Gemini App Introduces "Veo 3" Photo-to-Video Generation

Following last week’s global rollout of Veo 3, Google is now equipping the Gemini app with a powerful new feature: photo-to-video generation.

This capability allows users to transform static images into short, eight-second video clips, enriched with audio, using the latest Veo 3 model.

Once you select the “Videos” tab inside Gemini’s prompt bar, you can now upload a photo. Just as before, you’ll describe the scene and the desired audio elements—whether it’s dialogue, sound effects, or ambient noise. According to Google, you can “get creative by animating everyday objects, bringing your sketches and paintings to life, or adding movement to natural landscapes.”

The final result is downloadable and shareable as an MP4 file (720p, 16:9 landscape format).

Starting today, this feature is rolling out globally—but only for users subscribed to Google AI Pro ($19.99/month) or AI Ultra ($249.99/month). Initially, it’s being made available on gemini.google.com, with mobile app support to follow shortly after.

Built-in Safety and Responsible AI Measures

To address concerns around authenticity and misuse, every generated video is watermarked with “Veo” in the bottom-right corner. Additionally, an invisible synthetic digital watermark is embedded to ensure traceability. Users are also invited to provide feedback using thumbs-up or thumbs-down buttons.

Behind the scenes, Google has implemented extensive “red-teaming” procedures—actively stress-testing the system to uncover and resolve risks before they reach the public. This includes robust evaluations to understand real-world use cases and mitigate potential abuse. Google continues to evolve its content policies to protect against harmful or unsafe outputs.

According to Google, more than 40 million Veo 3 videos have been created within the last seven weeks across the Gemini app and its sibling tool, the Flu Film Generator, which already supported photo-to-video functionality.