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
2,371

AI/ML CVEs Tracked

344

Critical

449

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 1550 results — no patch
MEDIUM

TFLite: heap OOB read via negative indices in GatherNd

CVE-2021-37687
5.5
EPSS 0.2%
Data Extraction Supply Chain Framework Inference
tensorflow 3.7K 4 ATLAS
MEDIUM

TensorFlow Lite: OOB read leaks heap memory in expand_dims

CVE-2021-37685
5.5
EPSS 0.2%
Data Extraction Privacy Violation Framework Inference
tensorflow 3.7K 3 ATLAS
MEDIUM

TensorFlow TFLite: DoS via division by zero in pooling

CVE-2021-37684
5.5
EPSS 0.1%
DoS Framework Inference
tensorflow 3.7K 3 ATLAS
MEDIUM

TFLite: division by zero DoS in inference kernels

CVE-2021-37683
5.5
EPSS 0.2%
DoS Framework Inference
tensorflow CWE-369 3.7K 3 ATLAS
HIGH

TFLite: uninitialized quant params corrupt inference

CVE-2021-37682
7.1
EPSS 0.2%
Code Execution DoS Supply Chain Framework Inference
tensorflow 3.7K 3 ATLAS
HIGH

TensorFlow: heap over-read leaks memory via RaggedTensor

CVE-2021-37679
7.8
EPSS 0.2%
Data Leakage Code Execution Framework Training Data
tensorflow CWE-681 3.7K 3 ATLAS
HIGH EXPLOIT AVAIL

TensorFlow/Keras: RCE via YAML model deserialization

CVE-2021-37678
8.8
EPSS 0.5%
Code Execution Supply Chain Framework Model
tensorflow CWE-502 3.7K 4 ATLAS
MEDIUM

TensorFlow: DoS via invalid Dequantize axis argument

CVE-2021-37677
5.5
EPSS 0.1%
DoS Framework Inference
tensorflow CWE-1284 3.7K 3 ATLAS
MEDIUM

TensorFlow: DoS via MaxPoolGrad invalid tensor input

CVE-2021-37674
5.5
EPSS 0.2%
DoS Framework Training Data
tensorflow CWE-1284 3.7K 2 ATLAS
MEDIUM

TensorFlow: MapStage CHECK-fail causes process DoS

CVE-2021-37673
5.5
EPSS 0.2%
DoS Framework Training Data Inference
tensorflow 3.7K 3 ATLAS
MEDIUM

TensorFlow: heap OOB read in SdcaOptimizerV2

CVE-2021-37672
5.5
EPSS 0.2%
Data Extraction Privacy Violation Framework Training Data
tensorflow 3.7K 3 ATLAS
MEDIUM

TensorFlow: heap OOB read in sorting ops

CVE-2021-37670
5.5
EPSS 0.2%
Data Extraction Supply Chain Framework Training Data
tensorflow 3.7K 3 ATLAS
MEDIUM

TensorFlow: integer conversion DoS in NonMaxSuppression ops

CVE-2021-37669
5.5
EPSS 0.2%
DoS Framework Inference
tensorflow 3.7K 3 ATLAS
MEDIUM

TensorFlow: DoS via div-by-zero in UnravelIndex op

CVE-2021-37668
5.5
EPSS 0.2%
DoS Framework Inference
tensorflow CWE-369 3.7K 3 ATLAS
HIGH

TensorFlow MKL: null-ptr/heap-OOB in requantization ops

CVE-2021-37665
7.8
EPSS 0.2%
Code Execution DoS Framework Inference
tensorflow 3.7K 3 ATLAS
HIGH

TensorFlow: QuantizeV2 heap OOB/null-deref in quantization

CVE-2021-37663
7.8
EPSS 0.2%
Code Execution DoS Framework Training Data Inference
tensorflow 3.7K 3 ATLAS
MEDIUM

TensorFlow Lite: MLIR null ptr deref crashes inference

CVE-2021-37689
5.5
EPSS 0.2%
DoS Supply Chain Framework Inference
tensorflow 3.7K 4 ATLAS
MEDIUM

TensorFlow Lite: DoS via crafted TFLite model file

CVE-2021-37688
5.5
EPSS 0.2%
DoS Supply Chain Framework Inference Model
tensorflow 3.7K 3 ATLAS
MEDIUM

TFLite: infinite loop DoS via crafted strided slice model

CVE-2021-37686
5.5
EPSS 0.2%
DoS Supply Chain Framework Inference
tensorflow CWE-835 3.7K 4 ATLAS
HIGH

TensorFlow Lite: null ptr deref crashes SVDF inference

CVE-2021-37681
7.8
EPSS 0.2%
Code Execution DoS Framework Inference
tensorflow 3.7K 4 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