Google has launched global access to the SynthID Detector, a service designed to identify content generated by artificial intelligence. The tool allows users to verify whether images, videos, or audio recordings were created using Google models or those of its partners. Previously, the system was available only to journalists, media professionals, and researchers.
The detector relies on invisible watermarks embedded by the SynthID technology, which was introduced in 2023. This system is currently utilized by models and tools including Nano Banana, Veo, Lyria, Gemini, Flow, ProducerAI, and Vids. Support for the standard has been announced by OpenAI, Nvidia, and Kakao, with Apple expected to join in the future.
The service supports various file formats, such as JPG, PNG, WEBP, MP4, MOV, MP3, and WAV. Currently, the verification tool is available worldwide in English. Google has also integrated SynthID checks directly into the Gemini platform and the Chrome browser to streamline the identification process for users.
According to company data, tools integrated with SynthID process over 1 million requests every day. To date, the technology has been used to label more than 180 billion images and videos. Additionally, the system has processed an amount of audio content equivalent to 240,000 years of recordings.
