Task Injection – Exploiting agency of autonomous AI agents

Unmasking Harmful Content in a Medical Chatbot: A Red Team Perspective

Revival Hijack – PyPI hijack technique exploited in the wild, puts 22K packages at risk

Microsoft Copilot: From Prompt Injection to Exfiltration of Personal Information

Google AI Studio: LLM-Powered Data Exfiltration Hits Again! Quickly Fixed.

From MLOps to MLOops: Exposing the Attack Surface of Machine Learning Platforms