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

95

New This Week

25

In CISA KEV

Sort:DateCVSSEPSSKEV firstMost exploited
Filter:AllCriticalHighMediumKEV onlyActive exploitationHas patchNo patch

Latest AI Security Threats

Showing 20 of 1961 results — Active exploitation
HIGHEXPLOIT AVAIL

n8n: unrestricted file upload RCE via Chat Trigger

CVE-2025-56265
8.8
EPSS 0.6%
Code ExecutionData ExtractionAuth BypassFrameworkAgentPlugin
n8n165 ATLAS
MEDIUMEXPLOIT AVAIL

xgrammar: DoS via oversized JSON schema grammar parsing

CVE-2025-58446
--
EPSS 0.6%
DoSInferenceFramework
xgrammarPatch: 0.1.24CWE-7701323 ATLAS
HIGHEXPLOIT AVAILSCANNER

EverNoteLoader: XXE exposes host files in LangChain

CVE-2025-6984
7.5
EPSS 1.6%
Data ExtractionData LeakageFrameworkRAG
langchain-communityPatch: 0.3.27CWE-2001.3K4 ATLAS
CRITICALEXPLOIT AVAIL

n8n-workflows: path traversal in download_workflow endpoint

CVE-2025-55526
9.1
EPSS 0.8%
Data ExtractionAuth BypassCode ExecutionAgentFrameworkAPI
fastapi165 ATLAS
HIGHEXPLOIT AVAIL

llama-index: JSON parsing DoS via deep recursion

CVE-2025-5302
8.6
EPSS 0.3%
DoSFrameworkRAG
llama-index-corePatch: 0.12.38CWE-6741.3K3 ATLAS
HIGHEXPLOIT AVAIL

xgrammar: uncontrolled recursion in grammar parsing causes DoS

CVE-2025-57809
7.5
EPSS 0.5%
DoSSupply ChainFrameworkInference
xgrammarPatch: 0.1.21CWE-6741323 ATLAS
HIGHEXPLOIT AVAIL

Langflow: privilege escalation to full superuser via CLI

CVE-2025-57760
8.8
EPSS 0.5%
Auth BypassCode ExecutionFrameworkAgent
langflowCWE-2695 ATLAS
HIGHEXPLOIT AVAIL

vLLM: unauthenticated DoS via oversized HTTP header

CVE-2025-48956
7.5
EPSS 0.6%
DoSInferenceAPIFramework
vllmCWE-400943 ATLAS
HIGHEXPLOIT AVAIL

Keras: safe mode bypass enables RCE via model load

CVE-2025-8747
7.8
EPSS 0.1%
Code ExecutionSupply ChainFrameworkModel
kerasCWE-5024.1K5 ATLAS
HIGHEXPLOIT AVAIL

skops: joblib fallback enables RCE via model load

CVE-2025-54886
8.4
EPSS 0.2%
Supply ChainCode ExecutionFrameworkModel
skopsPatch: 0.13.0CWE-5027665 ATLAS
MEDIUMEXPLOIT AVAIL

ExecuTorch: integer overflow enables RCE via model loading

CVE-2025-54952
--
EPSS 0.6%
Code ExecutionSupply ChainFrameworkInferenceModel
executorchCWE-68074 ATLAS
CRITICALEXPLOIT AVAIL

ExecuTorch: OOB read in model loader enables RCE

CVE-2025-54950
9.8
EPSS 0.6%
Code ExecutionSupply ChainFrameworkModelInference
executorchPatch: 0.7.0CWE-12574 ATLAS
CRITICALEXPLOIT AVAIL

Azure OpenAI: SSRF EoP, no auth required (CVSS 10)

CVE-2025-53767
10.0
EPSS 1.1%
Auth BypassData ExtractionPrivacy ViolationAPIInference
azure_openai15.6K6 ATLAS
MEDIUMEXPLOIT AVAIL

Ollama: arbitrary file deletion via /api/pull

CVE-2025-44779
6.6
EPSS 0.2%
DoSData LeakageInferenceFramework
ollama1.9K4 ATLAS
MEDIUMEXPLOIT AVAIL

Transformers: ReDoS in TF-to-PyTorch weight converter

CVE-2025-5197
5.3
EPSS 0.4%
DoSSupply ChainFramework
transformersCWE-13339.8K4 ATLAS
LOWEXPLOIT AVAIL

CIRCL FourQ: low-order point flaw weakens key exchange

CVE-2025-8556
3.7
EPSS 0.5%
Data LeakageSupply ChainFrameworkTraining Data
rhelai1/instructlab-amd-rhel91 ATLAS
CRITICALEXPLOIT AVAIL

ChatGLM-Webui: arbitrary file read, no auth required

CVE-2025-45150
9.8
EPSS 0.6%
Data ExtractionAuth BypassFrameworkAPI
langchain-chatglm-webui3.2K5 ATLAS
CRITICALEXPLOIT AVAIL

BentoML: unauthenticated SSRF via file upload URLs

CVE-2025-54381
9.9
EPSS 13.2%
Supply ChainData ExtractionAuth BypassFrameworkInferenceAPI
bentomlCWE-918235 ATLAS
CRITICALEXPLOIT AVAIL

LangChain GmailToolkit: indirect prompt injection to RCE

CVE-2025-46059
9.8
EPSS 0.8%
Prompt InjectionCode ExecutionData ExtractionFrameworkAgentPlugin
6 ATLAS
CRITICALEXPLOIT AVAIL

smolagents: sandbox escape enables unauthenticated RCE

CVE-2025-5120
10.0
EPSS 31.2%
Code ExecutionSupply ChainData LeakageFrameworkAgent
smolagentsCWE-941115 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