YouTube expands privacy protections for AI-generated content

YouTube now allows users to request the removal of AI-generated or altered content mimicking their face or voice, considering realism, identifiability, and parody in its decisions.
YouTube expands privacy protections for AI-generated content

YouTube has announced an expansion of its privacy request process, allowing users to seek the removal of AI-generated or synthetic content that simulates their face or voice. This new measure builds upon the platform’s approach to responsible AI innovation, which was initially outlined in November 2023.

In their earlier announcement, the platform emphasized the need to balance the creative potential of generative AI with the responsibility to protect the YouTube community. They highlighted introducing disclosure requirements for altered or synthetic content and new labels to inform viewers about AI-generated material.

Now, YouTube is taking further steps to address privacy concerns related to AI-generated content. Users can use the platform’s privacy request process to report content miming their likeness or voice without permission. This move comes as AI-generated content becomes increasingly prevalent on the platform.

When evaluating these requests, YouTube will consider several factors before deciding on removal. These include whether the content could be mistaken for real, if the person making the request is identifiable, and whether the content is intended as parody or satire, particularly when it involves public figures or well-known individuals.

The announcement has also noted that privacy violations are treated separately from Community Guidelines strikes. Creators who receive a privacy complaint should be aware that it will not automatically result in a strike against their channel.

YouTube also says that users can report channels they believe are impersonating themselves or other creators as part of their ongoing efforts to maintain platform integrity.

The updated Privacy Complaint Process page consists of a six-step process for users to file privacy complaints regarding content that violates their privacy or safety. The platform encourages users to first contact the uploader directly, as the creator may be unaware of the discomfort caused by their content. If this approach is unsuccessful, users are advised to flag the video for review against YouTube’s Community Guidelines.

Before submitting a formal privacy complaint, users will need to confirm that they are uniquely identifiable in the content and acknowledge that abusing the privacy process may result in account suspension.

The final step allows users to specify the nature of their complaint, whether it involves the unauthorized sharing of personal images or full names, other sensitive information, or AI-generated content that alters or synthesizes their voice or image without permission.

Posted by Alex Ivanovs

Alex is the lead editor at Stack Diary and covers stories on tech, artificial intelligence, security, privacy and web development. He previously worked as a lead contributor for Huffington Post for their Code column.