News for hackers. Stuff that matters.
Re:New Prompt Injection Papers: Agents Rule of Two and the A... by simonw on Monday November 03, 2025 @10:57AM (#45800439)
Yeah, there remains a very real problem where a prompt injection against a system without external communication / ability to trigger harmful tools can still influence the model's output in a way that misleads the human operator.

There are no comments on this comment yet.