TL;DR
Get the latest gadgets delivered free — and shop member deals
- Fast, free delivery on millions of items
- Access to Prime Big Deal Days deals on October 6–7
- Prime Video, Amazon Music and more included
Google has made its SynthID Detector available globally in English, expanding access from an earlier version intended for media professionals. The tool checks whether images, video or audio carry SynthID watermarks associated with AI tools from Google and participating partners; it does not establish that every AI-generated file can be identified.
Google has made its SynthID Detector available globally in English, widening access to a tool that checks images, video and audio for watermarks associated with AI-generated media. The company says the detector can identify content made with AI from Google or participating partners, giving users another way to assess the origin of some media they encounter online.
The detector was previously offered in an early version for media professionals, Google said. The new release expands access to everyone and supports checks of images, video and audio. Google describes the system as looking for SynthID, an imperceptible watermark embedded in supported media.
Google lists OpenAI, NVIDIA and Kakao among the partners whose AI-made content can be checked, and says Apple will be included soon. The company does not specify in its announcement which products or file formats from each partner are covered, or when Apple support will arrive.
Google said it has watermarked more than 180 billion images and videos, as well as audio equivalent to 240,000 years. Those figures describe the amount of content the company says it has watermarked; they are not a measure of detector accuracy or the share of AI content online that the tool can identify.
A Wider Route to Media Checks
Making the detector publicly available gives people beyond newsrooms and other media organizations a way to check whether supported files carry an AI watermark. That may help readers, viewers and listeners add provenance information to their assessment of a piece of media, particularly when its origin is uncertain.
The value of the tool depends on what it can recognize. A positive result can point to a SynthID watermark, but the announcement does not say that the detector can identify all AI-generated content. Nor does a watermark alone establish whether a file is accurate, misleading, edited, or presented with the right context. The detector is best understood as one source of information, rather than a general-purpose verdict on authenticity.
Google says the detector joins verification features in Search, the Gemini app and Chrome. It reports that those built-in features handle more than one million requests daily. That is a company-reported request volume, not an independent measure of how often users successfully verify media.
As an affiliate, we earn on qualifying purchases.
SynthID and Earlier Verification Tools
Google introduced SynthID in 2023 as a way to mark AI-generated media with watermarks designed to be imperceptible to people. The detector is intended to find those marks later, offering information about a file’s possible AI origin without relying only on a visible label or the uploader’s description.
The earlier detector version was aimed at media professionals. The latest announcement changes who can access it: Google says the service is now open to everyone globally in English. The company frames that release as part of a broader effort to provide context about online media through tools in its own products and through partner participation.
The approach has a defined scope. It relies on detecting SynthID in content that has been marked with the system, rather than assessing every image, recording or video on the internet. The partner list indicates that the tool is intended to cover some content produced beyond Google’s own AI systems, but the announcement does not offer a full inventory of supported services.
““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, including OpenAI, NVIDIA, Kakao, and soon, Apple.””
watermark detector for images and videos
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Coverage and Detection Limits
Google’s announcement does not provide independent test results, accuracy rates, or details on how the detector performs when media is compressed, edited or otherwise altered. It also does not say whether a negative result means a file was not AI-generated, or simply that the detector did not find a supported watermark.
The precise range of partner content, supported file types, usage limits and access requirements is not specified in the source announcement. Google says Apple support is coming, but gives no launch date. It is also unclear how the company will update users about the detector’s coverage as partners and supported systems change.
AI-generated media verification tool
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Partner Support and Public Use
The detector is available globally in English, according to Google. The next stated addition is Apple support, which the company says is coming but has not dated. Users and media organizations can begin testing the tool’s practical usefulness, though Google has not announced a public timetable for publishing accuracy data or a detailed list of supported partner products.
For now, Google presents SynthID Detector as one element in a broader verification effort, alongside features in Search, Gemini and Chrome. The company’s announcement does not describe a separate independent review of those tools. Further details about coverage, performance and partner availability will determine how much the public can rely on the detector when checking specific media.
SynthID watermark detection device
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What does Google’s SynthID Detector do?
It checks images, video and audio for SynthID watermarks associated with supported AI-generated media. Google says the detector can check content made with AI from Google and participating partners.
Who can use the detector?
Google says the tool is now available to everyone globally in English. The company previously offered an early version for media professionals.
Does a negative result prove that content was made by a person?
No such conclusion is established in Google’s announcement. The detector looks for supported watermarks, and Google has not said that it identifies every form of AI-generated media.
Which partners are included?
Google names OpenAI, NVIDIA and Kakao as partners whose AI-made content can be checked. It says Apple support is planned, without giving a date or specifying the full set of supported products.
How accurate is SynthID Detector?
Google’s announcement does not provide accuracy rates or independent evaluation results. It also does not explain how detection performs after media has been edited or compressed.
Source: rss
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
