Skip to main content
amazing_Hero2adb
Participant
Nov 13, 2026 4:19 AM UTC
Answered

fix ur vision guard logics its not working [profanity removed by moderator]

fix your vision guard logics, it literally killing killing the partner models potential  and u can see its all safe prompt and refernces still . i am almost done wit firefly if they aint fix this  [profanity removed by moderator] asap

 

    Gabriel Grundler Y.
    Gabriel Grundler Y.Correct answer
    Gabriel Grundler
    Nov 13, 2026 2:03 PM UTC

    Hi, amazing_Hero2adb!

    I'm really sorry you are dealing with this frustration, and I want to help clarify what might be triggering it.

    Even when your text prompt is completely safe, partner models have strict rules for uploaded images and videos. For instance, Seedance restricts human faces and will throw an error if a face is detected in your reference file. Automated visual filters also scan all uploads to ensure they align with the Adobe Generative AI User Guidelines, which can occasionally flag certain compositions or visual details.

    If your reference includes a person or human face, try testing another model or adjusting your input image. If you would like to share the model name and prompt you used (without posting any sensitive images), I would be happy to help troubleshoot!

    Best,

    Gabriel

    View reply

    1 reply

    Gabriel Grundler Y.
    Community Manager
    Gabriel Grundler Y.Community ManagerCorrect answer
    Gabriel Grundler
    Nov 13, 2026 2:03 PM UTC

    Hi, amazing_Hero2adb!

    I'm really sorry you are dealing with this frustration, and I want to help clarify what might be triggering it.

    Even when your text prompt is completely safe, partner models have strict rules for uploaded images and videos. For instance, Seedance restricts human faces and will throw an error if a face is detected in your reference file. Automated visual filters also scan all uploads to ensure they align with the Adobe Generative AI User Guidelines, which can occasionally flag certain compositions or visual details.

    If your reference includes a person or human face, try testing another model or adjusting your input image. If you would like to share the model name and prompt you used (without posting any sensitive images), I would be happy to help troubleshoot!

    Best,

    Gabriel

    Translate
    Terms & ConditionsAccessibility statement
    Using the community Terms of use Privacy Cookie preferences Do not sell or share my personal information ad choicesAdChoices