Chatbots and the Alignment Problem

I think everybody who's worked with these systems has some concerns in the back of their mind about how it could all go wrong. One of the things that we have seen is lots of examples where chatbots in particular are like leading people astray. Or you know sometimes code is the right thing and it will just write some code for you and explain it to you if that's what's needed. And yeah I feel like we're just scratching the surface of like we've done a lot of looking at are there specific ways that LLM's can change existing processes to make them better. Things like co-pilot and all that stuff. But I think we haven't really looked at like