Style Living Self Celebrity Geeky News and Views
In the Paper BrandedUp Hello! Create with us Privacy Policy

Google launches watermarks for AI-generated images to fight misinformation

By Yoniel Acebuche Published Aug 31, 2023 6:30 pm

The use of artificial intelligence has been rampant in this fast-changing world. Images we see online, for one, are not all reliable because of AI.

To better identify images thar are real from those generated by computers, search engine company Google launched on Aug. 29 an invisible, permanent watermark that will identify photos that are AI-generated.

Google's new technology, called SynthID, generates a watermark that is unnoticeable to the human eye and, at the same time, cannot be removed or altered when the image is applied with common editing techniques, like cropping or applying filters.

The special watermarking tool has an attempt to find “the right balance between imperceptibility and robustness to image manipulations"to prevent the spread of misinformation. 

"We’re committed to connecting people with high-quality information, and upholding trust between creators and users across society. Part of this responsibility is giving users more advanced tools for identifying AI-generated images so their images —and even some edited versions—can be identified at a later date," said Google DeepMind in a blog post

The watermark is detectable even after modifications like adding filters, changing colours and brightness.

SynthID also allows Vertex AI customers to create AI-generated images responsibly and to identify them with confidence. Google admits that this technology "is not perfect" however believes that its accurate against many common image manipulation based on their trials and testings. 

SynthID's combined approach

SynthID has two combined approaches—one is called watermarking, where you can add an imperceptible watermark to synthetic images.‍ Identification, on the other hard, involves scanning an image for its digital watermark in order for SynthID to assess the likelihood of an image being created by Imagen.

This tool provides three confidence levels for interpreting the results of AI identification:  detected, not detected, and possibly detected.

SynthID Identification confidence levels chart.

Moving forward

Since SynthID is currently in beta, the team is gathering user input and considering ways to make improvements, "We hope our SynthID technology can work together with a broad range of solutions for creators and users across society, and we’re continuing to evolve SynthID by gathering feedback from users, enhancing its capabilities, and exploring new features."

They're also looking into integrating it with other Google products and even releasing it to third parties "in the near future empowering people and organizations to responsibly work with AI-generated content."