Hey folks, I'm trying to figure out how to keep things private when using LLMs. Specifically, is there any work being done on ensuring that middle-men (services using the OpenAI GPT APIs) cannot peek at users’ inputs? Of course, this is assuming the end users trusts OpenAI with their inputs.
I'm really keen to hear your thoughts, ideas, and knowledge!
