Out-of-bounds read in TensorFlow - CVE-2022-21728
Published: February 2, 2022 / Updated: May 3, 2026
Vulnerability identifier: #VU129178
CSH Severity: Medium
CVSSv4.0: CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:L/VI:N/VA:N/SC:N/SI:N/SA:N/E:U/U:Green
CVE-ID: CVE-2022-21728
CWE-ID: CWE-125
Exploitation vector: Remote access
Exploit availability:
No public exploit available
Vulnerable software:
TensorFlow
TensorFlow
Software vendor:
TensorFlow
TensorFlow
Description
The vulnerability allows a remote attacker to disclose sensitive information.
The vulnerability exists due to an out-of-bounds read in the shape inference for ReverseSequence when processing a ReverseSequence operation with a negative batch_dim value. A remote attacker can supply a specially crafted ReverseSequence operation to disclose sensitive information.
The issue occurs because negative batch_dim values are not fully validated, and excessively negative values can cause access before the start of the dimensions array.
Remediation
Install security update from vendor's website.