Hello OpenAI Team,
First, thank you for all the work you’ve put into ChatGPT and image generation. I’m a horror artist and enjoy creating original monster and creature designs inspired by the unsettling atmosphere of analog horror and cryptid art, similar in style to Trevor Henderson’s fictional creatures. My goal isn’t to create content that glorifies real violence or harms anyone—I’m simply passionate about designing frightening fictional monsters for horror fans and other artists.
Sometimes my prompts are automatically blocked with the message: “We’re so sorry, but the prompt may violate our guardrails around violence.” This can happen even when I’m only describing an original horror creature without graphic violence.
I hope you might consider refining the moderation system so it can better distinguish between fictional horror creature design and graphic violent content. It would be helpful if artists could create unsettling, creepy, nightmare-inspired monsters with disturbing anatomy, eerie expressions, and dark atmospheres, while still preventing truly graphic or harmful imagery.
A more nuanced approach would allow horror creators to express their creativity without encouraging real-world violence or unsafe content. Thank you for considering this feedback, and I appreciate all the effort you put into improving these tools for creative communities.