AI/News
Google opens its SynthID Detector to everyone: a free check for the watermark it has put on more than 180 billion images and videos - and that OpenAI, Nvidia and Kakao now add too, with Apple "soon"
Google's own page: the tool that was for media professionals last year is open "to everyone" from today, globally in English; its built-in checks in Search, the Gemini app and Chrome "now regularly handle over 1 million requests daily". The detector's own site draws the line coverage blurs: "This is not a general AI detector."
By Daily Aletheia · Checked against the primary source · 7 October 2026 · 2 min read

Google DeepMind opened its SynthID Detector to everyone on 7 October. Pushmeet Kohli, the lab's vice-president for science and strategic initiatives, wrote that since launching SynthID in 2023 Google has "helped people identify AI-generated media using imperceptible watermarks across images, video, and audio", and that "to date, we've watermarked over 180 billion images and videos, along with 240,000 years of audio content".
What changed today. "Last year, we introduced an early version of the SynthID Detector to help media professionals verify AI-generated content. Now, we're expanding access to everyone - with the tool available globally in English starting today."
What it checks. "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." The detector's own site says the same thing more plainly: "This is not a general AI detector. We can only detect media from the above companies, who have adopted SynthID technology." A file that comes back clean "was not likely to have been edited or generated by a SynthID-enabled AI model", the site says, "although this does not totally rule out AI generation or manipulation" - a model without the watermark, or heavy editing, produces the same result. "In very rare cases SynthID can also falsely trigger on non-watermarked content."
The scale of the existing checks. "This tool joins our built-in verification features in Search, the Gemini app, and Chrome, which now regularly handle over 1 million requests daily."
What the page does not say. No accuracy figure for the detector, no list of which OpenAI, Nvidia or Kakao products carry the watermark, and no date for Apple beyond "soon". The announcement is four paragraphs; the working detail sits on the detector itself, which accepts images, video and audio, and for video and audio reports "segments" that carry the watermark.
For anyone who publishes or buys content. A free check now exists for the watermark that four of the largest generators add. What it answers is narrow and reliable: whether this file carries a SynthID mark. What it does not answer is whether a file is real.
The Check
- What was said
- "The new SynthID website can now identify AI content from Google, OpenAI, and more." - Ars Technica, 7 October
- What the Source Says
- Google's page: the tool checks "if an image, video, or audio file was made with AI from Google or our partners, including OpenAI, NVIDIA, Kakao, and soon, Apple". The detector's own site: "This is not a general AI detector. We can only detect media from the above companies, who have adopted SynthID technology."
- The Difference
- It identifies the SynthID watermark, which four companies add to their generators' output. It does not identify AI content as such: a file from any other model, or one edited heavily, comes back as not detected, and the site says that result "does not totally rule out AI generation".