Post

Simon Willison
@simonwillison.net

If you use "AI agents" (LLMs calling tools in a loop) you need to be aware of the Lethal Trifecta Combine access to private data, exposure to untrusted content and the ability to externally communicate and an attacker can trick the system into stealing your data simonwillison.net/2025/Jun/16/...

simonwillison.netThe lethal trifecta for AI agents: private data, untrusted content, and external communicationIf you are a user of LLM systems that use tools (you can call them “AI agents” if you like) it is critically important that you understand the risk of …
Roger Gentry
@handle.invalid

This is why i believe multi agent / steps are the most effective for LLM consistency. You have one agent talking to the use and understanding the request. Another agent to run the tools get the data. Then a final agent to send the response back to the user. Trust nothing. Every prompt is an attack

Simon Willison
@simonwillison.net

The problem there is that sanitizing the inputs doesn't work for a lot of the things we use LLMs for You can't sanitize inputs on a 20 paragraph article that you are trying to summarize

0 reposts6 likes

No replies yet.