TL;DR: Testing an AI agent is really three different jobs. Evaluation tells you if an answer was good. Observability tells you what actually happened. Red teaming tells you what could go wrong next. M
TL;DR: AI agents don't behave the same way every time, so an attack that fails once can still work later. Automated red teaming runs each attack many times, judges what the agent actually did instead
If you have built a chatbot, you know how to test it. You send it questions and check whether the answers are right. An AI agent is different. It does not just answer. It acts. It reads your email, ca
