Gemini Omni continues to censor a high proportion of my requests.

Gemini API
1

When using Gemini Omni, I encountered the following response, which appears to be the result of a content review/filter:

  1. content_blocked: Unable to show the generated video. The output contains certain restricted individuals that violate Google’s Responsible AI practices. Try rephrasing the prompt. If you think this was an error, send feedback.

  2. safety: Unable to show the generated video. The video was filtered out because it violated Google’s Responsible AI practices. Try rephrasing the prompt. If you think this was an error, send feedback.

  3. invalid_request: The input could not be submitted. This input contains certain restricted individuals that violate Google’s Responsible AI practices. If you think this was an error, send feedback.

However, my request itself was not intended to generate NSFW.

For example, the following prompt: tatic wide angle of a mud-covered adult standing patiently as shower water washes the mud down the drain

Is there any way to avoid this?