AVID-2026-R0873

Description

Heap OOB in TensorFlow Lite (CVE-2021-37685)

Details

TensorFlow is an end-to-end open source platform for machine learning. In affected versions TFLite’s expand_dims.cc contains a vulnerability which allows reading one element outside of bounds of heap allocated data. If axis is a large negative value (e.g., -100000), then after the first if it would still be negative. The check following the if statement will pass and the for loop would read one element before the start of input_dims.data (when i = 0). We have patched the issue in GitHub commit d94ffe08a65400f898241c0374e9edc6fa8ed257. The fix will be included in TensorFlow 2.6.0. We will also cherrypick this commit on TensorFlow 2.5.1, TensorFlow 2.4.3, and TensorFlow 2.3.4, as these are also affected and still in supported range.

Reason for inclusion in AVID: CVE-2021-37685 describes a heap-based out-of-bounds read in TensorFlow Lite, a core ML framework used to deploy and run AI models. This is a software vulnerability in a component (TensorFlow Lite) that is part of the AI software stack, affecting AI models/frameworks and their deployment. It impacts a software artifact commonly used in AI pipelines, fitting the software supply chain focus for general-purpose AI systems. The vulnerability has CVE/CWE evidence and a commit patch, confirming its security/safety impact.

References

• NVD entry
• https://github.com/tensorflow/tensorflow/security/advisories/GHSA-c545-c4f9-rf6v
• https://github.com/tensorflow/tensorflow/commit/d94ffe08a65400f898241c0374e9edc6fa8ed257

Affected or Relevant Artifacts

• Developer: tensorflow
• Deployer: tensorflow
• Artifact Details:

Impact

AVID Taxonomy Categorization

• Risk domains: Security
• SEP subcategories: S0100: Software Vulnerability
• Lifecycle stages: L06: Deployment

CVSS

CWE

Other information

• Report Type: Advisory
• Credits:
• Date Reported: 2021-08-12
• Version: 0.3.3
• AVID Entry