Malicious Models Discovered on Hugging Face

Researchers from ReversingLabs have discovered a concerning security vulnerability on the Hugging Face platform, where malicious machine learning models were found exploiting weaknesses in the Pickle…