Facebook cracks down on deepfakes
MANILA, Philippines – Facebook announced on Monday, January 6 (January 7, Manila time) it was banning manipulated photographs and videos, or media classified as deepfakes.
In a blog post, Monika Bickert, Facebook vice president of global policy management, said the social network will "remove misleading manipulated media" based on two primary criteria.
First, the company will remove media that has been "edited or synthesized – beyond adjustments for clarity or quality – in ways that aren’t apparent to an average person, and would likely mislead someone into thinking that a subject of the video said words that they did not actually say."
Second, it will also remove media that "is the product of artificial intelligence or machine learning that merges, replaces or superimposes content onto a video, making it appear to be authentic."
Facebook said parody, satire or video that's been edited to omit or change the order of words will not fall under the policy, but media that doesn't fall under the ban on deepfakes can still be vetted by fact checkers or removed for other community standards violations.
In the Philippines, the most convoluted manipulated video that Rappler fact checked was one that was published on Facebook in August 2016. The video was spliced to appear as if Senator Leila de Lima admitted that she was a drug coddler, but the original video was actually of her criticizing the extrajudicial killings due to President Rodrigo Duterte's war on drugs.
The video had 2.8 million views and over 88,000 shares before Facebook flagged it for fact checking. The page that shared it is now unavailable; its last online activity was recorded on September 4, 2019. (READ: PREMEDITATED MURDER: The character assassination of Leila de Lima)
In its report, The Verge added the ban comes ahead of a House Energy and Commerce hearing on manipulated media set for January 8. Bickert is scheduled as Facebook's representative to the hearing. – Rappler.com