Google has launched a new photo-to-video feature in Gemini, allowing users to turn still images into short, animated video clips complete with sound. The update builds on the company’s Veo 3 video generation model, which was released in May and has since seen over 40 million videos created by users across the Gemini app and Flow platform in the past seven weeks.
This new feature is now being gradually rolled out to Google AI Pro and Ultra subscribers in select countries, with global access expected to expand in the coming weeks.
How the feature works
Gemini’s photo-to-video tool allows users to upload an image, describe the scene and any audio cues, and then generate a dynamic eight-second video clip. Users can choose the ‘Videos’ option in the tool menu within the Gemini prompt box to begin the process.
The tool is designed to encourage creative expression, offering possibilities such as animating everyday objects, turning drawings and paintings into moving visuals, or adding motion to natural landscapes. Once the video is generated, it can be shared directly with others or downloaded for personal use.
In addition to Gemini, the same video generation capabilities are also available in Flow, Google’s AI filmmaking tool.
Focus on responsible AI use
Google has reiterated its commitment to safety and responsible AI development with this latest feature. The company said it performs “extensive red teaming,” which involves proactively stress-testing systems to identify and address issues before public release.
The tech giant also carries out detailed evaluations to anticipate how its tools may be used and misused, and updates its policies accordingly to prevent the creation of harmful or inappropriate content.
To maintain transparency, every video generated through the platform includes both a visible watermark and an invisible SynthID digital marker, clearly indicating that the content is AI-generated.
User engagement and feedback
Google is encouraging users to provide feedback on the photo-to-video tool by using the thumbs up and down buttons available on each generated video. This input will be used to further improve the overall experience and strengthen safety protocols.
The photo-to-video tool is now accessible via gemini.google.com, where eligible subscribers can start experimenting with video creation. The same capabilities are mirrored in the Flow tool, broadening access for users interested in AI-powered video storytelling.
