TL;DR
Get business pricing on monitors, keyboards and dev 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 globally in English. It can check images, video and audio for SynthID watermarks associated with AI systems from Google and named partners, but it cannot identify every AI-generated file.
Google has expanded access to SynthID Detector, making the tool available to people worldwide in English to check whether images, videos or audio contain watermarks associated with AI content from Google and participating partners. The change takes a service first offered in an early version to media professionals and makes it available to a broader public, though the detector does not establish whether all AI-generated content is present or absent.
Google says users can submit an image, video or audio file for the detector to check for SynthID watermarks. The company says the tool can identify media made with AI from Google or its partners, naming OpenAI, NVIDIA and Kakao; Google says Apple support is coming soon. The announcement does not specify when Apple support will be added.
Google introduced SynthID in 2023, describing it as an imperceptible watermarking technology for AI-generated images, video and audio. The company reports that it has watermarked more than 180 billion images and videos, as well as audio equivalent to 240,000 years. Those figures are Google’s own reported totals, not an independent measure of how much AI media is online.
The detector’s wider release joins verification features Google says are built into Search, the Gemini app and Chrome. Google reports those features handle more than one million requests daily. The announcement gives no breakdown of those requests by product, region or media type.
A Wider Check for Watermarked Media
Opening the detector beyond media professionals gives more people a way to investigate whether some online media carries an AI watermark. That may help users, journalists and other media workers add context when checking material, especially where the file was created using a participating system that applies SynthID.
The tool’s usefulness depends on the content carrying a watermark it can recognize. A result is not a universal AI-authenticity verdict: the announcement describes checks for SynthID on content from Google and partners, not a method for detecting every AI system, edited file or synthetic media item. Readers should treat the detector as one verification signal rather than proof that content is authentic or fabricated.
As an affiliate, we earn on qualifying purchases.
From Media Teams to Public Access
Google says it launched SynthID in 2023 to mark AI-generated media with watermarks designed to be imperceptible. It later introduced an early version of the detector for media professionals. The current announcement expands access globally, with the interface available in English; it does not say that the tool is available in other languages.
The detector is part of a broader set of verification features Google says it has added to Search, Gemini and Chrome. The company presents these services as ways to give people more information about media they encounter online. The source material does not detail how those separate features work or how their results compare with the standalone 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, including OpenAI, NVIDIA, Kakao, and soon, Apple.””
media verification tools for AI content
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Limits of the Detector’s Coverage
Google’s announcement does not explain how the detector handles media without a SynthID watermark, files modified after creation, or content produced by tools outside the named partner group. It also does not describe accuracy rates, known failure cases, file-size limits, processing times or how submitted media is handled and stored.
The company says Apple support is coming soon but provides no release date. Although access is described as global, the tool is available in English according to the announcement; availability in other languages and any regional restrictions are not specified. The reported watermark totals and daily request volume are company figures, and the supplied material offers no independent verification.
AI-generated media watermark detector
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Broader Access and Partner Support
Users can now access SynthID Detector globally in English, according to Google. The next stated expansion is support for Apple-generated content, though Google has not announced a date or described how that integration will work. The company has not outlined a schedule for additional languages or other partners.
As use grows, practical questions will include how clearly the tool communicates uncertainty and how well its checks work across different formats and media that have been edited. Google has not announced further technical details or an independent evaluation. For now, its stated role is narrower: checking submitted media for watermarks associated with participating AI systems.
AI content authenticity verification
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
Who can use SynthID Detector now?
Google says the tool is available to anyone worldwide in English. It was previously offered in an early version for media professionals.
What kinds of files can it check?
Google says it can check images, video and audio for SynthID watermarks.
Does it detect all AI-generated content?
No such broad capability is stated in the announcement. Google describes checking for watermarks associated with its own and participating partners’ AI systems, so an absence of a detected watermark should not be treated as proof that media was made by a person.
Which partners are named?
Google names OpenAI, NVIDIA and Kakao as partners whose AI content can be checked. It says Apple support is coming soon but gives no date.
How many items have been watermarked?
Google reports more than 180 billion images and videos and audio equivalent to 240,000 years. The announcement does not provide an independent audit or a time window for those totals.
Source: rss
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
