TensorFlow Vulnerabilities

pipML Libraries

AI Threat Alert tracks 438 known vulnerabilities in TensorFlow, 18 rated critical — an AI/ML ml libraries in the pip ecosystem. Each CVE includes CVSS severity, EPSS exploit probability, patch status, and CISO-grade analysis.

67
Risk Score
438
Total CVEs
18
Critical
pip
Ecosystem
Aug 10, 2026
Last CVE
4%
Patch Rate
1372d
Avg Time to Patch
201,621 stars77,975 forks3,457 issues3,410 dependentsLast push Nov 15, 2026
View on GitHub
OpenSSF Scorecard7.7/10

Known Vulnerabilities(438 total, page 10 of 18)

SeverityCVE IDSummaryCVSSPublished
MEDIUMCVE-2021-41198TensorFlow: tf.tile integer overflow crashes ML process5.5Nov 5, 2021MEDIUMCVE-2021-41197TensorFlow: integer overflow in tensor dims causes DoS5.5Nov 5, 2021MEDIUMCVE-2021-41196TensorFlow: integer underflow crashes Keras pooling layers5.5Nov 5, 2021MEDIUMCVE-2021-41195TensorFlow: integer overflow in segment ops causes DoS5.5Nov 5, 2021MEDIUMCVE-2021-37690TensorFlow: use-after-free crashes training processes6.6Aug 13, 2021MEDIUMCVE-2021-37692TensorFlow: string tensor GC segfault causes process DoS5.5Aug 12, 2021MEDIUMCVE-2021-37691TensorFlow TFLite: DoS via crafted model in LSH kernel5.5Aug 12, 2021MEDIUMCVE-2021-37687TFLite: heap OOB read via negative indices in GatherNd5.5Aug 12, 2021MEDIUMCVE-2021-37685TensorFlow Lite: OOB read leaks heap memory in expand_dims5.5Aug 12, 2021MEDIUMCVE-2021-37684TensorFlow TFLite: DoS via division by zero in pooling5.5Aug 12, 2021MEDIUMCVE-2021-37683TFLite: division by zero DoS in inference kernels5.5Aug 12, 2021HIGHCVE-2021-37682TFLite: uninitialized quant params corrupt inference7.1Aug 12, 2021HIGHCVE-2021-37679TensorFlow: heap over-read leaks memory via RaggedTensor7.8Aug 12, 2021HIGHCVE-2021-37678TensorFlow/Keras: RCE via YAML model deserialization8.8Aug 12, 2021MEDIUMCVE-2021-37677TensorFlow: DoS via invalid Dequantize axis argument5.5Aug 12, 2021MEDIUMCVE-2021-37674TensorFlow: DoS via MaxPoolGrad invalid tensor input5.5Aug 12, 2021MEDIUMCVE-2021-37673TensorFlow: MapStage CHECK-fail causes process DoS5.5Aug 12, 2021MEDIUMCVE-2021-37672TensorFlow: heap OOB read in SdcaOptimizerV25.5Aug 12, 2021MEDIUMCVE-2021-37670TensorFlow: heap OOB read in sorting ops5.5Aug 12, 2021MEDIUMCVE-2021-37669TensorFlow: integer conversion DoS in NonMaxSuppression ops5.5Aug 12, 2021MEDIUMCVE-2021-37668TensorFlow: DoS via div-by-zero in UnravelIndex op5.5Aug 12, 2021HIGHCVE-2021-37665TensorFlow MKL: null-ptr/heap-OOB in requantization ops7.8Aug 12, 2021HIGHCVE-2021-37663TensorFlow: QuantizeV2 heap OOB/null-deref in quantization7.8Aug 12, 2021MEDIUMCVE-2021-37689TensorFlow Lite: MLIR null ptr deref crashes inference5.5Aug 12, 2021MEDIUMCVE-2021-37688TensorFlow Lite: DoS via crafted TFLite model file5.5Aug 12, 2021

Showing 226–250 of 438

Frequently asked questions

What is TensorFlow?

TensorFlow is an AI/ML ml libraries tracked by AI Threat Alert for security vulnerabilities in the pip ecosystem.

How many known vulnerabilities does TensorFlow have?

TensorFlow has 438 known CVEs, 18 of them critical, tracked from NVD and GitHub Advisory.

Which ecosystem is TensorFlow distributed in?

TensorFlow is distributed via the pip ecosystem and categorized as ml libraries.

Where does the TensorFlow vulnerability data come from?

Vulnerability data is sourced from NVD and GitHub Advisory, enriched with CVSS, EPSS, exploit signals, and patch status for each CVE.

How do I assess the risk of TensorFlow?

Review each CVE below — every entry shows CVSS severity, EPSS exploit probability, exploitation signals, and whether a patched version is available.

Monitor TensorFlow in your stack

Get instant alerts when new vulnerabilities affect TensorFlow. CISO analysis, ATLAS technique mappings, and compliance reports included.

Start Monitoring