Google is stepping up its creative AI offerings with the global rollout of a new photo-to-video capability in Gemini, its AI-powered assistant. The tool, which began rolling out to Google AI Pro and Ultra subscribers this week, enables users to turn still images into animated eight-second videos—complete with sound.

The feature builds on the success of Veo 3, Google’s video generation model launched in May and now widely available via Gemini and Flow, Google’s AI filmmaking platform. In the seven weeks since its debut, more than 40 million videos have been generated, showcasing a wide spectrum of user creativity—from modern influencer-style fairy tale retellings to lava-themed ASMR clips.

“Your imagination is the limit when you create videos with Gemini,” Google said in its blog, highlighting how users have animated everyday objects, personal artworks, and nature scenes.

To use the new feature, subscribers can navigate to the ‘Videos’ tool in the Gemini prompt box, upload an image, and enter a scene description along with audio prompts. The AI then generates a dynamic video from the static photo. Finished clips can be downloaded or shared directly from the platform.

Safety Front and Centre

As generative video tools grow in popularity and capability, so too do concerns around responsible AI use. Google says it is continuing its commitment to safety through “extensive red teaming” and rigorous evaluation processes aimed at preventing misuse.

All AI-generated videos include a visible watermark and an invisible digital tag powered by SynthID, a technology designed to signal the content’s artificial origin. Users are also encouraged to rate their videos using thumbs up or down feedback buttons—part of Google’s broader effort to refine its tools and maintain responsible content generation.

The new feature is now available at gemini.google.com for users in supported countries and also through Google’s Flow tool.