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
2,387

AI/ML CVEs Tracked

341

Critical

358

New This Week

18

In CISA KEV

Sort: Date CVSS EPSS KEV first Most exploited
Filter: All Critical High Medium KEV only Active exploitation Has patch No patch

Latest AI Security Threats

Showing 20 of 307 results — Medium severity, has patch
MEDIUM EXPLOIT AVAIL

open-webui: XSS in pending overlay enables session hijack

CVE-2026-44568
4.8
EPSS 0.2%
Data Extraction Code Execution Social Engineering Inference Framework
open-webui Patch: 0.9.0 CWE-79 4 ATLAS
MEDIUM

open-webui: RAG auth bypass exposes private files

CVE-2026-44560
6.5
EPSS 0.4%
Auth Bypass Data Extraction Privacy Violation RAG API Framework
open-webui Patch: 0.9.0 CWE-862 5 ATLAS
MEDIUM EXPLOIT AVAIL

open-webui: auth bypass exposes private group channels

CVE-2026-44561
5.4
EPSS 0.2%
Auth Bypass Data Extraction Privacy Violation API
open-webui Patch: 0.9.0 CWE-284 4 ATLAS
MEDIUM

open-webui: auth bypass in collaborative doc editing

CVE-2026-44564
5.4
EPSS 0.2%
Auth Bypass Prompt Injection DoS API Framework
open-webui Patch: 0.9.0 CWE-863 4 ATLAS
MEDIUM

open-webui: auth bypass exposes restricted LLM models

CVE-2026-44563
5.4
EPSS 0.2%
Auth Bypass Data Extraction Inference API
open-webui Patch: 0.9.0 CWE-862 5 ATLAS
MEDIUM

open-webui: missing authz enables model hijacking

CVE-2026-44562
6.5
EPSS 0.3%
Auth Bypass Model Poisoning Data Extraction Model API Framework
open-webui Patch: 0.9.0 CWE-862 5 ATLAS
MEDIUM EXPLOIT AVAIL

open-webui: private channel member list exposed to any user

CVE-2026-44559
4.3
EPSS 0.2%
Auth Bypass Data Extraction Privacy Violation API Framework
open-webui Patch: 0.9.0 CWE-862 4 ATLAS
MEDIUM

open-webui: auth bypass exposes all knowledge base metadata

CVE-2026-44557
4.3
EPSS 0.2%
Auth Bypass Data Extraction Privacy Violation RAG Framework API
open-webui Patch: 0.9.0 CWE-200 5 ATLAS
MEDIUM EXPLOIT AVAIL

open-webui: permission bypass exposes channels publicly

CVE-2026-44558
5.4
EPSS 0.2%
Auth Bypass Data Leakage API Framework
open-webui Patch: 0.9.0 CWE-863 3 ATLAS
MEDIUM

open-webui: mass assignment enables cross-user folder injection

CVE-2026-44550
5.0
EPSS 0.3%
Auth Bypass Social Engineering Privacy Violation Framework API
open-webui Patch: 0.9.0 CWE-862 4 ATLAS
MEDIUM EXPLOIT AVAIL

BentoML: symlink traversal exfiltrates host secrets at build

CVE-2026-40610
5.5
EPSS 0.3%
Data Extraction Supply Chain Framework
bentoml Patch: 1.4.39 CWE-59 21 4 ATLAS
MEDIUM

@axonflow/openclaw: credential exposure via insecure file permissions

GHSA-cqmh-pcgr-q42f
5.5
Data Leakage Auth Bypass Privacy Violation Plugin Agent
@axonflow/openclaw Patch: 2.0.0 CWE-552 4 5 ATLAS
MEDIUM EXPLOIT AVAIL

vLLM: speculative decoding DoS via penalty params

CVE-2026-44223
6.5
EPSS 0.4%
DoS Inference
vllm Patch: 0.20.0 CWE-131 130 2 ATLAS
MEDIUM

vLLM: token injection DoS via multimodal placeholders

CVE-2026-44222
6.5
EPSS 0.4%
DoS Prompt Injection Inference Model Framework
vllm Patch: 0.20.0 CWE-129 130 5 ATLAS
MEDIUM

openclaw: stale webhook secret survives credential rotation

GHSA-q8ff-7ffm-m3r9
6.0
Auth Bypass Agent Plugin
openclaw Patch: 2026.4.23 CWE-613 4 3 ATLAS 1 incident
MEDIUM

JupyterHub: CSRF bypass on spawn and share endpoints

CVE-2026-40864
5.4
EPSS 0.2%
Auth Bypass DoS Framework
jupyterhub Patch: 5.4.5 CWE-352 1.9K 4 ATLAS
MEDIUM

jupyter-server: auth cookie survives password reset

CVE-2026-40934
6.8
EPSS 0.3%
Auth Bypass Data Extraction Framework API
jupyter-server Patch: 2.18.0 CWE-613 1.9K 4 ATLAS
MEDIUM EXPLOIT AVAIL

jupyter-server: Open redirect enables credential phishing

CVE-2025-61669
--
EPSS 0.3%
Social Engineering Auth Bypass Framework API
jupyter-server Patch: 2.18.0 CWE-601 1.9K 5 ATLAS
MEDIUM

OpenClaw: symlink traversal exposes host filesystem

CVE-2026-43570
6.5
EPSS 0.3%
Supply Chain Data Extraction Agent Plugin
openclaw Patch: 2026.4.5 CWE-61 4 4 ATLAS 1 incident
MEDIUM

openclaw: auth bypass exposes Gateway bootstrap config

GHSA-93rg-2xm5-2p9v
--
Auth Bypass Data Leakage Agent API
openclaw Patch: 2026.4.22 CWE-287 4 4 ATLAS 1 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