Insights Security & Privacy

Google tool checks if a photo or video is AI-made, but only some

7 Oct 2026, 06:00 1 min read

The short version

Google now lets anyone upload an image, video or audio file to check for a hidden AI watermark. It only recognises content from Google and a few partners, so a clear result does not prove a file is genuine.

What happened

Google's SynthID Detector is a website where you upload a photo, video or audio file. It looks for an invisible watermark that some AI tools add to what they create. Last year it was limited to media professionals. It is now open worldwide, in English, and you must sign in to use it.

It recognises content made with AI from Google, OpenAI, NVIDIA and Kakao. Apple is expected to join later this year. Similar checks are also built into Google Search, the Gemini app and Chrome.

Why it matters to you

A positive result is useful evidence that AI was involved in making or editing something. A negative result means little. Google says this is not a general AI detector, so material from other tools, or with no watermark, will not be flagged.

The tool also cannot tell AI-made media from AI-edited media. Treat it as one check among several, not a final verdict on a suspicious image or clip.

What to do

If a photo or video shared on WhatsApp or Facebook looks doubtful, upload it to the detector and note the result. Also check who first posted it and whether trusted news sources carry the same story.

Do not forward something just because the tool found nothing. If you are unsure, leave it unshared.

Sources

Researched and written by our automated news system.

Free website check

Is your own website in good shape?

Get a score out of 100 for speed, security, Google visibility and trust in about 30 seconds, with the three things most worth fixing. Free, and no email needed.

Check my website