In a recent policy change, Google-owned YouTube allows individuals to request the takedown of AI-generated images of themselves. There are, of course, exceptions. From the Privacy Guidelines:

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. In order to qualify for removal, the content should depict a realistic altered or synthetic version of your likeness. We will consider a variety of factors when evaluating the complaint, such as:

…

● Whether the content is realistic

● Whether the content contains parody, satire or other public interest value

● Whether the content features a public figure or well-known individual engaging in a sensitive behavior such as criminal activity, violence, or endorsing a product or political candidate

It’s getting very easy to produce AI impersonations, From the Wall Street Journal last year:

As staff writer Sarah Perez notes at TechCrunch, Google is framing the problem as a privacy issue rather than one of misrepresentation: “Instead of requesting the content be taken down for being misleading, like a deepfake, YouTube wants the affected parties to request the content’s removal directly as a privacy violation.”

That probably makes sense because if they got into the question of misrepresentation as such on the internet, how would they ever get out of it?

YouTube has been working on means to address the problem for at least some months now:

The company didn’t broadly advertise the change in policy, though in March it introduced a tool in Creator Studio that allowed creators to disclose when realistic-looking content was made with altered or synthetic media, including generative AI. It also more recently began a test of a feature that would allow users to add crowdsourced notes that provide additional context on videos, like whether it’s meant to be a parody or if it’s misleading in some way.

YouTube is not against the use of AI, having already experimented with generative AI itself, including with a comments summarizer and conversational tool for asking questions about a video or getting recommendations. However, the company has previously warned that simply labeling AI content as such won’t necessarily protect it from removal, as it will still have to comply with YouTube’s Community Guidelines.

Sarah Perez, “YouTube now lets you request removal of AI-generated content that simulatesyour face or voice,” TechCrunch, July 1, 2024

A minimalist position

The video uploader seems to be trying to avoid accusations of complicity in bringing harm to an individual or violating privacy without harming its business model, which thrives on, among other things, sensation and controversy. At the same time, the technolopgy could become a political powderkeg:

At Futurism, Frank Landymore points out some of the details:

In keeping with its standards regarding any form of a privacy violation, YouTube says that it will only hear out first-party claims. Only in exceptional cases like the impersonated individual not having internet, being a minor, or being deceased will third-party claims be considered.

If the claim goes through, YouTube will give the offending uploader 48 hours to act on the complaint, which can involve trimming or blurring the video to remove the problematic content, or deleting the video entirely. If the uploader fails to act in time, their video will be subject to further review by the YouTube team.

Frank Landymore, “Youtube Now Lets You Request the Removal of AI Content That Impersonates You,” Futurism, July2, 2024

The court cases should be lively. For one thing, the Federal Trade Commission (FTC) is looking at new rules too.