Google DeepMind has opened its SynthID Detector to the public, giving people a new way to check whether digital content was created using AI. The tool can identify invisible SynthID watermarks in images, videos and audio generated by Google and participating AI partners.
As artificial intelligence makes it increasingly easy to create photorealistic images, convincing videos and synthetic audio, figuring out what is real and what was generated by AI is becoming a growing challenge. Google is now expanding one of its key content-verification technologies to help address that problem.
Google DeepMind announced on October 7 that its SynthID Detector is now available globally in English, moving beyond its earlier limited access for journalists, media professionals and researchers. The standalone platform allows users to upload digital media and check whether it contains a SynthID watermark.
Google’s SynthID Detector Opens to the Public
SynthID is Google’s digital watermarking technology designed specifically for AI-generated content. Unlike a visible label or logo, the watermark is embedded into the content in a way that is generally imperceptible to people but can be detected by specialized technology.
Google says the system can be used to identify AI-generated images, video and audio, while SynthID technology has also been developed for AI-generated text.
The public launch of the Detector represents a significant expansion from its initial rollout in 2025, when Google introduced the verification portal primarily for professionals working with AI-generated media.
Now, instead of relying solely on journalists, researchers or technology companies to investigate suspicious media, ordinary users can also perform a SynthID check.
How Does Google SynthID Work?
The technology takes a different approach from conventional AI detectors.
Rather than simply analyzing whether something looks AI-generated, SynthID embeds a digital signal into content when it is created by supported AI systems. That signal is designed to remain detectable even after the media undergoes certain common modifications.
For images and video, Google says SynthID can remain detectable after changes such as cropping, filters, frame-rate adjustments and lossy compression. For audio, the watermark is designed to withstand modifications including noise, MP3 compression and changes in playback speed.
This makes the technology particularly relevant to the modern internet, where images and videos are routinely downloaded, edited, compressed and reposted across different platforms.
It’s Not Just About Google AI Anymore
One of the most notable aspects of the latest announcement is the expansion of SynthID beyond Google’s own ecosystem.
Google says the SynthID Detector can identify content created with AI from Google and participating partners, including OpenAI, NVIDIA and Kakao, with Apple expected to join soon.
That broader participation could make watermarking more useful as the AI industry becomes increasingly fragmented across dozens of models and platforms.
OpenAI has separately confirmed that it is using SynthID as part of its content-provenance efforts. The company said supported AI-generated images and audio can carry SynthID signals alongside other provenance technologies such as C2PA Content Credentials.
The development points toward a future in which AI-generated content may carry recognizable provenance signals regardless of which major AI platform produced it.
Why AI Content Verification Matters
The rapid improvement of generative AI has created a difficult information environment.
AI systems can now produce realistic photographs, cinematic video, human-like voices and sophisticated written material in seconds. While these capabilities can benefit creators, businesses and everyday users, they can also make misinformation, impersonation and manipulated media harder to identify.
Google has been developing SynthID partly to address these concerns.
The company says the technology is intended to help people understand how digital content was created and make more informed decisions about what they encounter online.
The challenge is becoming particularly important as AI-generated media moves from experimental tools into mainstream social media, advertising, entertainment, education and business communication.
SynthID Has Already Reached Massive Scale
Google’s watermarking technology is no longer a small-scale experiment.
In an earlier update, Google said SynthID had already been integrated into its generative AI products and had watermarked more than 100 billion images and videos and 60,000 years of audio.
The latest announcement indicates that the scale has grown even further, with Google reporting that SynthID has been used across more than 180 billion images and videos and roughly 240,000 years of audio.
That scale matters because the effectiveness of provenance technologies increasingly depends on how widely they are adopted.
SynthID Is Not an AI Truth Detector
Despite its potential, SynthID should not be viewed as a universal solution for identifying every piece of AI-generated media.
The system is primarily designed to detect the watermark associated with supported AI-generated content. If a piece of media was created using an AI system that does not use SynthID, the tool cannot simply prove that it was made by a human.
Google itself has previously described SynthID as an important building block rather than a complete solution for AI-content identification.
That distinction is important. A result showing no SynthID watermark does not automatically mean that content is authentic or entirely human-created.
Google Is Building a Larger Content-Provenance Ecosystem
SynthID is also part of a broader effort by Google to make digital content easier to trace.
The company has been working with C2PA Content Credentials, an industry standard designed to provide information about how media was created or modified. Google has also expanded verification capabilities across products including Search, Gemini and Chrome.
The approach combines multiple layers of information rather than relying on a single detection method.
For users, the ultimate goal is straightforward: provide more context about digital media at a time when seeing is no longer necessarily believing.
A New Step Toward More Transparent AI
The public launch of SynthID Detector comes as governments and technology companies increasingly focus on transparency around AI-generated content.
Google recently signed the EU AI Act Code of Practice on Transparency of AI-Generated Content, while continuing to work on watermarking and provenance technologies with other companies.
The move also reflects a broader shift in the AI industry. The conversation is no longer limited to how quickly companies can generate content; it is increasingly about how people can understand, verify and trust that content.
With SynthID Detector now available to everyone, Google is putting one more verification tool directly into the hands of internet users.
As AI-generated media becomes harder to distinguish from reality, technologies that can reveal a digital trail could become an increasingly important part of the online information ecosystem.
SEO Keywords: Google SynthID Detector, Google AI content detector, SynthID, AI-generated content detector, Google DeepMind SynthID, AI watermarking, AI content verification, AI-generated images, AI-generated videos, AI-generated audio, Google AI watermark
Read more: Norway’s Top 10 Women Entrepreneurs to Watch in 2026
