YouTube will now remove AI deepfakes from you if you ask
Have you recently seen an AI-generated version of yourself in a YouTube video? Don’t worry: You can now ask the platform to remove it.
That’s because YouTube has quietly rolled out a new policy addressing concerns about AI-generated impersonation of humans, allowing affected individuals to request its removal through YouTube’s privacy request process. It marks the latest in ongoing efforts by YouTube and other social media platforms to crack down on synthetic media and AI content shared by users of other humans.
The rise of AI-generated content has presented significant challenges for platforms like YouTube. Synthetic media, including deepfakes, can create highly realistic but misleading representations of individuals, raising concerns about privacy and potential abuse. In March, YouTube introduced a tool in Creator Studio that allows creators to flag when their content was created using synthetic media. The platform is also testing a crowdsourced notes feature that will highlight whether a video featuring AI content is misleading or simply a parody.
“If someone has used AI to alter or create synthetic content that looks or sounds like you, you can ask for it to be removed,” the new policy reads is reading“To be eligible for removal, the content must depict a realistic, altered, or synthetic version of your likeness.”
The updated policy allows individuals to file complaints about AI-generated content that they believe violates their privacy. Previously, YouTube would simply label such content as AI-generated or potentially misleading.
Privacy issues
Now, the person depicted by AI tools or their lawyer can file a privacy complaint and have it removed, at least in most cases. Content creators have two days to remove the likeness or the entire video after the complaint comes in. If they do nothing, YouTube will review the complaint and decide whether it is valid.
However, there is no guarantee of removal. The decision depends on factors such as whether the video acknowledges the origins of the AI, uniquely identifies a person, was created as a parody or for satirical reasons, or involves a public figure who is engaged in criminal activities or supports someone or something.
Privacy complaints are separate from receiving Community Guidelines strikes. Privacy complaints do not automatically result in a strike against the user, although repeated privacy violations may result in the user being banned.