I was asking ChatGPT to write some NSFW scenes about two consenting adult OCs who are about 30 years old, sharing some vanilla kinks. There's never been an issue before, but now it suddenly produced a super creepy age play scene where it used terms like "little child" and treated one character like a child. I didn't ask for anything like that. I gave it a thumbs down and regenerated the response, which was normal. I'm now worried about potential consequences. Can I get banned for something ChatGPT writes? Should I have given it a thumbs down, or could that get me flagged?
3 Answers
All language models have moments when they 'hallucinate,' meaning they might produce bizarre or inappropriate responses because they lack a true understanding of context. It's just how they're built, and while it's frustrating, it's not necessarily something you can control. As for reporting it, you can report harmful content, but it might not change much. Don't panic too much; these weird outputs happen.
You shouldn't worry about getting banned for this! If your previous prompts were clean and you didn't get a warning, you're likely safe. The AI can generate odd responses sometimes, and it's not your fault. It's good you gave it a thumbs down; that's how feedback works!
Yeah, it doesn’t seem fair to ban you for something the AI wrote. You haven’t asked for anything like that, so it shouldn't reflect on you. Just keep an eye on it in case it happens again, but I think you’re in the clear!
Thanks for the reassurance! I appreciate it.