
Cloud Vulnerability DB
A community-led vulnerabilities database
TensorFlow through 2.5.0 contains a vulnerability in the tf.keras.utils.getfile function when used with extract=True parameter. This function allows attackers to overwrite arbitrary files via a crafted archive. The vulnerability was discovered and disclosed on June 29, 2021. However, this vulnerability is disputed as the vendor's position is that tf.keras.utils.getfile is not intended for untrusted archives (MITRE, NVD).
The vulnerability exists in the get_file function of TensorFlow's Keras utility module when the extract parameter is set to True. The function fails to properly validate archive contents before extraction, which could allow path traversal attacks through crafted archive files (RyotaK).
If exploited, this vulnerability could allow an attacker to overwrite arbitrary files on the system through a specially crafted archive file when processed by the tf.keras.utils.get_file function with extract=True parameter (RyotaK).
The recommended mitigation is to avoid using tf.keras.utils.get_file with extract=True parameter on untrusted archive files. Security-conscious users should verify archive contents before extraction (RyotaK).
Source: This report was generated using AI
Free Vulnerability Assessment
Evaluate your cloud security practices across 9 security domains to benchmark your risk level and identify gaps in your defenses.
Get a personalized demo
"Best User Experience I have ever seen, provides full visibility to cloud workloads."
"Wiz provides a single pane of glass to see what is going on in our cloud environments."
"We know that if Wiz identifies something as critical, it actually is."