There is no solution to prompt injection in systems where LLMs call the shots.
- There is no solution to prompt injection in systems where LLMs call the shots.
- LLMs seeing raw data and being asked to make load-bearing security decisions cannot be made safe, no matter how good the model gets.
- Even if the model is great, the trolley problem of having the model, not the user, be tricked, shifts the blame.
- No mechanistic system can handle all the open-ended inputs that LLMs can cover, and LLMs are fundamentally confusable.
- You need a new kind of approach that has mechanistic software at the core, and LLMs marbled inside in intentional, limited ways.