🚨 Fake OpenAI model on Hugging Face spreads malware

A malicious repository falsely claiming to be an OpenAI “Privacy Filter” model went viral on Hugging Face, hitting ~244K downloads in under 18 hours before being removed.

AI

0.0326

+13.98%

🔒 Key details:

657/667 likes flagged as bot-like activity
Deployed a multi-stage infostealer (Windows)
Stole browser passwords, Discord tokens, crypto wallet keys & SSH credentials

Data sent to attacker-controlled servers

A reminder: always verify model sources before downloading.
#AI #OpenAI