HighVulnerability
CVE-2022-21727: Tensorflow integer overflow in shape inference for Dequantize
- Source
- NVD(opens in a new tab)
- Identifier
- CVE-2022-21727
- Published
- Record updated
- 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
- 0.7%
Summary
TensorFlow's shape inference for `Dequantize` contains an integer overflow flaw. The `axis` argument can be `-1` or any positive value up to the input's number of dimensions, but the upper bound is never checked, and the code computes `axis + 1`, so an attacker can trigger the overflow.
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.
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