Three critical zero-day vulnerabilities affecting PickleScan, a widely used tool for scanning Python pickle files and PyTorch models, have been uncovered by cybersecurity researchers. The flaws, all ...

Picklescan flaws allowed attackers to bypass scans and execute hidden code in malicious PyTorch models before the latest patch.

Researchers have concocted a new way of manipulating machine learning (ML) models by injecting malicious code into the process of serialization. The method focuses on the "pickling" process used to ...