Google has officially launched Gemini 3 with major improvements, but one of the most impactful improvements is not its enhanced generative UI and other expanded capabilities. Instead, it is a new feature that allows Gemini to detect whether an image was created with the help of artificial intelligence, due to Google’s upgraded SynthID watermark system.
These days, it seems to be more difficult to identify what is real from what is created by an AI, as it is possible to generate photos, videos, and audio in seconds. Google’s latest update aims to do so, providing a user with a simple way to check the legitimacy of online content.
Gemini can now spot invisible AI Watermarks
The new detection option is based on SynthID Detector, a technology Google introduced earlier this year. SynthID can embed an invisible, tamper-resistant watermark into AI-created images, videos, audio clips, and even text. Unlike traditional visible watermarks, SynthID operates silently in the background, adding identifiable markers to an image’s metadata and pixel structure without changing its appearance.
The latest version of Gemini now allows users to upload an image and tag @SynthID in the chatbot. Gemini will instantly scan the file and identify the hidden watermark, revealing whether it was created using Google’s AI tools or edited with them.
The system can also identify partial watermarks. If only part of an image was edited using AI, Gemini will show how much of the watermark remains detectable.
Discover more from WikiTechLibrary
Subscribe to get the latest posts sent to your email.
