Google Watermarked 100 Billion Files, Now Lets Rivals Use SynthID

Google says SynthID, its digital watermarking technology, has now marked more than 100 billion images and videos plus 60,000 years of audio since launching three years ago. The company announced an expansion of its content transparency tools across Search, Gemini, Chrome, Pixel and Cloud on Thursday, and said OpenAI, Kakao and ElevenLabs will start applying SynthID to their own AI-generated content.

The scale matters. Generative media has become cheap, fast and convincing, and platforms, insurers and fact-checkers lack reliable ways to distinguish synthetic files from camera-captured originals. Provenance infrastructure — watermarks embedded at generation, and capture credentials embedded at the camera — is the industry's main technical answer, and Google is now pushing it beyond its own products.

SynthID and C2PA Credentials spread across Google's stack

Google introduced SynthID three years ago. The technology embeds imperceptible signals into AI-generated content, and Google has since integrated it into its generative media models and products. Alongside the watermark, Google uses C2PA Content Credentials — the industry standard that records how media was created and modified, with or without AI — across a growing number of its generative media tools.

Pixel 10 was the first smartphone to provide Content Credentials for images in its native camera app. In the coming weeks, Google will extend the technology to video on Pixel 8, 9 and 10 phones. By applying credentials at the point of capture, Pixel documents when content came from a real camera sensor. Google frames this as equally important as flagging AI-made files: "In an era of generative media, we believe that identifying authentic, unedited content can be just as important as knowing when a file was made or edited using AI."

Verification moves into Search and Chrome

Google recently added SynthID verification for image, video and audio to the Gemini app. Users have already invoked it 50 million times globally, according to the company. Starting today, that verification capability expands to Search, and it will arrive in Chrome over the coming weeks.

Users can check an image's origin through Search features like Lens, AI Mode and Circle to Search, as well as Gemini in Chrome, by asking "Is this made with AI?" or "Is this AI generated?"

Google is also adding verification for C2PA Content Credentials, which lets users check whether content is an unaltered original from a camera, or whether it has been modified and by which tools. That feature rolls out in the Gemini app starting today and will reach Search and Chrome in the coming months. The company notes this builds on existing work, including labels on YouTube that identify AI-generated content and testing with trusted partners on Backstory, a project to make detection tools faster and more reliable.

Rivals adopt Google's watermark

Digital media travels across many platforms, so interoperability is the industry's central problem. Google's answer is to license its watermark to competitors. OpenAI, Kakao and ElevenLabs will bring SynthID technology to more of their AI-generated content, meaning more material across the web will soon carry Google's imperceptible watermarks.

The move builds on earlier steps: Google open-sourced its SynthID text watermarking technology and partnered with NVIDIA to watermark AI-generated video produced by NVIDIA's Cosmos world foundation models.

A detection API for businesses

For organizations that need to identify AI-generated media at scale, Google is launching an AI Content Detection API on Google Cloud's Gemini Enterprise Agent Platform. The API detects AI content made by both Google's models and other popular models, and Google positions it as a tool for businesses deciding how to evaluate and manage media on their own platforms.

Use cases span backend operations — sorting feeds, preventing insurance fraud — and user-facing functions such as fact-checking and labeling synthetic media. Google is launching with a group of trusted partners and says it will refine the API based on their feedback.

Standards work reaches Instagram

Google sits on the C2PA steering committee and continues to advocate for global provenance standards, arguing that transparency tools built into devices should work seamlessly across everyday platforms. A concrete example: Meta, a fellow C2PA Steering Committee member, will start labeling camera-captured media with Content Credentials on Instagram. Photos and videos shot natively on Pixel phones will soon be recognized and labeled as authentic when shared there.

Google says content transparency remains "a complex challenge" and pledges to keep developing the technology and setting "a high bar for the industry." The stated goal is to give users tools to determine the history of any content they encounter — a bar the company will be measured against as rivals' watermarked output accumulates across the web.