Looking at an internal tool that is trying, yet again, to do a #threatmodel by using an #LLM. I got 12 threat actors on a fairly large project. What I'm really interested in is how different these two threats are from each other, and why we have distinct mitigations that address one but not the […]