Interpreter abort via mishandled quantization scale
Published May 20, 2022 · Updated Apr 22, 2025
Improper input validation in Google TensorFlow allows local users to abort a TFLite interpreter by loading a crafted quantized model. The quantized comparison kernel passes input scales above 1 to QuantizeMultiplierSmallerThanOneExp, whose TFLITE_CHECK_LT assertion terminates the process. Exploitation requires local execution or an application that accepts attacker-controlled TFLite models, and the demonstrated consequence is a process crash.
Summary
What happened
Improper input validation in Google TensorFlow allows local users to abort a TFLite interpreter by loading a crafted quantized model. The quantized comparison kernel passes input scales above 1 to QuantizeMultiplierSmallerThanOneExp, whose TFLITE_CHECK_LT assertion terminates the process. Exploitation requires local execution or an application that accepts attacker-controlled TFLite models, and the demonstrated consequence is a process crash.
The record
- CVE
- CVE-2022-29212
- Published
- May 20, 2022
- Updated
- Apr 22, 2025
- Vendor
- Product
- TensorFlow
- Classifications
- CWE-20, T1499
- Attack vector
- local
- Privileges
- authenticated
Timeline
How it unfolded
- May 20, 2022CVE publishedPublication date reported by the CVE source.
- Apr 22, 2025Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=< 2.6.4
- Affected versionversion=>= 2.7.0rc0, < 2.7.2
- Affected versionversion=>= 2.8.0rc0, < 2.8.1
- Affected versionversion=>= 2.9.0rc0, < 2.9.0
What conditions does exploitation require?
What is affected?
Published CVSS scores
CVSS describes severity. EPSS estimates exploitation probability.
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Public exploit references
- Core dumped when invoking TFLite model converted using nightly TFLite converterproof of concept · demonstrated
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo