Google opened its SynthID verification website to everyone, allowing checks for watermarks in supported image, video, and audio files.
Google opened its SynthID verification website to the public on October 6, 2026, expanding access beyond the journalists, media professionals, and researchers who had tested the tool after its introduction at Google I/O the previous year. The site lets users submit supported image, video, and audio files to check whether they contain SynthID marks associated with AI-generated media.
Supported image formats include JPG, JPEG, PNG, BMP, WEBP, AVIF, HEIC, HEIF, TIFF, TIF, and GIF. The video formats are MP4, MOV, and WEBM; supported audio files include WAV, MP3, OGG, FLAC, AAC, and M4A. Google introduced SynthID in 2023, and uses it to watermark media generated by products and models including Nano Banana, Veo, Lyria, Gemini, Flow, ProducerAI, and Vids.
The website adds a public route to verification that Google had already integrated into the Gemini app and Chrome. The company says people make one million requests to check content each day. OpenAI, Nvidia, and Kakao also support SynthID, while OpenAI operates its own content-checking site. TechCrunch reported that Apple is expected to add support, but did not give a timing.
The launch gives people a way to check for SynthID in a range of media files, but it is not a universal test for whether something was made with AI: the check depends on the media carrying a recognizable watermark. The article also notes that Microsoft and Meta have their own watermarking and verification approaches, and that such systems can fail to identify content made by their creators’ models. The public site therefore expands access to Google’s watermark checks without establishing that every AI-generated image, video, or recording can be identified.