Twitter Proposes Warning Labels for Deepfakes Instead of Removal
Twitter reveals a new draft policy that warns users about synthetic media but refuses to remove most deepfakes. The platform only considers taking down manipulated content that poses a direct threat to physical safety.
Twitter unveils a draft policy that addresses the rising concern of deepfakes by placing warning labels on manipulated media rather than removing it. Under these proposed rules, the platform adds notices next to deceptive tweets, alerts users before they like or share such content, and provides links to news articles explaining the falsified media. Twitter states it only removes deepfakes if they threaten someone's physical safety or cause serious harm, and it invites public feedback on these plans until late November.
While many people worry that doctored videos subvert democracy by making politicians appear to say or do false things, the actual political deepfakes encountered so far either turn out to be regular edits or are created by awareness campaigns rather than malicious actors. The true primary threat of deepfake technology currently targets women, as a recent report reveals that 96% of deepfakes are non-consensual pornographic videos featuring famous actors and musicians. Because this explicit content already violates Twitter's existing rules, the platform removes it regardless of the new synthetic media policy.
Tech companies struggle to agree on a unified approach to manipulated videos, as seen when YouTube, Facebook, and Twitter all handled the infamous Nancy Pelosi video differently. Automated deepfake detection technology alone cannot solve this complex issue, meaning the industry needs broader social and legal solutions to manage the evolving threat effectively.