While Google creates tools to ensure people aren’t duped by images generated by, or manipulated using artificial intelligence, the company’s Gemini AI is proving a dab hand at getting around copyright.
Reports emerging over the weekend showed Gemini 2.0 Flash being used to remove watermarks from copyrighting images. The above image is an iStock image, which is credited to a photographer named JMroceck. So we apologise for using the doctored version of his image to illustrate the point.
The AI tool was shown removing images from the photographs, filling in the gaps and adding a subtle Gemini watermark of its own (via TechCrunch).
The examples shown by Reddit and X users show an exceptionally good effort at removing the watermarks, up there with existing tools. However, in some cases it appears as if they company has almost crafted an entirely new image rather than simply filling in the gaps left by removing the watermark.
Back in December Google revealed a new tool for the Google Photos app that would tell users when generated images are captured photograph, presented authentically.
In a blog post the company wrote: “To further improve transparency, we’re making it easier to see when AI edits have been used in Google Photos. Starting next week, Google Photos will note when a photo has been edited with Google AI right in the Photos app.” However, the feature stopped short of placing a watermark or something on the image to make it clear as day.
While this is annoying and potentially costly for photographers, you shouldn’t let that distract you from checking out the best camera phones.
Opinion
Unacceptable, but par for the course
Google needs to get a handle on this. There’s already enough for imaging professionals to worry about with AI generation tools robbing them of paying gigs at the touch of a button. Now Google is making it easy to remove protections on their existing protected work?
