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

vLLM Gateway: logs leak bearer tokens & chat PII

CVE-2026-15574
7.5
EPSS 0.5%
Data LeakagePrivacy ViolationInferenceAPI
CWE-5383 ATLAS
HIGHEXPLOIT AVAIL

PraisonAI: weak default block threshold lets prompt injection through

CVE-2026-61439
7.5
EPSS 0.3%
Prompt InjectionData ExtractionData LeakageAgentFrameworkPlugin
praisonaiCWE-118815 ATLAS1 incident
HIGHEXPLOIT AVAIL

PraisonAI: SSRF bypass via DNS rebinding/redirects

CVE-2026-61429
8.5
EPSS 0.3%
Data LeakageAuth BypassAgentPlugin
praisonaiCWE-91813 ATLAS
HIGHEXPLOIT AVAIL

PraisonAI: insecure defaults expose agent secrets

CVE-2026-61426
8.6
EPSS 0.4%
Auth BypassData LeakageData ExtractionAgentAPI
praisonaiCWE-20014 ATLAS
HIGH

h2o: QPACK alloca overflow crashes HTTP/3 server

CVE-2026-55213
7.5
EPSS 0.5%
DoSInference
CWE-7892 ATLAS
HIGH

Phantom MCP: unconfined output path enables file write

GHSA-52vm-mxx8-f227
7.7
Code ExecutionDoSAgentPlugin
CWE-224 ATLAS
HIGHEXPLOIT AVAIL

Gradio: SSRF via file_fetch() leaks cloud IAM creds

CVE-2026-59806
7.4
EPSS 0.4%
Data LeakageSocial EngineeringFrameworkAPI
gradioCWE-6017355 ATLAS
HIGH

n8n: workflow:read scope bypass triggers execution

CVE-2026-56776
7.4
EPSS 0.3%
Auth BypassCode ExecutionAgentPlugin
n8nCWE-863163 ATLAS
HIGHEXPLOIT AVAIL

OpenLLM: command injection via model repo name

CVE-2026-15035
7.8
EPSS 2.2%
Code ExecutionInferenceFramework
2 ATLAS
HIGHEXPLOIT AVAIL

Foxit PDF: use-after-free in attachment panel via JS page deletion

CVE-2026-57252
7.8
EPSS 0.2%
Code ExecutionDoSPlugin
HIGH

better-auth: race condition mints duplicate OAuth tokens

CVE-2026-53518
8.1
EPSS 0.4%
Auth BypassData LeakageAgentAPIFramework
CWE-2941 ATLAS
HIGH

Joomla: broken access control exposes com_privacy data

CVE-2026-48957
8.8
EPSS 0.4%
Auth BypassData LeakagePrivacy ViolationAPITraining Data
CWE-2843 ATLAS
HIGH

AMP for WP: unsafe ZIP extraction enables file write

CVE-2026-6101
7.5
EPSS 1.0%
Code ExecutionPlugin
HIGH

vLLM: unbounded regex compile hangs inference worker (ReDoS)

CVE-2026-55574
7.5
EPSS 0.6%
DoSInference
xgrammarCWE-13331323 ATLAS
HIGHEXPLOIT AVAIL

fickling: denylist gaps let malicious pickles execute

CVE-2026-14534
8.8
EPSS 0.6%
Code ExecutionSupply ChainFrameworkModel
fickling625 ATLAS
HIGHEXPLOIT AVAIL

HestiaCP: cronjob flaw lets low-priv users get root

CVE-2026-12196
--
EPSS 0.4%
Auth BypassCode ExecutionFramework
1 ATLAS
HIGH

n8n: authenticated RCE via Execute Command node

CVE-2025-71380
8.8
EPSS 0.6%
Code ExecutionAuth BypassAgentPlugin
n8nCWE-284163 ATLAS
HIGHEXPLOIT AVAIL

picklescan: detection bypass enables pickle RCE

CVE-2025-71375
8.1
EPSS 0.5%
Supply ChainCode ExecutionModelFramework
picklescanCWE-50234 ATLAS
HIGHEXPLOIT AVAIL

picklescan: methodcaller bypass enables pickle RCE

CVE-2025-71373
8.1
EPSS 0.6%
Code ExecutionSupply ChainModelFramework
picklescanCWE-69334 ATLAS
HIGHEXPLOIT AVAIL

Picklescan: RCE gadget bypasses pickle safety scanner

CVE-2025-71372
8.1
EPSS 0.5%
Supply ChainCode ExecutionModel
picklescanCWE-50235 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