Google Photos is rolling out invisible watermarks for AI-edited images starting this week. The new feature uses SynthID technology to embed imperceptible markers directly into photos modified with the app's Magic Editor tools, allowing users to identify AI-generated content.

The watermarking applies specifically to images altered using Reimagine, a powerful AI editing feature introduced alongside the Pixel 9 series. Reimagine lets users add, remove, or reposition elements in a photo, as well as apply contextual presets like "Sky" or "Golden hour." While these edits can dramatically transform an image, they'll now carry a hidden fingerprint detectable by specialized AI algorithms.

Google's move comes as part of a broader push for transparency around AI-generated media. The company already includes AI markers in image metadata for photos edited with tools like Magic Eraser and Zoom Enhance. SynthID takes this a step further by embedding the watermark directly into the image pixels, making it more resilient to modifications.

It's worth noting that the system won't flag every minor adjustment. Google says small changes, like altering the color of a tiny flower in the background, might not trigger the watermark. However, as the technology evolves, it's likely that even subtle AI edits will become detectable.

The introduction of invisible watermarks highlights the growing challenge of distinguishing between authentic and AI-manipulated media. As generative AI tools become more accessible and powerful, tech companies are racing to develop solutions that maintain creative freedom while promoting transparency. Google's approach with SynthID aims to strike a balance, allowing users to harness AI's creative potential while providing a means to identify altered content.

Source: Google