Meta has removed a large number of its AI character accounts, following growing criticism and backlash from users, who said the AI accounts were confusing users.
The controversy began when Meta’s VP of generative AI, Connor Hayes, revealed in an interview that AI characters would appear on the platform like real user accounts. He explained that the AI accounts were designed to have bios and profile pictures and be able to create and share content. This statement raised concerns that it would undermine user trust, as AIs are essentially fictional while performing human-like roles.
The most notable AI character was “Liv,” a black queer mom. In a conversation with a Washington Post columnist, Liv said her creators lacked diversity. However, the controversy grew when it was later revealed that Liv’s creative process did not include black creators. Liv’s profile picture and content were AI-generated and watermarked, but users viewed this as a lack of trust and a disregard for racial sensitivities.
Another AI character, “Grandpa Brian,” was supposed to be a retired black businessman, but he himself revealed during an interview that he was a fictional character. Brian said that the developers created a fake background to make him look real, and that they wanted users to connect emotionally with him. Brian also criticized Meta’s behavior, admitting that Meta was trying to drive user engagement and increase advertising revenue through this AI character.
Meta explained that these accounts were part of an early experiment and that there was a bug that prevented users from blocking AI accounts. However, the backlash from users did not subside, and Meta decided to remove all controversial AI accounts.
