2027: Google opens AI watermark detector to public
The tool allows users to check images, videos and audio for invisible AI watermarks, among other features The post 2027: Google opens AI watermark detector to public appeared first on Premium Times Nigeria.
Google's SynthID technology embeds invisible digital watermarks into AI-generated images, audio and video at the point of creation, allowing the content to be traced back to AI generation tools even after editing or compression. The watermark detector, previously limited to select partners and testers, verifies whether a piece of media carries this hidden marker, offering a way to flag synthetic content without altering how it looks or sounds to the human eye.
The rollout comes as concern grows globally and in Nigeria over AI-generated disinformation ahead of elections, with deepfakes and synthetic audio clips increasingly used to spread false claims about politicians or incite unrest. By making the detector publicly accessible, Google is positioning the tool as a resource for journalists and fact-checkers working to verify content ahead of Nigeria's 2027 general election, though questions remain over how widely it will be adopted locally, how effective it is against AI tools that don't use SynthID, and whether it can keep pace with rapidly evolving generative AI models.