPLOIT AVAIL

PraisonAI: XSS via no-op HTML sanitizer in agent output

CVE-2026-40112
5.4
EPSS 0.2%
Prompt InjectionCode ExecutionData ExtractionAgentFrameworkRAG
praisonaiCWE-7914 ATLAS
MEDIUM

OpenClaw: credential exposure via gateway channel URLs

CVE-2026-35644
6.5
EPSS 0.2%
Data ExtractionAuth BypassData LeakageAgentAPIFramework
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: auth bypass injects restricted agent events

CVE-2026-35642
4.3
EPSS 0.2%
Auth BypassCode ExecutionAgentFramework
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: session isolation bypass exposes parent sessions

CVE-2026-35636
6.5
EPSS 0.3%
Auth BypassData LeakageAgent
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: webhook route hijack bypasses DM access controls

CVE-2026-35635
4.8
EPSS 0.2%
Auth BypassData LeakageData ExtractionAgentPlugin
OpenClaw34 ATLAS1 incident
MEDIUM

OpenClaw: auth bypass grants unauthenticated Canvas access

CVE-2026-35634
5.1
EPSS 0.1%
Auth BypassData ExtractionAgentPlugin
OpenClaw35 ATLAS1 incident
MEDIUM

OpenClaw: unbounded memory alloc DoS via crafted HTTP errors

CVE-2026-35633
5.3
EPSS 0.4%
DoSAgent
OpenClaw33 ATLAS
MEDIUM

OpenClaw: auth bypass on ACP mutating commands

CVE-2026-35631
6.5
EPSS 0.2%
Auth BypassData ExtractionAgentPlugin
OpenClaw35 ATLAS1 incident
MEDIUM

OpenClaw: webhook brute-force bypasses AI agent auth

CVE-2026-35628
4.8
EPSS 0.3%
Auth BypassPrompt InjectionAgentAPI
OpenClaw34 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