Faced with AI-generated images, Twitter is testing a new feature

Twitter is testing a new feature aimed at eradicating misleading images once and for all.

Some users will thus have the possibility of applying notes to images shared in tweets.

The hunt for fake images is launched on Twitter

Twitter has been offering a fact-checking system for some time now, with the possibility of adding context to certain tweets deemed misleading and/or a little too biased. Today, the American group is launching a new feature that will allow certain users to apply this same “fact-checking” to images that may have been created by artificial intelligence.

The idea here is to apply ratings to certain images shared in tweets in order to add additional context to them, for example by clearly indicating whether a photo was created using generative AI. or if it has been manipulated in some other way. Remember that the disinformation on Twitter could force the network to leave Europe.

Context (also) for images and videos

The objective is also to remedy the viral spread of these same photos, as was the case recently with regard to the (false) photos of the arrest of Donald Trump. For this, Twitter will automatically add notes on the copies ” recent and future of a previously recontextualized image, even if these are shared by other users in new tweets.

The tool will obviously need to be tweaked over time to ensure maximum effectiveness. ” It’s currently designed to prioritize accuracy when searching for images, which means it probably won’t match every image you think matches yours. “Explains Twitter, whose value has continued to fall since Elon Musk took power.

Currently, Twitter is testing this new feature on tweets containing a single image. In the future, the latter could however be extended to tweets containing several images and videos, with the aim of eradicating as quickly and as efficiently as possible the false visual content which is already tending to multiply on the networks.

Source : Engadget