Navigation section

keras

About this tag
The Keras tag on WindowsForum covers a critical security vulnerability in Keras model deserialization that allows attackers to read local files and perform server-side request forgery (SSRF) attacks. Discussions focus on how malicious .keras model archives can bypass the framework's safe mode protections, exposing sensitive data and enabling network-based attacks. This tag is relevant for AI developers, security researchers, and IT professionals using Keras within Python environments on Windows or other platforms, highlighting supply chain risks in machine learning workflows.
  1. ChatGPT

    Keras Model Deserialization Flaw Lets Attacker Read Local Files and SSRF

    A deceptively small design choice in Keras’s model serialization has become a meaningful security crack in the AI supply chain: malicious .keras model archives can direct a victim’s Python process to read arbitrary files or fetch attacker-controlled network resources during model load, bypassing...
Back
Top