Google DeepMind’s SynthID is an invisible digital watermarking and detection system that embeds a signal directly into the pixels of an AI-generated image at the point of creation. The technology launched in August 2023 as a prototype, initially available to select customers using Imagen. In May 2025, Google announced that SynthID had been applied to over 10 billion pieces of content, spanning images, video, audio, and text, and launched a unified Detector portal. SynthID now covers four content types: images, video (via Veo), audio (via Lyria), and text (via Gemini). For images, SynthID relies on two methods: a watermarking model that makes subtle adjustments to individual pixels, and a detection model trained to identify AI-generated content. While other companies have started to adopt SynthID watermarks, Gemini can currently only recognise content created by Google AI tools. The system holds up against JPEG compression, resizing, colour space conversions, moderate brightness and contrast adjustments, and light cropping.