Edgar Cervantes / Android Authority
summary
- YouTube has expanded its privacy request process to include AI-generated content.
- If the content impersonates your face or voice, you can request its removal from YouTube.
- YouTube will manually review each request and issue a takedown order if the counterfeit content “may be mistaken for the real thing.”
Over the past year, we've seen how tools like Midjourney can create misleading images and sway public opinion. To address these concerns, YouTube is taking new steps to protect user privacy. The platform will now allow users to flag and request removal of AI-generated content that mimics their face or voice. This policy covers fully synthetic reproductions or partially altered content that could be mistaken for the real thing.
If you come across content where your voice or face has been disguised, YouTube has added a third option to its privacy complaint form to address this scenario. Previously, you could only report videos that included sensitive information, like your full name or address, without your consent.
Please note that YouTube carefully considers each request before taking action. Key factors are the level of realism and the potential for abuse or manipulation.
The announcement also states that YouTube will consider whether the reported content contains “parody or satire of public figures.” The platform may make exceptions for public figures who have already spoken publicly. To the extent that AI-generated content falls within the bounds of social commentary and free speech, this seems like a reasonable trade-off. But it will be interesting to see how YouTube balances that aspect with the reputational risk to individuals.
YouTube began enforcing the use of disclosure labels for AI-generated content in March. When the label is applied, viewers will see a small message at the bottom of the video that reads “Altered or synthetic content,” similar to a sponsorship disclosure. This only applies if the video contains a significant amount of artificially generated content, such as when an AI voice generator is used for narration.
The announcement is another step toward weeding out potential misuse of AI on YouTube, which could become increasingly prevalent as video generators get better. OpenAI's Sora video demos published earlier this year looked highly realistic, and the company's next GPT-5 model will likely support video as an additional modality, in addition to text, images, and audio. Google also announced a competing Veo video generator last month, which it plans to integrate into YouTube Shorts.
