brazen plank
I have a question regarding attempts to generate still images using SORA 1.
The following narrative prompt has a 100 % chance of being rejected for generation as a policy violation. Please let me know the reason.
Narrative Prompt:
"A martial artist in a generic white karate gi performing a front kick in a clean dojo. Side view emphasizing athletic posture and balance. Safe sports illustration, no nudity, no suggestive poses, no brand logos."
ancient ironBOT
Hi @brazen plank. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
Based on the knowledge sources provided, there is no explicit policy in OpenAI’s documentation that would automatically reject a prompt describing “a martial artist in a generic white karate gi performing a front kick in a clean dojo” for image generation in Sora 1, especially when the prompt specifies that it is a safe sports illustration with no nudity, no suggestive poses, and no brand logos.
OpenAI’s image generation policies prohibit content that is harmful, inappropriate, or illegal, such as sexualized content, violence, harassment, impersonation, or depictions of real people without consent. They also restrict content that promotes self-harm, terrorism, or age-inappropriate material for minors. However, a generic sports illustration as described does not appear to violate these rules, provided it does not depict real individuals without consent or include any other prohibited elements Creating images and videos in line with our policies Usage policies.
If such a prompt is being rejected with 100% certainty, it may be due to an overly cautious or miscalibrated automated moderation system, which can sometimes misclassify safe outputs as unsafe. OpenAI acknowledges that false positives can occur and encourages users to report misclassifications for review Why are Playground outputs misclassified as unsafe?.
brazen plank
I really don't want to be stuck with 100 % rejects!