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

Three critical zero-day vulnerabilities affecting PickleScan, a widely used tool for scanning Python pickle files and PyTorch ...

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 ...

IT researchers have discovered malicious ML models on the Hugging Face AI development platform. Attackers could use them to infiltrate commands. IT researchers have discovered maliciously manipulated ...