Gotta love it that they made an AI that supposedly can talk about everything except the growing list of things that are "inapprotiate".Correct. It won't generate any NSFW responses. Even with the API it can't be manipulated to create NSFW responses and that is built into the model itself. People outside OpenAI don't have the ability to change that. Closest people have come to it is DAN, tricking the AI into roleplaying to give politically incorrect responses but even then it's very hit and miss with its responses. People need to know while ChatGPT will have many applications, it can't be applied to everything.
Whats the point of exploring the possibilities if you cut them down because someone might get offended about something other people have typed in?