Product/Service: Google VEO / Imagen (via Vertex AI / Google Cloud)Issue Type: Overaggressive Safety Filter / Sudden Regression

I am suddenly unable to generate videos due to a persistent error message stating: “This prompt might violate our policies about generating prominent people. Please try a different prompt or send feedback.”

This is a clear false positive. I am not using, naming, or referencing any celebrity, public figure, or real prominent person.

Key Details:

  • Asset Origin: The character images used are strictly [choose one: fully AI-generated / my own original, custom images]. They do not depict or mimic any real human being.
  • Previous Success: I have successfully generated videos using this exact same character asset and prompt structure multiple times in the past without any issues.
  • Current Behavior: The error now triggers consistently every time I try to run the generation, completely breaking my established workflow.

Steps Taken:
I have tried running the generation multiple times and have submitted feedback via the in-app tool, but the loop persists.

Could a community manager or engineer look into this filter calibration? It seems a recent backend filter update is incorrectly flagging previously approved, entirely synthetic character assets.

Hello @VASANTH_JANY,

For issues related to the Vertex API, please contact your sales representative if you have one. Otherwise, you can find more relevant assistance on the Google Cloud Forum.