Following its major launch yesterday, Google announced two new features that will fundamentally change the user experience in the Gemini app. You can now give much more specific commands to the Nano Banana model by drawing on or adding notes to images. Additionally, the SynthID technology, which checks whether videos are AI-generated, has been expanded.
Add Notes to Images, Let Gemini Understand
When you add an image to the command box in Gemini, you can now select that image to launch the new “Mark-up” editor. This toolset offers two main options:
Drawing (Sketch): You can draw arrows on the image, circle specific areas, or freely draw the parts you want to highlight. You can easily correct your drawings with the undo/redundant buttons in the upper corner. Text: You can write your command directly on the image.
Shiftdelete.netThis feature allows you to communicate your needs simply by marking the image, instead of writing long texts. Moreover, this new capability has been simultaneously rolled out on Android, iOS, and web versions.
SynthID verification, which was introduced for images in November, now also covers videos. If you’re wondering whether a video you come across online is real or an AI creation, Gemini can now give you definitive answers.
How it works: Simply upload a video up to 100 MB in size and 90 seconds in length to Gemini and ask a question like, “Was this video generated by Google AI?”
Detailed Analysis: Google scans the video for hidden SynthID digital watermarks. The result doesn’t just say “yes” or “no”; it also specifies, second by second, which seconds of the image or which parts of the audio contain AI interference. For example: “SynthID detected in audio between seconds 10-20, but no AI traces found in images.”
The video and image verification feature is now available worldwide in all languages and countries where the Gemini app is supported.
