Is TensorFlow Keras “Safe Mode” Actually Safe? Bypassing safe_mode Mitigation to Achieve Arbitrary Code Execution
March 12, 2025
Update: This issue was discovered and disclosed independently to Keras by JFrog’s research team and Peng Zhou. Machine learning frameworks often rely on serialization and deserialization mechanisms to store and load models. However, improper code isolation and executable components in the models can lead to severe security risks. The structure of the Keras v3 ML Model …
