Making Creator Headlines · Page 6
X Tests AI Labels for Posts That May Have Had an Efficient Assistant
· New York time
X is testing an automated way to detect and label posts that may have been generated using artificial intelligence. The experiment addresses a growing question for social platforms: how to give users context about material that can resemble ordinary human-made posts. A label could help readers consider how a post was produced as they decide what to make of it. For creators, it would introduce another signal attached to work appearing in a feed, where a short tag may influence attention before someone has examined the content itself. The word 'may' matters. An automated system assessing whether material was generated by AI is making a determination about a post, not providing a complete account of how it was created. The usefulness of any label depends on what it communicates and how people interpret it. The test reflects a familiar platform dilemma. The easier it becomes to produce and circulate material, the more appealing a quick indicator becomes. Yet a compact label has to carry a complicated distinction across a screen designed for rapid scrolling. The feed has spent years encouraging everyone to post more. It is now considering a small sign asking whether everyone was actually involved.
Read this story in the full newspaper →Satire. Not factual reporting. Read the original reporting from the linked publishers.