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 4307 results
HIGHEXPLOIT AVAIL

CodeWhale: rlm_eval RCE bypasses approval policy

CVE-2026-75858
7.8
EPSS 0.4%
Prompt InjectionCode ExecutionAuth BypassAgentPlugin
codewhalePatch: 0.8.64CWE-945 ATLAS
HIGH

Kestra: stored XSS via crafted [[link]] flow description

CVE-2026-55839
8.7
EPSS 0.4%
Code ExecutionData LeakageFramework
io.kestra:kestraPatch: 1.3.24CWE-795142 ATLAS
UNKNOWN

n8n: Git node symlink race leads to RCE

CVE-2026-71539
--
EPSS 0.3%
Code ExecutionSupply ChainAgentPlugin
n8nCWE-367164 ATLAS
MEDIUMEXPLOIT AVAIL

tract: malformed ONNX model triggers buffer flaw

CVE-2026-75093
4.3
EPSS 0.6%
Supply ChainDoSModelInference
CWE-1203 ATLAS
HIGHEXPLOIT AVAIL

MLflow: missing authz check leaks private artifacts

CVE-2026-69148
7.1
EPSS 0.4%
Auth BypassData LeakageModelAPI
mlflowPatch: 3.15.0CWE-8626953 ATLAS
MEDIUMEXPLOIT AVAIL

MLflow: missing authZ lets users forge dataset lineage

CVE-2026-69146
6.5
EPSS 0.4%
Auth BypassModel PoisoningFrameworkTraining DataAPI
mlflowPatch: 3.15.0CWE-8626952 ATLAS
CRITICALKEVSCANNER

MLflow: unauth SSRF via webhook hits cloud metadata

CVE-2026-64849
9.3
EPSS 10.0%
Data ExtractionAuth BypassData LeakageFrameworkAPI
mlflowPatch: 3.15.0CWE-9186952 ATLAS
MEDIUM

go-pkcs12: PBMAC1 flaw lets wrong-password files decode

GHSA-mpwr-8vm7-h73f
--
Auth BypassInference
software.sslmate.com/src/go-pkcs12Patch: 0.7.2CWE-206.3K
HIGH

atomic-agents: unpriced models silently bypass cost cap

GHSA-j659-8xh6-5pq5
--
DoSAgentFramework
atomic-agents-stackPatch: 1.1.0CWE-770942 ATLAS
HIGHEXPLOIT AVAIL

Evidently: unauth path traversal leaks any file

CVE-2026-75111
7.5
EPSS 0.6%
Data LeakagePrivacy ViolationTraining DataFramework
CWE-223 ATLAS
MEDIUMEXPLOIT AVAIL

Transformers: path traversal via checkpoint shards

CVE-2026-75104
5.5
EPSS 0.3%
Data LeakageSupply ChainFrameworkModel
transformersCWE-229.8K3 ATLAS
MEDIUMEXPLOIT AVAIL

vLLM: SSRF/LFI bypass in MiMo-V2 multimodal input

CVE-2026-73560
6.5
EPSS 0.4%
Data ExtractionAuth BypassInferenceModel
vllmPatch: 0.26.0CWE-918943 ATLAS
HIGHEXPLOIT AVAIL

Budibase: SSRF in automations reaches cloud metadata

CVE-2026-35219
--
EPSS 0.5%
Data ExtractionAuth BypassAgentPlugin
3 ATLAS
MEDIUM

vLLM: unbounded /derender inputs cause resource DoS

CVE-2026-71486
4.3
EPSS 0.4%
DoSInferenceAPI
vllmPatch: 0.26.0CWE-400943 ATLAS
HIGH

PyCharm: unauthenticated Jupyter MCP tools enable RCE

CVE-2026-75060
8.4
EPSS 0.2%
Code ExecutionAuth BypassAgentPlugin
CWE-3063 ATLAS
HIGHEXPLOIT AVAIL

SiYuan: unthrottled brute-force exposes published notes

CVE-2026-74868
7.5
EPSS 0.5%
Auth BypassData LeakageRAG
2 ATLAS
HIGHEXPLOIT AVAIL

Edimax EW-7478APC: command injection via rootAPmac

CVE-2026-19960
7.4
EPSS 2.0%
Code ExecutionAPI
HIGHEXPLOIT AVAIL

SiYuan: brute-force bypass of notebook publish passwords

CVE-2026-73045
7.5
EPSS 0.5%
Auth BypassPlugin
CWE-3072 ATLAS
CRITICALEXPLOIT AVAIL

SiYuan: SQL injection enables cross-notebook DB access

CVE-2026-72811
10.0
EPSS 0.4%
Code ExecutionData ExtractionPrivacy ViolationRAGPlugin
github.com/siyuan-note/siyuan/kernelPatch: 0.0.0-20260723004839-1a5b3431d5abCWE-893.0K3 ATLAS
UNKNOWN

JupyterLab: settings CSS injection leads to code exec

CVE-2026-73417
--
EPSS 0.7%
Code ExecutionSocial EngineeringFramework
CWE-793 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