innethread

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!

Perhaps it could be my lack of understanding of the financial world, but are there (or why aren’t there any) banks that simply store your balance in cash? My understanding is that banks invest the deposits of their customers to make a profit. However, why can’t they just charge a fee to sustain the bank’s operations while being a completely safe bank that wont fail if everyone decides to withdraw their cash? Wouldn’t that be an appealing model to the market?