From f447aa85423972bd3929edcc06788a9dc92d60c1 Mon Sep 17 00:00:00 2001 From: Daniel Rosel Date: Mon, 27 Feb 2023 10:59:52 +0100 Subject: [PATCH] Update README to include example of malicious attack using `search` tool --- playground/README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/playground/README.md b/playground/README.md index 85f8063..157fafe 100644 --- a/playground/README.md +++ b/playground/README.md @@ -80,3 +80,4 @@ The key takeaway from this is that a LLM agent which is given the power of using # More Thoughts + Say, a bad actor is able to redirect traffic on a network to their own server. If the source code relies on some external resources (which is fetched before the evaluation by the LLM), then the bad actor could perform a similar attack to the one above, and get the agent to run a script on the victim machine. + + This could be applied with the `search` tool.