MediumVulnerability

CVE-2021-37686: TensorFlow Lite strided slice infinite loop via crafted model

Source: NVD(opens in a new tab)
Identifier: CVE-2021-37686
Published: August 12, 2021
Record updated: October 9, 2026

Affected: tensorflow >= 2.6.0rc0, < 2.6.0rc2
tensorflow-cpu >= 2.6.0rc0, < 2.6.0rc2
tensorflow-gpu >= 2.6.0rc0, < 2.6.0rc2
Fixed in: 2.6.0rc2
released August 4, 2021, 9 days before disclosure
Known exploitation: Not listed in the CISA Known Exploited Vulnerabilities catalog at the last check.
EPSS: 0.2%

Summary

CVE-2021-37686 affects TensorFlow Lite's strided slice implementation, which contains a logic bug introduced with new support for ellipsis in axis definition. An attacker can craft a model where `ellipsis_end_idx` is smaller than `i`, so the inner loop never increments `i` and the `continue` statement skips the outer loop's preincrement, causing an infinite loop. TensorFlow 2.6.0 is the only affected version.

Mitigation

We have patched the issue in GitHub commit dfa22b348b70bb89d6d6ec0ff53973bacb4f4695.

Details

Scores and exploitation

CVSS score: 5.5(medium)
Patch available: Yes
EPSS: 0.2%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-835

Affected

Packages: tensorflow@>= 2.6.0rc0, < 2.6.0rc2 (fixed: 2.6.0rc2)
tensorflow-cpu@>= 2.6.0rc0, < 2.6.0rc2 (fixed: 2.6.0rc2)
tensorflow-gpu@>= 2.6.0rc0, < 2.6.0rc2 (fixed: 2.6.0rc2)
Vendors: Google

Classification

Attack type: Denial of service
Attack sophistication: Moderate
AI component targeted: Framework
Impact: 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-2021-37686

First tracked: February 15, 2026 at 08:39 PM ET, from NVD/CVE Database

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

Classifier confidence: 90% (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