What does SynthID Detector actually check?

It checks whether images, video, or audio were made by AI that has adopted Google DeepMind's SynthID watermark, including Google, OpenAI, NVIDIA, and Kakao services. Content from AI without SynthID is out of scope.

Is there a limit on how many times I can use it?

Yes, the terms of service set a daily limit on how many checks you can run, though Google has not disclosed the specific number.

AI Safety & AlignmentITmedia AI＋3m agoOct 8, 2026, 16:00 JST

Google opens SynthID Detector to all, covers OpenAI and NVIDIA AI

Send on LINE

3 Key Points

What happened
Google publicly launched SynthID Detector on October 7, a website where you upload images, video, or audio to check if they came from AI made by Google, OpenAI, NVIDIA, or Kakao; Apple support is coming soon.
Why it matters
The tool now checks other companies' AI output, not just Google's own, and Google warns it is not a general-purpose AI detection tool.

WHO IT HITSJournalists, researchers, and content moderators who need to verify whether an image, video, or audio file was AI-generated can now use the tool without waiting for waitlist approval, though it only works on content from AI that has adopted SynthID.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, in your inbox every morning.

Get news like this every morning, free

Context & Analysis

SynthID Detector was first announced in May 2025 as an initial version, but access was limited to journalists and researchers who signed up to a waitlist. The public release removes that waitlist step, opening the tool to anyone who wants to check a file.

The tool checks for SynthID, an invisible watermark that Google DeepMind developed to embed in AI-generated images, video, and audio. Until now, watermark checks were available in the Gemini app, but only for content made with Google's AI. SynthID Detector expands that to other companies' AI output.

The limits are worth noting: AI that has not adopted SynthID is out of scope, and even if no watermark is found, the content could still be AI-made. Google itself cautions on the site that the tool is not a general-purpose AI detection tool.

FAQ

What does SynthID Detector actually check?: It checks whether images, video, or audio were made by AI that has adopted Google DeepMind's SynthID watermark, including Google, OpenAI, NVIDIA, and Kakao services. Content from AI without SynthID is out of scope.
Is there a limit on how many times I can use it?: Yes, the terms of service set a daily limit on how many checks you can run, though Google has not disclosed the specific number.
What languages does the tool support?: English only.

ITmedia AI＋Read Original Article

Send on LINE

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Get it free

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.