Twitter launches Community Notes for images

In an effort to put more onus on crowdsourced moderation, Twitter has launched Community Notes for images in posts. The company is aiming to address scenarios of morphed images or AI-generated images across the platform where the photos are posted. In other words, the platform wants to flag content like “the Balenciaga Pope” that went viral a few weeks ago. The launch comes days after an AI-generated image about an attack on the Pentagon spread quickly as prominent accounts retweeted it.

Twitter said that notes written for an image will appear on “recent & future” posts containing it. At the moment, the feature is in the test phase and will work for tweets with a singular image.

The company said that only Community Notes contributors with an impact score — a score that measures the helpfulness of contributor’s notes — of 10 will be able to see an option to submit notes about an image rather than a tweet. This way, if other users tweet the same image but with different captions, the Community Notes will remain the same.

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

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

— Community Notes (@CommunityNotes) May 30, 2023

Tweets containing images with notes will have a label saying “This note is about the image and could be shown on other Tweets including the image.”

Image Credits: Twitter

The social network said it is working on expanding this feature to support videos and tweets with multiple media. It also noted that since this is an early version of Community Notes for media content, the matching algorithm might miss a few tweets that includes an image with notes.

It’s currently intended to err on the side of precision when matching images, which means it likely won’t match every image that looks like a match to you. We will work to tune this to expand coverage while avoiding erroneous matches.

— Community Notes (@CommunityNotes) May 30, 2023

The Pengatgon attack hoax spreading quickly was a combination of paid verified accounts and AI-generated images. Even though the original image and the tweet were deleted, the photo was still being circulated on the social network. This issue is also spreading to fake AI-generated ads misleading people into believing that prominent personalities like former footballer Ian Wright and chef Gordon Ramsey have passed away.

Welcome to the @elonmusk advertising nightmare. Friend sent me this tweet asking wtaf the @guardian url is doing on this weird AI generated ad suggesting (incorrectly) that Ian Wright is dead . The link goes to an old page on the Guardian site 1/ pic.twitter.com/ICHEoyO5l6

— emily bell (@emilybell) May 29, 2023

After Elon Musk’s takeover of Twitter, the company has heavily relied on crowdsourced moderation through Community Notes as it let go of staff working on trust and safety to cut costs.

Last year, Twitter rolled out Community Notes worldwide. Earlier this year, it started accepting contributions from various countries. In February, Twitter rolled out a feature to make people aware of the context related to a tweet — the social network started sending notifications when a post a user has retweeted or replied to get a note after the fact.

Twitter launches Community Notes for images by Ivan Mehta originally published on TechCrunch

from TechCrunch https://ift.tt/9eyvL0x