Advisory

PYSEC-2026-1954

Also known as: BIT-tensorflow-2022-41911, CVE-2022-41911, GHSA-pf36-r9c6-h97j, PYSEC-2026-3222, PYSEC-2026-939

Summary

Invalid char to bool conversion when printing a tensor

Details

Impact
When printing a tensor, we get it's data as a const char* array (since that's the underlying storage) and then we typecast it to the element type. However, conversions from char to bool are undefined if the char is not 0 or 1, so sanitizers/fuzzers will crash.

Patches
We have patched the issue in GitHub commit 1be743703279782a357adbf9b77dcb994fe8b508.

The fix will be included in TensorFlow 2.11.0. We will also cherrypick this commit on TensorFlow 2.10.1, TensorFlow 2.9.3, and TensorFlow 2.8.4, as these are also affected and still in supported range.

For more information
Please consult our security guide for more information regarding the security model and how to contact us with issues and questions.

Attribution
This vulnerability was discovered via internal fuzzing.

Severity

Medium Basis: CVSS v3 score 4.8, vector CVSS:3.1/AV:N/AC:H/PR:L/UI:R/S:U/C:N/I:N/A:H

Dates

Published
2026-07-07 14:34 UTC
Modified
2026-07-13 16:43 UTC

Affected packages and fixes

tensorflow-gpu

PyPI · All systems

Affected

  • introduced 0, fixed 2.8.4, introduced 2.9.0, fixed 2.9.3, introduced 2.10.0, fixed 2.10.1 (ecosystem)

Fixed versions

2.8.4, 2.9.3, 2.10.1

Fix

  • Fixed version 2.8.4
    pip install --upgrade tensorflow-gpu==2.8.4
  • Fixed version 2.9.3
    pip install --upgrade tensorflow-gpu==2.9.3
  • Fixed version 2.10.1
    pip install --upgrade tensorflow-gpu==2.10.1

Fix commands are the export's own generic commands. Ojuju never runs them. Check them against your setup first.

Check your own computer: guard packages check

View this record on OSV.dev

Source

Source: PyPA Advisory Database · CC BY 4.0

Sources & licences