New Delhi: At a time when artificial intelligence is making fake photos, videos and audio clips look increasingly real, Google is strengthening its SynthID technology to help users identify AI generated content with greater confidence.
SynthID is Google DeepMind’s watermarking tool designed to mark and detect content created or edited through artificial intelligence. The company says the technology adds an invisible digital signal to AI generated media, helping people understand whether a piece of content has been created or altered using AI tools.
According to Google’s support page, SynthID Verification is available inside the Gemini app for signed in users. It can help identify images, videos and audio generated or edited by Google’s AI models. The feature scans the uploaded media and checks for the SynthID watermark.
The move is important because deepfakes and AI edited content are becoming a serious challenge for ordinary users, journalists, fact checkers and public institutions. During elections, conflicts and major breaking news events, fake visuals can spread quickly on social media before people get time to verify them.
Google has said that when SynthID detects a watermark, it can highlight the parts of an image, audio file, video or text that are most likely to carry the AI watermark. For images, it can indicate the likely watermarked areas. For audio, it can point to specific segments where the watermark is detected.
The latest development also connects with a wider industry push for content transparency. Reports from Google I/O say the company is expanding support for SynthID along with C2PA content credentials, a system that records the origin and edit history of digital media. This can allow users to know not only whether a file is AI generated, but also how it may have been changed.
For users, this means AI detection may become more accessible across Google products. Reports suggest that verification features are being integrated with tools such as Gemini, Google Search, Lens and Chrome, making it easier for people to check suspicious images without depending only on external fact checking websites.
The technology is not a complete solution to the deepfake problem. Watermarks can help when content is created through supported AI platforms, but not every fake image or video carries such a signal. Metadata can also be removed in some cases. That is why experts believe watermarking must work alongside media literacy, fact checking and platform level moderation.
Still, SynthID marks a significant step in the fight against digital misinformation. As AI tools become more powerful, the ability to identify what is real and what is synthetic will become essential for public trust.
For India, where political clips, celebrity videos and viral social media posts often reach millions within minutes, tools like SynthID could become especially useful. The challenge ahead is clear: AI will keep creating more convincing content, and technology must keep improving to help people separate truth from manipulation.