When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works.

Google announces SynthID Detector that identifies AI-generated content

GOogle SynthID Detector Portal AI-generated content detector

At the Google I/O 2025 developer conference, Google announced updates to the SynthID tool, which watermarks AI-generated images. The search giant introduced a new verification portal called SynthID Detector to quickly and efficiently identify AI-generated content made with Google AI.

"Advances in generative AI are making it possible for people to create content in entirely new ways — from text to high quality audio, images and videos. As these capabilities advance and become more broadly available, questions of authenticity, context and verification emerge," the company said.

SynthID Detector can spot AI-generated content across different modalities and highlight which parts of the content are most likely watermarked using SynthID. Google introduced its AI watermaking tool in 2023 and later updated it to support AI-generated text and video. The watermarking technique was added to Magic Editor in Google Photos earlier this year.

SynthID has been used to watermark over 10 billion content pieces; it supports Gemini, Imagen, Lyria and Veo models. Google claims that SynthID can preserve the content's quality and "acts as a robust watermark that remains detectable even when the content is shared or undergoes a range of transformations." It says that the watermark is detectable even after tweaks like adding filters, changing colours, and brightness.

Google SynthID Detector portal for AI watermarking

When using the SynthID detector, users can upload content types such as an image, audio track, video, or piece of text created using Google's AI tools. The portal will scan it for watermarks and present the results, highlighting the affected parts. It will pinpoint specific segments where a SynthID watermark is detected when dealing with audio content.

SynthID detector is still in the development phase and rolling out to early testers. It will be available to more users in the future, Google said, adding that interested researchers and media professionals can sign up for the waitlist.

Google's offering comes almost a month after Adobe launched its Content Authenticity watermarking tool in public beta. The search giant is working to build an ecosystem around SynthID. It has already open-sourced the watermarking tool and partnered with NVIDIA to watermark videos generated by its AI models.

Report a problem with article
Google AI Ultra
Next Article

Google announces AI Ultra plan that costs $249.99/month

Google AI Overviews reaches 15 billion users
Previous Article

Google AI Overviews reaches 1.5 billion monthly users, expanded to 200 countries

Join the conversation!

Login or Sign Up to read and post a comment.

2 Comments - Add comment