AVID-2026-R0773

Description

Reference binding to null in ParameterizedTruncatedNormal (CVE-2021-29568)

Details

TensorFlow is an end-to-end open source platform for machine learning. An attacker can trigger undefined behavior by binding to null pointer in tf.raw_ops.ParameterizedTruncatedNormal. This is because the implementation(https://github.com/tensorflow/tensorflow/blob/3f6fe4dfef6f57e768260b48166c27d148f3015f/tensorflow/core/kernels/parameterized_truncated_normal_op.cc#L630) does not validate input arguments before accessing the first element of shape. If shape argument is empty, then shape_tensor.flat<T>() is an empty array. The fix will be included in TensorFlow 2.5.0. We will also cherrypick this commit on TensorFlow 2.4.2, TensorFlow 2.3.3, TensorFlow 2.2.3 and TensorFlow 2.1.4, as these are also affected and still in supported range.

Reason for inclusion in AVID: The candidate CVE-2021-29568 describes a software vulnerability in TensorFlow (an AI framework) where binding to an uninitialized pointer due to an empty shape can cause undefined behavior. This is a vulnerability in a component widely used to build/deploy AI systems, impacting the software supply chain of general-purpose AI stacks. The description includes affected versions, remediation, and references (NVD entry, GitHub advisory), providing sufficient signal for AVID curation.

References

• NVD entry
• https://github.com/tensorflow/tensorflow/security/advisories/GHSA-4p4p-www8-8fv9
• https://github.com/tensorflow/tensorflow/commit/5e52ef5a461570cfb68f3bdbbebfe972cb4e0fd8

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: 2021-05-14
• Version: 0.3.3
• AVID Entry