Google has taken another major step toward improving AI-generated content identification with an updated version of its SynthID watermarking technology. The company has expanded SynthID to videos, allowing creators to embed an invisible digital watermark that can help identify content produced or modified using artificial intelligence.

Google says the new technology can add an invisible watermark to videos while maintaining their visual quality. The watermark is designed to remain detectable even when the content is edited, helping distinguish AI-generated videos from authentic footage. This could become increasingly important as generative AI makes it easier to create realistic images, videos and other digital media.

The upgraded SynthID system also works alongside Google’s existing tools for identifying AI-generated images and other content. According to Google, the technology can provide useful information about the origin and authenticity of digital media, giving viewers and platforms another way to assess whether content was created with AI.

The development comes as concerns grow over deepfakes, misinformation and AI-generated media. As synthetic content becomes more realistic, invisible watermarking could play an important role in improving transparency and digital trust.

Google’s latest move reflects a broader industry effort to develop responsible AI safety and content-authentication tools. While watermarking alone cannot prevent misuse of generative AI, technologies such as SynthID could help platforms, publishers and users better identify AI-created content and make more informed decisions about what they see online.