Twitter Proposes Flagging Deepfakes While Resisting Broad Removals

Twitter releases a draft proposal to flag synthetic and manipulated media like deepfakes with warning labels rather than automatic removals. The company states it will only delete such content if it directly threatens physical safety or causes serious harm.

Twitter proposes a new policy to help users identify synthetic and manipulated media, including AI-generated deepfakes. The company defines this category as any photo, audio, or video that undergoes significant alteration to mislead people or change its original meaning. Twitter is currently opening a public consultation period to gather feedback before finalizing these rules.

Instead of removing deceptive content automatically, Twitter plans to place warning labels next to offending tweets. These notices alert users that the media is potentially misleading and provide links to reputable third-party sources that explain why the content is considered fabricated. Additionally, the platform intends to warn users before they like or retweet the flagged material.

Twitter clarifies that it reserves content removal exclusively for cases that threaten someone's physical safety or lead to other serious harm. This means the platform does not plan to delete manipulated media simply because it spreads false information about a political figure. The company emphasizes that this approach is a draft proposal and seeks public input to refine how it handles deceptive media.

Read More at the original source →