CVE-2021-37659: TensorFlow undefined behavior in binary cwise operations without broadcasting

Summary

TensorFlow's binary cwise operations that do not require broadcasting assume both inputs have the same number of elements but never check it. An attacker can exploit this to trigger heap out-of-bounds reads and undefined behavior from binding a reference to a null pointer, as tracked in CVE-2021-37659. Affected versions include those still in supported range, such as TensorFlow 2.5.1, 2.4.3 and 2.3.4.

Mitigation

Patched in GitHub commit 93f428fd1768df147171ed674fee1fc5ab8309ec. The fix will be included in TensorFlow 2.6.0, and the commit will be cherry-picked onto TensorFlow 2.5.1, 2.4.3 and 2.3.4.

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

Related items

• MediumCVE-2026-121936: easyVoice server-side request forgery through openaiBaseUrl in /api/v1/ttsSimilar attack · NVD/CVE Database
• MediumCVE-2026-118628: MLflow get-artifacts server-side request forgery information disclosureSimilar attack · NVD/CVE Database
• HighCVE-2026-118625: MLflow Tracking Server privilege escalation via get_endpointsSimilar attack · NVD/CVE Database
• InfoWhen trust becomes the payload: Inside the threats, etc quarterly reportSimilar attack · Elastic Security Labs
• MediumCritical Langflow flaws enable authentication bypass and RCESimilar attack · CSO Online