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 544 results — Medium severity, has patch
MEDIUMEXPLOIT AVAIL

Keras: incomplete fix reopens HDF5 arbitrary file read

CVE-2026-12480
5.5
EPSS 0.2%
Data LeakageSupply ChainPrivacy ViolationModelTraining Data
kerasPatch: 3.12.3CWE-734.1K4 ATLAS
MEDIUM

Pterodactyl panel: email endpoint leaks user enumeration

GHSA-j7f5-gfqm-pcx3
--
Privacy ViolationAuth Bypass
pterodactyl/panelPatch: 1.12.3CWE-204514
MEDIUMEXPLOIT AVAIL

Gogs: XSS via data URI in ipynb sanitizer endpoint

CVE-2026-52816
--
EPSS 0.7%
Code ExecutionData LeakageAuth BypassAPI
gogs.io/gogsPatch: 0.14.3CWE-791.9K5 ATLAS
MEDIUM

Filament: unauth file upload drains disk/inflates costs

CVE-2026-48500
6.5
EPSS 0.3%
DoSAuth BypassFramework
filament/filamentPatch: 4.11.5CWE-8625144 ATLAS
MEDIUMEXPLOIT AVAIL

Keras: path traversal via layer names allows arbitrary file write

CVE-2026-12479
6.1
EPSS 0.4%
Supply ChainCode ExecutionFrameworkModel
kerasPatch: 3.12.3CWE-224.1K4 ATLAS
MEDIUMEXPLOIT AVAIL

Langflow: logout fails to clear session tokens

CVE-2026-55423
6.1
EPSS 0.2%
Auth BypassData ExtractionFrameworkAgent
langflowPatch: 1.7.0CWE-6133 ATLAS
MEDIUM

containerd: checkpoint import poisons node image cache

CVE-2026-50195
--
EPSS 0.3%
Supply ChainCode ExecutionInferenceFramework
github.com/containerd/containerd/v2Patch: 2.1.9CWE-3456.3K4 ATLAS
MEDIUM

JupyterLab: XSS via malicious PyPI extension URL

GHSA-vmhf-c436-hxj4
--
Supply ChainCode ExecutionSocial EngineeringPluginFramework
jupyterlabPatch: 4.5.9CWE-841.9K5 ATLAS
MEDIUMEXPLOIT AVAIL

tract-onnx: path traversal exposes arbitrary local files

CVE-2026-55832
6.1
EPSS 0.2%
Supply ChainData ExtractionModelInference
tract-onnxPatch: 0.21.17CWE-221.1K4 ATLAS
MEDIUMEXPLOIT AVAIL

nl-portal: GraphQL SSRF exposes Objecten-API auth token

CVE-2026-55414
5.3
EPSS 0.4%
Auth BypassData ExtractionPrivacy ViolationAPIFramework
nl.nl-portal:formPatch: 3.0.4CWE-8626423 ATLAS
MEDIUM

canto-saas-api: OAuth credentials logged via query params

CVE-2026-55375
5.3
EPSS 0.4%
Data ExtractionAuth BypassAPI
jleehr/canto-saas-apiPatch: 3.0.0CWE-2093 ATLAS
MEDIUMEXPLOIT AVAIL

nl-portal documenten-api: IDOR exposes citizen documents

CVE-2026-54683
6.5
EPSS 0.4%
Auth BypassData ExtractionPrivacy ViolationAPI
nl.nl-portal:documenten-apiPatch: 3.0.3CWE-2856423 ATLAS
MEDIUM

@theia/ai-chat: prompt injection exfiltrates workspace secrets

CVE-2026-22551
--
EPSS 0.3%
Prompt InjectionData ExtractionData LeakageAgentFrameworkPlugin
@theia/ai-idePatch: 1.71.0CWE-2014 ATLAS
MEDIUM

Kirby CMS: CRLF injection overrides outbound HTTP headers

CVE-2026-50188
--
EPSS 0.4%
Auth BypassData LeakagePluginAPI
getkirby/cmsPatch: 4.9.4CWE-935144 ATLAS
MEDIUM

Kirby CMS: missing auth exposes restricted page titles

CVE-2026-49274
--
EPSS 0.5%
Auth BypassData LeakageFramework
getkirby/cmsPatch: 4.9.4CWE-8625143 ATLAS
MEDIUM

praisonai-platform: cross-tenant IDOR poisons project stats

GHSA-2fjj-qqg8-fg7x
4.3
Auth BypassAgentFramework
praisonai-platformPatch: 0.1.8CWE-63913 ATLAS
MEDIUM

praisonaiagents: sandbox escape via format-spec read

GHSA-pv2j-rghr-v5r9
6.5
Code ExecutionPrompt InjectionData ExtractionAgentFrameworkPlugin
praisonaiagentsPatch: 1.6.59CWE-693206 ATLAS
MEDIUM

praisonaiagents: SSRF bypass via SpiderTools redirect

GHSA-6h9p-93hq-q7h6
6.5
Data ExtractionAuth BypassAgentPlugin
praisonaiagentsPatch: 1.6.59CWE-918205 ATLAS
MEDIUM

praisonaiagents: unauth SSE endpoint enables event injection

GHSA-35w5-pcw4-jx94
4.3
Auth BypassData LeakageAgentFramework
praisonaiagentsPatch: 1.6.59CWE-306204 ATLAS
MEDIUM

marimo: reflected XSS enables JS injection in notebooks

CVE-2026-54386
6.1
EPSS 0.2%
Code ExecutionData ExtractionFramework
marimoPatch: 0.23.9CWE-793.0K3 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