Google Photos could soon flag AI-generated images

Generative artificial intelligence can do wonders in creating high-quality images and even videos. An opportunity, but also a real threat for artists, knowing that their copyrighted works can also be used to train AI without their consent. The company Procreate, which offers a graphic design application of the same name, even goes so far as to say that generative artificial intelligence is “built on a foundation of flight”.

For the user, it is also becoming more and more difficult to differentiate an image created by artificial intelligence from an image made by humans. Google plans to develop its Google Photos application to remedy this, if its source code (APK teardown) is to be believed.

Google Photos should soon flag AI-generated images

Digging into the source code of Google Photos v7.3, Android Authority discovered that the app could soon determine whether an image is the product of artificial intelligence. A credit could appear as on images generated by Google Gemini, for example. The mention “compositeWithTrainedAlgorithmicMedia” suggests that Google Photos could also signal when an image has been corrected or improved using generative artificial intelligence.

It’s hard to know exactly what the “@id/ai_info” line refers to. Maybe Google Photos will be able to report which AI model generated contentbut there is nothing to confirm this at the moment. To consult all the information concerning a work, you will probably have to go to the image information.

All that remains is to wait for official confirmation from Google. To see if Apple will end up doing the same, in particular because of the upcoming arrival of Image Playground. This new feature discovered by Android Authority is reminiscent of the application recently unveiled by Adobe, called Content Authenticity. It will allow artists to place a digital signature on their works in order to protect their content.