HighVulnerability

CVE-2022-21728: Tensorflow heap out-of-bounds read in ReverseSequence shape inference

Source: NVD(opens in a new tab)
Identifier: CVE-2022-21728
Published: February 3, 2022
Record updated: October 9, 2026

Affected: tensorflow < 2.5.3, fixed in 2.5.3, released February 2, 2022, 1 day before disclosure
tensorflow >= 2.6.0, < 2.6.3, fixed in 2.6.3, released February 2, 2022, 1 day before disclosure
tensorflow = 2.7.0, fixed in 2.7.1, released February 2, 2022, 1 day before disclosure
and 6 more
Known exploitation: Not listed in the CISA Known Exploited Vulnerabilities catalog at the last check.
EPSS: 1.1%

Summary

TensorFlow's shape inference for `ReverseSequence` does not validate negative values of `batch_dim`, allowing a heap out-of-bounds read. When the value is too negative, the `Dim` implementation accesses elements before the start of an array.

Mitigation

The fix will be included in TensorFlow 2.8.0. The commit will also be cherry-picked to TensorFlow 2.7.1, TensorFlow 2.6.3, and TensorFlow 2.5.3.

Details

Scores and exploitation

CVSS score: 8.1(high)
Patch available: Yes
EPSS: 1.1%Estimated probability that this CVE is exploited in the next 30 days, from the FIRST Exploit Prediction Scoring System. Read on November 4, 2026.

Weaknesses and techniques

Weakness (CWE): CWE-125
CWE-125
Attack pattern (CAPEC): CAPEC-540

Affected

Packages: tensorflow@< 2.5.3 (fixed: 2.5.3)
tensorflow@>= 2.6.0, < 2.6.3 (fixed: 2.6.3)
tensorflow@= 2.7.0 (fixed: 2.7.1)
tensorflow-cpu@< 2.5.3 (fixed: 2.5.3)
tensorflow-cpu@>= 2.6.0, < 2.6.3 (fixed: 2.6.3)
tensorflow-cpu@= 2.7.0 (fixed: 2.7.1)
tensorflow-gpu@< 2.5.3 (fixed: 2.5.3)
tensorflow-gpu@>= 2.6.0, < 2.6.3 (fixed: 2.6.3)
tensorflow-gpu@= 2.7.0 (fixed: 2.7.1)
Vendors: Google

Classification

Attack type: Other
Attack sophistication: Moderate
AI component targeted: Framework
Impact: Confidentiality,
Availability
Labels: security

How this record was produced

Summaries, mitigations and labels on this page are generated automatically from the source. The linked source is authoritative. The source gives this advisory a description and no title, so the title above is a headline generated from the source text.

Original source: https://nvd.nist.gov/vuln/detail/CVE-2022-21728

First tracked: February 15, 2026 at 08:40 PM ET, from NVD/CVE Database

Classifier: claude-haiku-5-5, prompt v4

Classifier confidence: 95% (self-assessed)

Affected packages in the Exposure Registry

Matched by package name and ecosystem. Each entry shows whether the package delegates to a language model and how many tracked packages depend on it.

tensorflowPyPINo LLM dependency detected · 26 tracked dependents
tensorflow-gpuPyPINo LLM dependency detected · 2 tracked dependents

Summary

Mitigation

Affected packages in the Exposure Registry

Related items