Google’s SynthID Detector Lets You Check If Media Was Made With AI

Google's SynthID Detector identifies AI-generated images, videos, audio, and text by detecting digital watermarks. It's now available via a public web portal for media verification.

Categorized in: AI News PR and Communications
Published on: May 21, 2025
Google’s SynthID Detector Lets You Check If Media Was Made With AI

Google Launches SynthID Detector to Identify AI-Generated Media

Google has introduced SynthID Detector, a new tool that can identify whether images, videos, audio, or text have been generated using Google's AI models. Initially, SynthID was a digital watermark applied to images created with Google's Imagen model back in 2023. Now, the company is expanding this technology with a public-facing web portal that scans uploaded media for those watermarks.

How SynthID Detector Works

The tool offers a simple upload interface where users can submit various types of content. SynthID Detector then analyzes the media and reports if it contains AI-generated elements marked by SynthID watermarks. For images and videos, the detector can highlight specific sections most likely to have been watermarked. Audio files are also supported, with the tool able to pinpoint the exact segment of a track carrying the watermark.

Collaboration and Open-Source Efforts

While SynthID was designed specifically for Google's AI outputs, the company encourages other organizations to adopt this watermarking system. An open-source version focused on text watermarking is already available. As part of the rollout, Google is also partnering with NVIDIA to apply SynthID marks on media created by NVIDIA's Cosmos model.

Additionally, third-party security firms like GetReal Security will have capabilities to verify the presence of SynthID marks, providing broader options for verification beyond Google's own tool.

Why PR and Communications Professionals Should Care

With AI-generated content becoming more common across images, video, audio, and text, tools like SynthID Detector offer a way to verify the authenticity of media. This can be invaluable for PR teams managing brand reputation, ensuring transparency, and preventing misinformation. Knowing if content was AI-generated helps maintain trust and credibility in communications.

That said, the effectiveness of SynthID Detector depends on the widespread use of watermarks by AI content creators. Until then, it serves as an important but partial solution in identifying AI-generated media.

How to Access SynthID Detector

Journalists, researchers, and developers interested in testing the tool can join a waitlist through Google's online form. The announcement came during Google's annual I/O developer conference, which started on May 20, 2025.


Get Daily AI News

Your membership also unlocks:

700+ AI Courses
700+ Certifications
Personalized AI Learning Plan
6500+ AI Tools (no Ads)
Daily AI News by job industry (no Ads)
Advertisement
Stream Watch Guide