Social Media Platforms Warn of Increased AI Moderation Mistakes Amid Pandemic
Major social media companies rely heavily on automated content moderation as human reviewers work from home during the coronavirus pandemic. This shift leads to slower response times and a higher rate of erroneous content takedowns.
Major social media platforms warn that they make more content moderation mistakes as they rely entirely on artificial intelligence during the coronavirus pandemic. Companies like Facebook, YouTube, and Twitter send their human content reviewers home and temporarily shift to automated takedown software to enforce community guidelines.
This automated software identifies offensive material by analyzing digital clues, but it lacks the accuracy of human judgment. As a result, these platforms erroneously remove more videos and posts for policy violations, and users experience slower turnaround times when they submit appeals.
Each platform handles the transition differently, with Twitter explicitly refusing to ban users based solely on automated enforcement due to accuracy concerns. Facebook acknowledges the limitations of its automated tools and warns that the public sees longer response times and increased errors while human reviewers remain out of the office.