More tools for prompt injection exploits
2025-08-22 01:40:09.594837+02 by Dan Lyke 0 comments
This means an adversary can hide instructions in regular text, but also have the LLM create responses containing text that his hidden to the user as Kai tweeted.
And to state the obvious, such hidden instructions can be on websites, pdf documents, databases, or even inside GPTs (yes, I already built one of these).