Home » Database

AVID-2026-R1197

Description

TensorFlow vulnerable to Heap Buffer Overflow in AvgPoolGrad (CVE-2023-25664)

Details

TensorFlow is an open source platform for machine learning. Prior to versions 2.12.0 and 2.11.1, there is a heap buffer overflow in TAvgPoolGrad. A fix is included in TensorFlow 2.12.0 and 2.11.1.

Reason for inclusion in AVID: CVE-2023-25664 describes a heap-based buffer overflow in TensorFlow’s AvgPoolGrad, a software vulnerability in a widely-used ML framework. Affects TensorFlow versions prior to 2.12.0 and 2.11.1; mitigated by updates. TensorFlow is a core AI software stack used to build/train/deploy AI systems, so this is a software supply-chain issue in AI tooling. The CVSS/description indicate a security-related vulnerability with potential denial-of-service via memory corruption. Sufficient public signal is provided (CVE entry, advisory, commit).

References

Affected or Relevant Artifacts

  • Developer: tensorflow
  • Deployer: tensorflow
  • Artifact Details:
TypeName
Systemtensorflow

Impact

AVID Taxonomy Categorization

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

CVSS

Version3.1
Vector StringCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
Base Score7.5
Base Severity🔴 High
Attack VectorNETWORK
Attack Complexity🟢 Low
Privileges RequiredNONE
User InteractionNONE
ScopeUNCHANGED
Confidentiality ImpactNONE
Integrity ImpactNONE
Availability Impact🔴 High

CWE

IDDescription
CWE-120CWE-120: Buffer Copy without Checking Size of Input (‘Classic Buffer Overflow’)
CWE-122CWE-122: Heap-based Buffer Overflow

Other information

  • Report Type: Advisory
  • Credits:
  • Date Reported: 2023-03-24
  • Version: 0.3.3
  • AVID Entry