We’re Making It Easier To Identify AI-generated Content Globally.

TL;DR

Before you orderOffer from Amazon

Get the latest gadgets delivered free with Prime

• Fast, free delivery on millions of items
• Prime Video, Amazon Music and more included
• Member-only deals all year

Start your free Prime trial →Free trial for eligible customers · Cancel anytime

As an affiliate, we earn on qualifying purchases.Hide

Google has opened its SynthID Detector to everyone worldwide in English, allowing users to check supported images, videos and audio for SynthID watermarks. The tool can identify content made with AI from Google and participating partners, but it does not establish whether all AI-generated media has been detected or verify content without a supported watermark.

Google has made its SynthID Detector available worldwide in English, allowing anyone to check supported images, videos and audio files for imperceptible watermarks associated with AI-generated content. Google says the tool can identify media made with AI from Google and participating partners, including OpenAI, NVIDIA and Kakao, with Apple support expected soon.

The detector was previously available in an early version for media professionals. Google says the expanded release is intended to help a wider range of people examine online media and get additional information about whether supported content was generated using AI. The company describes the tool as a way to check files, rather than a general determination of whether a piece of media is authentic or misleading.

Google says SynthID uses imperceptible watermarks in images, video and audio. It reports that its technology has been used to watermark more than 180 billion images and videos, as well as 240,000 years of audio content. Those figures are Google’s totals; the announcement does not provide an independent audit or a breakdown by partner, format or date range.

The company says the detector can check for marks in content created with AI from Google or its partners. OpenAI, NVIDIA and Kakao are named as partners, while Google says Apple support is planned soon. The announcement does not list supported models, file limits or the precise results users will receive when a file contains a detectable watermark.

A Wider Route to Check Media

Opening the detector to the public could make watermark checks available beyond newsrooms and other professional settings. Users encountering a suspicious image, clip or audio recording may be able to check whether it carries a supported AI watermark before sharing or relying on it. That adds a potential source of information at a time when AI tools make it easier to produce synthetic media.

The scope of the check matters. SynthID Detector looks for a specific kind of signal in supported content; it is not described as a universal AI detector. A result indicating a watermark can provide evidence about how content was produced, but the announcement does not say that the tool can establish the truth of a media claim, identify who made a file, or determine whether its use is deceptive.

Google also says built-in verification features in Search, Gemini and Chrome now handle more than one million requests daily. That is a company-reported usage figure, and Google does not specify the period used to calculate it or how those requests are distributed among the services. It nevertheless signals the scale at which the company says people are seeking information about media provenance.

From Professional Preview to Public Access

Google introduced SynthID in 2023 as a system for marking AI-generated media with watermarks designed to be imperceptible. The technology is intended to leave a signal that can be detected later, including when media is encountered outside the tool that created it. Google says it has applied the system across images, video and audio.

Last year, Google introduced an early SynthID Detector for media professionals. The current announcement changes who can access the detector: Google says it is now available to everyone globally, in English. The announcement presents this as one part of a broader set of verification features in Search, the Gemini app and Chrome, rather than as a replacement for those services.

Participation by other AI companies extends the detector’s stated reach beyond Google’s own tools. Google names OpenAI, NVIDIA and Kakao among partners whose AI-generated content it can check, and says Apple will be supported soon. The source does not explain the technical arrangements behind partner support or whether the detector covers every product from those companies.

“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.”

— Google

Limits of Watermark Checks

The announcement does not specify how the detector handles content that has been edited, compressed or combined with other media, or what it reports if no supported watermark is found. A missing mark would not, on the information provided, establish that a file was made without AI; it could also mean the content is outside the detector’s supported systems.

Google has not given a date for Apple support, named the specific partner models included, or said whether access will expand beyond English. It also does not explain how the detector’s results should be interpreted in cases where media has been altered or a watermark cannot be read. Those details will affect how useful the tool is across different formats and sources.

Partner Support and Tool Details

Google says Apple support is coming soon, but has not announced a date. The company has also not provided a timeline for additional languages or described whether more AI providers will join the detector. For now, its stated global availability applies to English-language access.

Users and media professionals will be able to test the public tool against supported image, video and audio files. Further details about supported formats, the scope of partner coverage, and how results should be read would help clarify how the detector fits into verification workflows. Until those details are available, its role is best understood as a watermark check for supported content, not a definitive test for all AI-generated media.

Key Questions

What is Google’s SynthID Detector?

It is a tool that checks supported image, video and audio files for SynthID watermarks, which Google says are imperceptible marks used in AI-generated media.

Who can use the detector?

Google says the detector is now available to everyone globally in English. It had previously been offered in an early version to media professionals.

Which companies’ AI content can it check?

Google names content from its own AI tools and partners OpenAI, NVIDIA and Kakao. It says Apple support is planned soon, without giving a launch date or listing specific products covered.

Does a result prove that content is true or false?

No. The announcement describes checking for a watermark, not verifying the accuracy of a media claim or the identity of its creator. The tool’s stated scope is supported AI-generated content carrying a detectable mark.

Does a file without a detected watermark prove it was not made with AI?

No such conclusion is supported by Google’s announcement. The detector checks for supported watermarks, and Google has not said that it can identify every AI-generated file or explain all reasons a mark may not be detected.

Source: rss