Google Gemini can now turn your photos into videos with Veo 3

If you’ve ever stared at an old vacation photo and wished it could just spring to life, complete with crashing waves, ambient chatter, or even whispered dialogue, Google’s making that daydream a reality.

The company has just rolled out a new photo-to-video feature for its Gemini AI platform, letting users transform static images into dynamic eight-second clips, complete with AI-generated audio.

This new capability is powered by Veo 3, Google’s latest-gen video model that’s quietly been shaping the future of generative filmmaking. Now, it’s landing directly in the hands of Gemini Ultra and Pro subscribers, as part of a wider rollout that hits web users today and mobile users over the next few days, though access is still limited to select regions.

How to turn your photos into videos using Veo 3 on Gemini

Here’s how it works – upload a photo, write a short description of what you want it to do, whether that’s a wind-swept tree rustling in the breeze or a streetlamp flickering under the rain, and Gemini does the rest.

You can even add sound design details, like footsteps, dialogue, or environmental ambience. The system then generates a 720p MP4 video in 16:9, all “perfectly synced with the visuals,” according to Google.

ALSO READ: YouTube Shorts, Canva get a major AI upgrade with Veo 3

There’s no need to jump between platforms either. While a similar tool already lives inside Flow, Google’s filmmaking-focused AI tool released in March, this update brings visual storytelling straight into Gemini’s core UI. Users can find the new tool under the “video” option in the prompt bar.

All outputs come stamped with a visible watermark, as well as Google’s invisible SynthID digital watermarking tech to flag the content as AI-generated.

Why the feature is more important than you think

So, what’s the use case? Think less deepfake and more creative spark. You can animate your sketches, breathe life into a still frame of your pet, or make your childhood doodles move, sound, and feel like something out of a dream sequence. It’s storytelling at the intersection of imagination and machine learning.

As a bonus, Google also announced that Flow is expanding to 75 additional countries starting today; so if you’ve been waiting to dip your toes into AI video creation, now’s your moment.