Meta wants AI content in political advertising to be labelled

This applies, for example, to cases in which the impression is falsely created that a real person has said or done something, Meta explained in a blog post.

AI can generate realistic images partly from simple descriptions alone. This makes it easier to create photos - and videos - with false information.

"Advertisers running these ads do not need to disclose when content is digitally created or altered in ways that are inconsequential or immaterial to the claim, assertion, or issue raised in the ad," Meta said, such as cropping an image or colour correction "unless such changes are consequential or material to the claim, assertion, or issue raised in the ad".

Online platforms such as Facebook and Instagram, Elon Musk's X, formerly Twitter, and TikTok recently came under scrutiny from the EU Commission following reports of falsified images of the Gaza war.

If advertisers do not label the use of AI in their images and videos, Meta will reject the ad, it said. In the event of repeated offences, the advertiser could be penalised.