Humans have been labeled the weakest link for ages. But what about AI?
In this session, Cato CTRL experts reveal how attackers have weaponized AI systems and agents for harm. They’ll walk through real attacks from the wild that show how adversaries use prompt injection and model manipulation to make AI tools perform actions they shouldn’t.
This session reframes how you should think about AI risk, and what happens when machines become the ones making mistakes.
Inside the session:
- Real world attacks from this year where both humans and AI were manipulated for harm
- The techniques attackers used to influence AI output and behavior
- How to combine humans + AI to make the strongest link