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
3,628

AI/ML CVEs Tracked

484

Critical

101

New This Week

25

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 1042 results — High severity, no patch
HIGH EXPLOIT AVAIL

Qwen-Agent: path traversal in doc parser leaks files

CVE-2026-82275
7.5
EPSS 0.4%
Data Extraction Data Leakage Agent Plugin
CWE-22 3 ATLAS
HIGH EXPLOIT AVAIL

Qwen-Agent: unauth SSRF via document URL parser

CVE-2026-82268
7.5
EPSS 0.3%
Data Extraction Data Leakage Agent Plugin
CWE-918 3 ATLAS
HIGH

vLLM: unbounded media URLs cause memory-exhaustion DoS

CVE-2026-37237
7.5
EPSS 0.5%
DoS Inference
vllm 92 3 ATLAS
HIGH

Spring AI: predictable ONNX cache path enables model swap

CVE-2026-47852
7.5
EPSS 0.2%
Supply Chain Code Execution Framework Model Inference
3 ATLAS
HIGH EXPLOIT AVAIL

Penpot: stored XSS in comments enables session takeover

CVE-2026-47665
8.7
EPSS 0.3%
Code Execution Data Leakage Framework
CWE-79
HIGH EXPLOIT AVAIL

Pterodactyl Wings: SFTP quota bypass causes node-wide DoS

CVE-2026-61617
7.7
EPSS 0.2%
DoS Inference
CWE-400 1 ATLAS
HIGH EXPLOIT AVAIL

Cohere North: WebSocket leak exposes session data

CVE-2025-61164
7.5
EPSS 0.3%
Data Leakage Auth Bypass Agent API
3 ATLAS
HIGH

Cohere North AI: IDOR allows overwriting any user record

CVE-2025-61162
7.5
EPSS 0.2%
Auth Bypass Privacy Violation API Agent
2 ATLAS
HIGH

Vocos: RCE via unrestricted class loading in model config

CVE-2026-79784
8.8
EPSS 0.3%
Supply Chain Code Execution Model Inference
CWE-470 4 ATLAS
HIGH EXPLOIT AVAIL

mcp-shell: insecure defaults allow full shell RCE

CVE-2026-55580
--
EPSS 0.2%
Code Execution Prompt Injection Agent Plugin
CWE-78 4 ATLAS
HIGH EXPLOIT AVAIL

mcp-shell: bash allowlist bypass grants RCE, no auth

CVE-2026-55581
8.4
EPSS 0.3%
Code Execution Auth Bypass Agent Plugin
CWE-78 3 ATLAS
HIGH EXPLOIT AVAIL

mcp-shell: secure mode bypass via git alias RCE

CVE-2026-55582
8.4
EPSS 0.2%
Code Execution Auth Bypass Agent Plugin
CWE-78 4 ATLAS
HIGH EXPLOIT AVAIL

Ech0: scoped admin token bypass exposes full DB export

CVE-2026-79667
7.6
EPSS 0.2%
Auth Bypass Data Leakage Privacy Violation Framework
1 ATLAS
HIGH

Filament: recovery codes bypass app-based MFA

CVE-2026-77567
8.1
EPSS 0.3%
Auth Bypass Framework
CWE-287
HIGH EXPLOIT AVAIL

Label Studio: IDOR exposes cross-org annotations

CVE-2026-76073
8.8
EPSS 0.3%
Auth Bypass Data Leakage Model Poisoning Training Data API
label-studio 2 4 ATLAS
HIGH EXPLOIT AVAIL

Xinference: RCE via forced trust_remote_code=True

CVE-2026-76841
8.8
EPSS 0.7%
Code Execution Supply Chain Inference Model Framework
CWE-94 4 ATLAS
HIGH

Velociraptor: NOTEBOOK_EDIT users can overwrite artifacts

CVE-2026-19200
8.9
EPSS 0.2%
Auth Bypass Code Execution Framework Plugin
2 ATLAS
HIGH EXPLOIT AVAIL

ONNX: TOCTOU symlink bug allows arbitrary file overwrite

CVE-2026-49114
7.1
EPSS 0.2%
Code Execution Supply Chain Model Framework
onnx 1.1K 3 ATLAS
HIGH

Headroom: SSRF via client-set upstream base URL

CVE-2026-77775
8.6
EPSS 0.4%
Data Leakage Auth Bypass Privacy Violation API Inference
2 ATLAS
HIGH

Rocket.Chat: unauth stored HTML injection baits Livechat agents

CVE-2026-65644
7.5
EPSS 0.3%
Social Engineering Code Execution API
2 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