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

96

New This Week

25

In CISA KEV

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

Latest AI Security Threats

Showing 20 of 4307 results
MEDIUM

Langflow: SSRF in built-in components exposes cloud metadata

GHSA-j8f7-x8jm-wmm4
--
Data LeakageData ExtractionFrameworkAgentPlugin
lfx-doclingPatch: 0.1.2CWE-9184 ATLAS
MEDIUMEXPLOIT AVAIL

vLLM: OOB read in Mamba2 mixer enables remote DoS

CVE-2026-105775
4.3
EPSS 0.3%
DoSInferenceAPIFramework
vLLMCWE-119943 ATLAS
HIGHEXPLOIT AVAIL

simple-git: command injection via git clone config bypass

CVE-2026-102826
8.1
EPSS 0.5%
Code ExecutionSupply ChainFrameworkAgent
CWE-773 ATLAS
MEDIUM

vLLM: unbounded video frame decoding enables DoS

CVE-2026-105758
5.3
EPSS 0.3%
DoSInferenceAPI
vllmPatch: 0.30.0CWE-770943 ATLAS
MEDIUM

vLLM: unbounded video frame options enable DoS

CVE-2026-105760
5.3
EPSS 0.3%
DoSInferenceAPI
vllmPatch: 0.30.0CWE-400943 ATLAS
MEDIUM

vLLM: input validation flaws let attacker crash EngineCore

CVE-2026-105754
6.5
EPSS 0.3%
DoSData LeakageAuth BypassInferenceAPIModel
vllmPatch: 0.30.0CWE-20944 ATLAS
MEDIUM

vLLM: unhandled structured-output errors crash engine (DoS)

CVE-2026-105757
6.5
EPSS 0.3%
DoSInferenceFrameworkAPI
vllmPatch: 0.30.0CWE-20943 ATLAS
MEDIUM

vLLM: request-ID cache key collision corrupts rerank scores

CVE-2026-105755
4.2
EPSS 0.2%
Auth BypassDoSInferenceAPIRAG
vllmPatch: 0.30.0CWE-639944 ATLAS
MEDIUM

vLLM: unvalidated cache_salt crashes EngineCore (DoS)

CVE-2026-105756
6.5
EPSS 0.3%
DoSInferenceAPI
vllmPatch: 0.30.0CWE-20942 ATLAS
MEDIUM

OpenClaw channels: allowlist bypass exposes chat reads

GHSA-g7fw-3gjp-g5hf
6.5
Auth BypassData LeakageData ExtractionAgentPlugin
openclawCWE-86233 ATLAS1 incident
HIGHEXPLOIT AVAIL

vm2: unhandled Promise rejection in construct trap kills host

CVE-2026-100722
8.6
EPSS 0.3%
DoSFrameworkAgentPlugin
vm2Patch: 3.12.2CWE-2483.0K3 ATLAS
HIGHEXPLOIT AVAIL

Joplin Server: app-auth flaw enables account takeover

CVE-2026-105786
--
EPSS 0.4%
Auth BypassSocial EngineeringData LeakageAPI
CWE-3062 ATLAS
HIGHEXPLOIT AVAIL

vLLM: unbounded metrics labels allow memory-exhaustion DoS

CVE-2026-105759
7.5
EPSS 0.4%
DoSInferenceAPI
vllmCWE-400942 ATLAS
MEDIUM

vLLM: reachable assertion in cache causes DoS

CVE-2026-105753
6.5
EPSS 0.4%
DoSInferenceFramework
vllmPatch: 0.28.0CWE-617943 ATLAS
LOW

vLLM: cache_salt dropped on tool continuations leaks prefixes

CVE-2026-105752
3.1
EPSS 0.2%
Data LeakagePrivacy ViolationInferenceAPI
vllmPatch: 0.30.0CWE-200944 ATLAS
MEDIUMEXPLOIT AVAIL

Joplin Server: reset-token reuse enables account takeover

CVE-2026-105785
4.8
EPSS 0.2%
Auth BypassData LeakageAPIPlugin
CWE-6202 ATLAS
MEDIUMEXPLOIT AVAIL

Joplin: CSS injection via jsoncanvas note rendering

CVE-2026-105784
4.6
EPSS 0.2%
Data LeakageSocial EngineeringPlugin
CWE-79
HIGHEXPLOIT AVAIL

Joplin: Web Clipper CORS flaw leaks permanent API token

CVE-2026-105783
8.0
EPSS 0.1%
Auth BypassData LeakageSocial EngineeringAPI
CWE-3463 ATLAS
HIGHEXPLOIT AVAIL

fsspec: Jinja2 SSTI in Kerchunk refs allows RCE

CVE-2026-104851
8.8
EPSS 0.5%
Code ExecutionSupply ChainFrameworkTraining Data
fsspecPatch: 2026.6.0CWE-943614 ATLAS
HIGHEXPLOIT AVAIL

vm2: unhandled host Promise rejection crashes Node host

CVE-2026-92954
8.6
EPSS 0.5%
DoSFrameworkAgentPlugin
vm2Patch: 3.11.8CWE-2483.0K3 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