Meta’s Oversight Board has ordered the company to remove two AI-generated videos from Facebook and called for stronger measures to address the spread of deepfake content.
One of the cases involves a Scottish politician who was falsely portrayed in an AI-generated video making inflammatory comments about refugees. The video showed the councillor appearing to say: “Refugees are welcome here, even if they r*** our women, because white people do that too.”
The Oversight Board determined that the video appeared to be synthetically generated, citing a lack of synchronization between the audio and the councillor’s facial movements. The councillor also described the use of their replicated voice as “quite traumatic.”
The Guardian reports that Meta had previously decided not to remove the video after it was reported.
According to the company, the post had not been flagged by any of its “trusted partner” organizations and did not appear to interfere with voting or other electoral processes. The video was also not given an AI label.
The board says Meta should have removed the video because it violated the company’s hateful conduct rules by attributing criminal and predatory behavior to refugees as a group. It also says the video should have been prominently labeled as AI-generated.
In its decision, the board describes Meta’s existing safeguards as “consistently and fundamentally inadequate” to address the rapid growth of AI-generated content on its platforms.
“The majority finds that Meta needs more robust policies on deepfakes, including expanding the situations when ‘high risk’ labels can be applied, more measures to reduce the spread of deceptive AI content, increasing the penalties for accounts that repeatedly share it, and more transparency on data around when AI labels are applied,” it says.
The board recommends nine changes to Meta’s policies. Among them are measures to ensure that algorithms reduce the distribution of content classified as high risk and make such material less prominent in users’ feeds. It also proposed measures such as warning screens that would require users to click through before viewing certain AI-generated content.
“From politicians to private citizens, AI-generated deepfakes are increasingly being used to harass and silence women from engaging in public discourse,” says Pamela San Martin, an Oversight Board co-chair.
“These cases demonstrate a broader, troubling pattern in which women who engage publicly on issues are disproportionately subjected to harassment and misinformation. Meta and other social media platforms need more robust policies to address the proliferation of deepfakes.”
The Oversight Board was established by Meta in 2020 to review difficult content-moderation decisions and make policy recommendations. Its individual content decisions are binding on the company, while its broader policy recommendations are not. Meta has 60 days to respond to the latest recommendations.
Image creditsHeader photo licensed via Depositphotos. Affiliate Disclosure PetaPixel articles may include affiliate links; we may earn a commission if you buy through one.