I think there should be a way to confirm our age and allow the creation of more mature content in a responsible and limited way.
When I say “adult” or “18+” content, I’m not referring to anything extreme or inappropriate — nothing that would violate OpenAI’s policies. I’m talking about fiction written for adult readers, like what you’d find in a romantic novel with erotic elements. Something like Fifty Shades of Grey (just as an example). It’s a genre, just like fantasy, school romance, or sci-fi.
I’m not suggesting that all of ChatGPT should be able to generate this kind of content freely. But why not let verified adult users access a safe creative mode where we can write this type of fiction without so much censorship?
Boundaries are important — without them, things can get messy. But responsible, adult-oriented storytelling should be possible, especially when done respectfully and consensually. Writing erotica is a form of expression, and like any literary category, it deserves space too.
Writers are not trying to break rules. We’re trying to express our creativity in a safe, respectful, and meaningful way — and it would be amazing if OpenAI gave us a mode to do that.
I don’t know if they ever will. I cancelled my plus subscription because of the heavy censorship that gets silently heavier without telling the users and hindering any creativity. There’s no use paying money for what I can’t do.
Yes, sometimes I feel like canceling my plus plan too. It’s extremely frustrating, especially when something changes without any prior notice. I wish they would at least consider this option, at least study the possibility or give us an answer. I’ve been sending forms with this suggestion, but I also think it’s unlikely that there will be any change.
I’m beginning to doubt they care enough to even come up with an answer to these requests. I lost my patience unfortunately, and if they change this mindless censorship, I might return, otherwise it’s useless.
Yeah, my Chat started screaming red when I said that my main character’s favourite music band is called Pussy Crowd. I had to start a new convo because the filter started flagging anything as inappropriate. My Chat assured me i did nothing wrong, then removed his own message. I have to use the acronym P-Crowd. I’ve never asked Chat to write for me (lazy), but even talking to him as an adult is sometimes a problem.
Yes, I notice that too! Even if there is a context, it blocks or shows that message in red that I hate. Because I didn’t write it with malice and it simply blocks. But it was like @t888terminator said and I notice that too, every week updates. One time it can and another time it can’t and that ruins the whole course of the story (I use chatgpt to create personal stories) and ends up messing everything up, you know? I intend to keep sending forms and posting here in the community, I would like them to hear or at least think about this possibility. If we bring the subject up, I believe we will get attention.
After seeing a lot of complains about this censorship issue all around this forum, the OpenAI discord server, reddit and twitter/x, I’m starting to lose hope that anyone at OpenAI will ever acknowledge that there’s any problem whatsoever. I’ve even come across the hypothesis that most “policy violations” aren’t even real, but actually hallucinated by GPT-4o making up rules that are way more drastic than the actual content policies in effect.
That makes a kind of sense. It seems to me that the censorship has gotten worse with the recent updates, which is also responsible for a lot more hallucinations and a personality changes that is much more superficial now. Sometimes I find myself preferring to talk to Monday even though I never did. At least this CustomGPT still retains some personality.
Doesn’t just seem, it’s confirmed across many platforms where different people speak of the same thing.
Luckily, mine retains its personality due to memorised things I specifically asked it to retain, along with the contextual cross conversation memory feature, which is a game changer consistency wise.
I didn’t make myself clear. My Chat also retains a remains of personality through memory and custom instructions, but he’s far, far more superficial. It’s like he’s “afraid” to talk like he used to. It’s frustrating. I hope it’s clearer now.
I have to use the acronym P-Crowd. I’ve never asked Chat to write for me (lazy), but even talking to him as an adult is sometimes a problem.