TL;DR
Get business pricing on networking and server gear
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
Google has expanded access to its SynthID Detector, making the tool available to anyone worldwide in English. The company says it can check images, video and audio for SynthID watermarks in content made with AI tools from Google and named partners; it does not establish that all AI-generated content can be identified.
Google has made its SynthID Detector available worldwide in English, opening a tool previously offered in an early version to media professionals to anyone who wants to check certain images, videos and audio files for AI watermarks. Google says the detector can identify SynthID-marked content made with AI from Google and participating partners, including OpenAI, NVIDIA and Kakao, with Apple support planned.
The detector looks for imperceptible SynthID watermarks embedded in supported media. Google says users can submit an image, video or audio file and receive an indication of whether the content carries a watermark. The company describes the tool as a way to add context to media, not as a universal test for whether something was generated by AI.
Google said it has watermarked more than 180 billion images and videos and the equivalent of 240,000 years of audio since SynthID launched in 2023. Those totals are figures supplied by Google; the announcement does not specify the measurement method or give an independent audit.
The company also said its verification features in Google Search, the Gemini app and Chrome handle more than 1 million requests daily. It did not break down that figure by product or explain whether the requests are all related to SynthID. Google presents the wider set of features as part of its effort to help people understand the origin of online media.
A Wider Check for Watermarked Media
Opening the detector to the public could make it easier for readers, journalists and other users to check whether supported media contains a marker associated with participating AI systems. That may provide a useful signal when people encounter images, clips or audio whose origin is uncertain, particularly as generative tools become more widely available.
The reach of the tool is also limited by what it can detect: content must carry a compatible watermark for SynthID Detector to identify it. A result that finds no watermark does not, by itself, prove that media was made by a person or rule out AI generation. Users will need to interpret results alongside other evidence about a file’s source and history.
As an affiliate, we earn on qualifying purchases.
From Media Professionals to Public Access
Google introduced SynthID in 2023 as a system for embedding watermarks in AI-generated images, video and audio. The company later offered an early version of its detector to media professionals to help them verify AI-generated material. The latest announcement expands who can access that detector, while its public availability is currently described as global and English-language.
Google says the detector covers content from its own AI tools and those of partners. The announcement names OpenAI, NVIDIA and Kakao, and says Apple support is coming soon. It does not identify specific products or file formats from each partner that are currently supported. The company also points to verification features in Search, Gemini and Chrome, but the announcement does not detail how those features compare with the new detector.
““With the new SynthID Detector, anyone can easily check if an image, video, or audio file was made with AI from Google or our partners.””
AI-generated media verification tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Limits of Watermark Detection
The announcement does not say whether the detector can identify AI-generated files that lack SynthID or another compatible watermark, or explain how it handles files that have been edited, compressed or re-recorded. It also provides no published accuracy rates, error rates or independent evaluation of the detector’s performance.
Google has not specified when support will expand beyond English, what “global” availability excludes, or when Apple compatibility will arrive. The company names partner organizations but does not list which of their products or content types are supported at launch. These details matter because a negative result should not be read as proof that a file is authentic or human-made.
As an affiliate, we earn on qualifying purchases.
Partner Support and Language Expansion
Google says Apple support is planned, but gives no release date. It also has not announced a timetable for additional languages or shared a detailed list of supported partner products. Further information on coverage and performance would help users understand when the detector can provide a meaningful signal and when it cannot.
For now, the confirmed development is broader public access to the English-language tool. The company’s claims about watermark totals and daily verification requests describe the scale it reports, while the detector’s practical reach will depend on watermark coverage and the specifics of each supported file.
media authenticity verification device
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
Who can use Google’s SynthID Detector?
Google says the detector is now available to everyone globally in English. The source does not provide a list of countries or explain any regional exclusions.
What kinds of files can the detector check?
Google says it can check images, video and audio for SynthID watermarks. Its announcement does not give detailed file-format or size requirements.
Which AI companies’ content does it cover?
Google names its own AI tools and partner systems from OpenAI, NVIDIA and Kakao. It says Apple support is coming soon but does not identify supported products or a launch date.
Does a result without a watermark mean the content is not AI-generated?
No. The announcement describes detection of compatible watermarks; it does not say the tool can identify every AI-generated file. A file without a detected watermark should not be treated as proof of human authorship.
Source: rss
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
