Twitter: Community will be able to report fake AI-generated images

With the frequent circulation of fake and artificial intelligence (AI) images that contribute to the spread of fake news, Twitter has decided to expand its Community Notes functionality to include tweets containing media files. The news was announced by microblog on Tuesday (30).

The feature that currently allows adding context to text broadcasts can now also be used to: flag posts with suspicious images. Initially, the tool will be limited to posts containing only one photo, but will support posts containing multiple images as well as GIFs and videos in the future.

According to Elon Musk’s social network, flagged tweets will be displayed with the hashtag: “This note is about and may appear on other tweets, including the image”. The platform also explained that at this initial moment, the algorithm may show some inaccuracies when comparing similar environments.

From AI-generated images to manipulated videos, it’s common to come across misleading media. Today we are piloting a feature that gives contributors a superpower: Notes on Media

Notes attached to an image will automatically appear on recent and future matching images. pic.twitter.com/89mxYU2Kir

— Community Notes (@CommunityNotes) 30 May 2023

First, Community notes for images on Twitter can only be created by contributors with a writing impact score of 10 or higher.. When using this feature, people will have the opportunity to choose whether the hashtag will be applied to the tweet in question or to the image it contains and to all posts in which it appears.

latest examples

Two AI-generated images that have gone viral on Twitter recently, making many netizens doubt their veracity, could certainly garner community ratings for the photos. One is seen wearing a Pope Francis Balenciaga jacket, and the other is allegedly a terrorist attack on the Pentagon.

According to the microblock, The new collaborative moderation feature is currently being tested, meaning it’s not yet visible to all users.. Also, the hashtag may appear on “recent and future” tweets if posts are flagged by the system.