Updated
Updated · TechCrunch · Oct 7
Google Opens SynthID Site to Public as 1 Million Daily Checks Target AI Media
Updated
Updated · TechCrunch · Oct 7

Google Opens SynthID Site to Public as 1 Million Daily Checks Target AI Media

3 articles · Updated · TechCrunch · Oct 7

Summary

  • Google on Tuesday opened its SynthID verification website to everyone, letting users check whether images, video or audio were generated with AI rather than limiting access to journalists, researchers and media professionals.
  • 1 million verification requests are already made each day, and the service now accepts major file types including JPG, MP4 and MP3 across image, video and audio uploads.
  • SynthID works by detecting invisible watermarks embedded by Google's AI tools—including Gemini, Veo and Lyria—and the checker also supports partner content from OpenAI, Nvidia and Kakao, with Apple expected soon.
  • Google has also built SynthID verification into the Gemini app and Chrome, though the broader watermarking push still faces limits because detection tools can miss content made by their own companies' models.

Insights

If SynthID only flags participating AI models, are we leaving the door wide open for untraceable deepfakes from rogue developers?
Why did Google release a simplified public AI detector while keeping its most powerful forensic tracking tools locked away internally?
Could heavy image editing strip away Google's invisible AI watermark, giving malicious actors a perfect loophole to spread misinformation?