Content Credentials generative AI training and usage preference

Last updated on Sep 2, 2025

Learn how to set the generative AI training and usage preference to prevent supported generative AI models from using or training on your content.

Adobe supports giving creators control over how their content is used. By setting this preference in Adobe Content Authenticity (Beta) or other supported apps, you can indicate that you don't want generative AI to train on your content or use it to create new content, such as Style Reference or Structure Reference

This preference is embedded within each file through Content Credentials. Currently, this preference is supported by Adobe Firefly.

Generative AI training and usage preference interface in Adobe Content Authenticity (Beta), showing the option: “I request that generative AI models not train on or use my content,” with a note about the requirement for a verified name or connected social media account.
Interface for setting generative AI training and usage preference in Adobe Content Authenticity (Beta)

Note

You must include a verified name or at least one connected social media account in your Content Credentials to select this preference. 

Impact on Adobe Firefly and Stock workflows

If this preference is enabled, your work can't be uploaded to Adobe Stock, as Firefly is trained only on licensed content, such as Adobe Stock assets. 

Setting this preference prevents your work from being used in Adobe Firefly tools like Style Reference, Structure Reference, or Generate Similar. These tools use generative inference technology to influence outputs, even though they don’t train on your content. 

Dialog box showing a message that the uploaded image has Content Credentials preventing it from being used for generative AI training or creation. Options include “Cancel” and “Upload another image.”
Image upload blocked due to generative AI restrictions in Content Credentials