AI Security
2 articles
-
AI Model Supply Chain Security: Can You Trust a Downloaded Model?
Downloaded AI models are software from strangers. The risks in model files, containers and dependencies, and a checklist for vetting them.
Wei Fei
·
-
Prompt Injection Explained: How the Attack Works and How to Defend Against It
Prompt injection is text that makes an AI system follow an attacker's instructions. How direct and indirect attacks work, and the layered defences.
Wei Fei
·
Wei Fei
