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
2,388

AI/ML CVEs Tracked

342

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 1074 results — Active exploitation
Severity CVE ID Summary CVSS EPSS Package Date
MEDI E CVE-2025-5197 Transformers: ReDoS in TF-to-PyTorch weight converter 5.3 0.4% transformers Aug 6 CRIT E CVE-2025-45150 ChatGLM-Webui: arbitrary file read, no auth required 9.8 0.6% langchain-chatglm-webui Aug 1 CRIT E CVE-2025-54381 BentoML: unauthenticated SSRF via file upload URLs 9.9 11.1% bentoml Jul 29 CRIT E CVE-2025-46059 LangChain GmailToolkit: indirect prompt injection to RCE 9.8 0.7% — Jul 29 CRIT E CVE-2025-5120 smolagents: sandbox escape enables unauthenticated RCE 10.0 17.7% smolagents Jul 27 HIGH E CVE-2025-54413 skops: RCE via MethodNode unsafe deserialization — 0.1% skops Jul 26 HIGH E CVE-2025-54412 skops: OperatorFuncNode type confusion → RCE — 0.1% skops Jul 26 MEDI E CVE-2025-51471 Ollama: auth token hijack via crafted WWW-Authenticate 6.9 3.8% ollama Jul 22 MEDI E CVE-2025-51481 Dagster: path traversal exposes arbitrary file read via gRPC 6.6 0.5% — Jul 22 MEDI E CVE-2025-3933 Transformers: ReDoS in DonutProcessor causes DoS 5.3 0.4% transformers Jul 11 MEDI E CVE-2025-6716 Contest Gallery WP Plugin: Stored XSS in OpenAI integration 6.4 0.2% — Jul 11 MEDI E CVE-2025-7021 OpenAI Operator: fullscreen spoofing captures credentials 6.5 0.3% operator Jul 10 MEDI E CVE-2025-6211 llama-index: DocugamiReader MD5 hash collision drops chunks 6.5 0.3% llama-index-readers-docugami Jul 10 HIGH E CVE-2025-6209 llama_index: path traversal allows arbitrary file read 7.5 0.5% llama-index-core Jul 7 MEDI E CVE-2025-6210 llama-index Obsidian reader: hardlink path traversal leaks files 6.2 0.3% llama-index-readers-obsidian Jul 7 MEDI E CVE-2025-5472 llama-index: JSONReader DoS via recursive JSON parsing 6.5 0.3% llama-index-core Jul 7 HIGH E CVE-2025-3225 llama-index Papers Loader: XML expansion DoS 7.5 0.4% llama-index-readers-papers Jul 7 HIGH E CVE-2025-3046 LlamaIndex Obsidian: symlink traversal exposes host files 7.5 0.6% llama-index-readers-obsidian Jul 7 MEDI E CVE-2025-3044 llama-index ArxivReader: MD5 collision corrupts training data 5.3 0.3% llama-index-readers-papers Jul 7 LOW E CVE-2025-3777 Transformers: URL validation bypass exposes image pipeline 3.5 0.3% transformers Jul 7

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