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
3,590

AI/ML CVEs Tracked

481

Critical

95

New This Week

24

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 472 results — Medium severity, Active exploitation
MEDIUM EXPLOIT AVAIL

NLTK: dead sandbox check enables arbitrary file read

CVE-2026-65915
6.5
EPSS 0.4%
Data Leakage Auth Bypass Framework
nltk Patch: 3.10.0 CWE-284 3.1K 2 ATLAS
MEDIUM EXPLOIT AVAIL

NLTK: symlink escape allows arbitrary file read

CVE-2026-70626
6.2
EPSS 0.2%
Data Leakage Privacy Violation Training Data Framework
nltk Patch: 3.9.4 CWE-59 2.9K 3 ATLAS
MEDIUM EXPLOIT AVAIL

Ollama: integer overflow in GGUF decoder causes DoS

CVE-2026-86289
4.3
EPSS 0.4%
DoS Supply Chain Inference Model
ollama CWE-189 1.8K 3 ATLAS
MEDIUM EXPLOIT AVAIL

tokenizers: malformed vocab crashes process (DoS)

CVE-2026-85670
6.5
EPSS 0.3%
DoS Supply Chain Inference Training Data
tokenizers CWE-787 15.9K 3 ATLAS
MEDIUM EXPLOIT AVAIL

SiYuan: broken authz leaks DB schema & block IDs

CVE-2026-72800
5.8
EPSS 0.2%
Auth Bypass Data Leakage API
github.com/siyuan-note/siyuan/kernel Patch: 0.0.0-20260724103335-f36331956ae9 CWE-862 3.1K 2 ATLAS
MEDIUM EXPLOIT AVAIL

SiYuan: auth bypass exposes private notebook content

CVE-2026-72805
5.8
EPSS 0.2%
Auth Bypass Data Leakage Privacy Violation RAG API
github.com/siyuan-note/siyuan/kernel Patch: 0.0.0-20260723163028-931ba693375e CWE-862 3.1K 1 ATLAS
MEDIUM EXPLOIT AVAIL

ckan-mcp-server: cache key collision enables poisoning

CVE-2026-73846
6.5
EPSS 0.2%
Prompt Injection Agent Plugin
CWE-345 2 ATLAS
MEDIUM EXPLOIT AVAIL

pdfme: XSS via unsanitized i18n labels in Designer

CVE-2026-82865
4.4
EPSS 0.1%
Code Execution Data Leakage Framework
CWE-79 2 ATLAS
MEDIUM EXPLOIT AVAIL

browser-use web-ui: stores LLM API keys in cleartext

CVE-2026-82640
5.5
EPSS 0.1%
Data Leakage Privacy Violation Agent API
CWE-312 4 ATLAS
MEDIUM EXPLOIT AVAIL

browser-use: arbitrary directory creation via paths

CVE-2026-82637
5.3
EPSS 0.2%
Auth Bypass DoS Agent Framework
browser-use CWE-73 72 1 ATLAS
MEDIUM EXPLOIT AVAIL

ash_paper_trail: sensitive fields leak in version logs

CVE-2026-75847
--
EPSS 0.1%
Data Leakage Privacy Violation Framework
1 ATLAS
MEDIUM EXPLOIT AVAIL

Chainlit: IDOR lets users tamper with others' feedback

CVE-2026-82290
5.3
EPSS 0.2%
Auth Bypass Framework Training Data
chainlit CWE-639 41 2 ATLAS
MEDIUM EXPLOIT AVAIL

crewai-tools: SQLi in NL2SQLTool allows DB compromise

CVE-2026-37009
6.5
EPSS 0.3%
Code Execution Data Extraction Agent Plugin
3 ATLAS
MEDIUM EXPLOIT AVAIL

Booking Calendar: IDOR enables employee account takeover

CVE-2026-14212
4.7
EPSS 0.3%
Auth Bypass Plugin
MEDIUM EXPLOIT AVAIL

PraisonAI: TOCTOU SSRF in webhook validation

CVE-2026-55535
6.8
EPSS 0.2%
Data Leakage Auth Bypass Agent API
PraisonAI Patch: 4.6.58 CWE-367 1 3 ATLAS
MEDIUM EXPLOIT AVAIL

PraisonAI: missing authz lets agent mass-rewrite files

CVE-2026-55530
6.1
EPSS 0.1%
Auth Bypass Code Execution Agent Framework
praisonaiagents Patch: 1.6.58 CWE-862 20 3 ATLAS
MEDIUM EXPLOIT AVAIL

PraisonAI: unauth memory exhaustion via MCP sessions

CVE-2026-55531
6.5
EPSS 0.2%
DoS Agent
PraisonAI Patch: 4.6.58 CWE-400 1 2 ATLAS
MEDIUM EXPLOIT AVAIL

PraisonAI: MCP origin-check bypass runs local tools

CVE-2026-55529
6.9
EPSS 0.1%
Auth Bypass Code Execution Agent Plugin
PraisonAI Patch: 4.6.58 CWE-306 1 4 ATLAS
MEDIUM EXPLOIT AVAIL

Ech0: scope bypass exposes comment admin endpoints

CVE-2026-79672
5.5
EPSS 0.2%
Auth Bypass Framework
CWE-862
MEDIUM EXPLOIT AVAIL

BentoML: SSRF via unfiltered CGNAT range

CVE-2026-78205
5.8
EPSS 0.3%
Data Leakage Auth Bypass Inference Framework
bentoml CWE-918 20 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