NEW YORK – Meta Platforms Inc.’s independent Oversight Board issued a ruling on Monday determining that a doctored video suggesting President Biden is a pedophile does not violate the company’s current policies on manipulated media. However, the board deemed Meta’s rules on such content “incoherent” and too narrowly focused on AI-generated media.
The board took on the case of the Biden video in October 2022 after a user complaint about a manipulated 7-second clip posted on Facebook. The video showed real footage of Biden giving his granddaughter “I Voted” stickers during the 2022 midterms but was edited to make it appear Biden was acting inappropriately.
This first ruling addressing Meta’s manipulated media policy highlighted significant gaps that fail to protect users from misleading content. The board suggested urgent updates were needed to make the rules comprehensible and applicable beyond a limited scope of AI-generated deepfakes.
Flaws in Meta’s Policies Enable Spread of Misinformation
Meta informed the board during the review that it was already planning to revise the manipulated media policy to address new AI technologies that can generate increasingly realistic fake videos and audio.
However, the board emphasized non-AI altered content remains highly prevalent and potentially misleading without current safeguards. Meta’s focus exclusively on AI-manipulated media lacks clear justification and causes confusion for users.
The oversight board said Meta made the correct decision in keeping the Biden video online under the existing narrow policy. But it advised significant expansions of the rules to match the scale of threats posed by advancing technologies.
Recommendations for Improved Protections Against Manipulated Content
To offer more coherent protections, the board told Meta to broaden the policy to cover both video and audio media, regardless of the editing techniques used. It said users face equal risks of being misled by AI and non-AI manipulated content.
The board further recommended labels be applied to identify manipulated media instead of just removing posts. However, it stopped short of urging rules against doctored photographs due to enforcement challenges with visual media at Meta’s vast scale.
The oversight board gave Meta 60 days to publicly respond to its ruling and suggestions for shoring up the policy’s shortcomings. With pivotal elections approaching in 2024, pressure is rising for social platforms to combat new forms of political disinformation enabled by advancing generative AI.
Monday’s ruling highlighted that even without cutting-edge algorithms, basic editing tools remain potent weapons for spreading false narratives if not properly checked. Meta and its oversight board now face urgent decisions on how to update protections to match the pace of technological progress.
I enjoy your website, obviously, but you should check the spelling on a number of your posts. A number of them have numerous spelling errors, which makes it difficult for me to tell the truth, but I will definitely return.
Its like you read my mind You appear to know so much about this like you wrote the book in it or something I think that you can do with a few pics to drive the message home a little bit but other than that this is fantastic blog A great read Ill certainly be back