fictional image

Question

I am writing to express my frustration with the repeated and unnecessary blocking of harmless, fictional image requests in Copilot. I have followed every guideline, removed every prop, avoided any illusion context, and simplified my requests exactly as instructed — yet the system continues to block basic, safe images such as a fictional character simply standing in a rehearsal room.

This is not a dangerous request. This is not inappropriate content. This is not a safety issue. This is a system failure.

The safety filters are clearly misfiring and incorrectly flagging harmless creative content. This is severely limiting Copilot’s usefulness and actively damaging the creative experience. An AI designed to support creativity should not be shutting down simple, fictional, non‑harmful image generation.

I expect this issue to be investigated and corrected. The current behaviour is unacceptable, and it prevents Copilot from functioning as advertised. Please escalate this problem to the appropriate team and provide a response explaining why these blocks are happening and what will be done to fix them.

Answer 1

Hello phil pentecost

Thank you for taking the time to share this feedback, and we sincerely apologize for the frustrating experience you've had with image generation in Copilot.

I understand how disappointing it can be when requests that appear harmless, fictional, and compliant with the provided guidelines are repeatedly blocked. In particular, when you have already made efforts to simplify prompts, remove potentially ambiguous elements, and follow recommended practices, you would reasonably expect those requests to be processed successfully.

Copilot's security and safety systems are designed to help prevent the generation of content that may violate policy or create safety concerns. However, these systems are always perfect and may, in some cases, incorrectly classify guideline-compliance requests and produce what are commonly referred to as false. When this occurs, legitimate creative scenarios may be blocked even though they do not contain unsafe or prohibited content.

We understand your concern that this behavior can limit Copilot's usefulness for creative work and negatively impact user experience. Feedback such as yours is valuable because it helps identify situations and improve the safety guideline. Since this is a user-to-user forum and us Moderator do have any access nor authorization to escalate your feedback directly to Copilot team, I would suggest you submit your feedback via Microsoft Feedback Portal . Once submitted, I believed your feedback will reach the correct team, and it will surely contribute to future updates.

Thank you for your understanding.