I’m finding the censorship on AI to be a HUGE negative for LLMs in general, since in my mind they’re basically an iteration of search engines. Imagine trying to just search for a basic term or for some kind of information and being told that that information is restricted. And not just for illegal things, but just historical facts or information about public figures. I guess I understand them censoring the image generation just because of how that could be abused, but the text censorship makes it useless in a large number of cases. It even tries to make you feel bad for some relatively innocuous prompts.