Google Photos is getting a new feature that will allow people to use artificial intelligence (AI) to check if images are enhanced. Images edited in the Magic Editor using Reimagine will be marked with SynthID, a watermarking technology based on mountain view technology giants. The tag can be viewed in the About This Image section of the image, or it can be detected by a dedicated tool. This move may be aimed at reducing potential deep-strike instances that can be used to use Google’s AI image editing to generate.
Add AI watermark to Google Photos
The tech giant announced in a blog post that Google Photos will now automatically add AI watermarks to images that are enhanced with Magic Editor. This will mainly apply to the Reimagine tool, which will add new elements to the image using the Imagen 3 AI model.
However, other uses of the Magic Editor will also result in watermark addition. Google has developed a new technology called SynthID, which will add an invisible and imperceptible tag to the image.
Although the label can be seen in this image menu, it can also be detected by a special detection tool. When information is added to the pixels of an image, this watermark cannot be deleted by cropping, adding filters, or copying the image.
It is worth noting that SynthID is developed by Google DeepMind and can be added to AI-generated text, images, audio and video. In each case, the watermark is not added on the surface but is added within the medium structure. For example, in a video, add it to each frame of the video.
Move behind adding a watermark in an AI image may reduce the instances of people falling for deep strikes. DeepFakes is any type of realistic media that is digitized using AI or enhanced with other digital tools, with the aim of spreading misinformation or misleading people.
It is worth noting that Google clarifies that if the edits made by Reimagine are too small, the SynthID may not be added to the image. This will work for scenes, such as the user changing the color of the flower in the background. Users can check whether an AI watermark has been added by turning on metadata.
