Google Gemini can now turn photos into videos with Veo 3

Turning a static image into motion used to require complex animation software, a lot of patience, or heavy post-production. Google wants to cut that entire workflow down to a single upload and a text prompt.

Google just rolled out a new photo-to-video feature in Gemini, powered by its flagship Veo 3 video model launched back in May. The update allows users to take existing pictures and convert them into dynamic, eight-second clips complete with generated audio.

Animating stills with synchronized sound

The workflow sits right inside the standard Gemini interface. Users click Videos in the tool menu inside the prompt box, upload a still photo, and type out motion descriptions alongside audio instructions.

From there, the system generates an eight-second clip that includes realistic sound effects and movement. Google suggests animating everyday objects, bringing hand-drawn sketches to life, or adding dynamic movement and ambient audio to still landscape shots.

The feature is currently rolling out to Google AI Pro and Ultra subscribers in select countries at gemini.google.com. It is also being added simultaneously to Flow, Google’s dedicated AI filmmaking workspace.

Google also dropped a massive adoption stat: over the past seven weeks since Veo 3 launched, users have generated more than 40 million videos across Gemini and Flow. Last week, standard Veo 3 access expanded to Pro subscribers in more than 150 countries.

On the safety side, Google says each generated clip includes both a visible watermark declaring it as AI-generated and an invisible SynthID digital watermark embedded directly into the file. The company also pointed to extensive internal red teaming to curb abuse.

Why it matters

Text-to-video has always suffered from a control problem. Prompting from scratch feels like pulling a slot machine handle, where you never quite know what character or setting you will get.

By anchoring generation to an existing photograph, photo-to-video provides immediate consistency. For creators, animators, and social marketers, being able to take a static concept art piece or product photo and generate an eight-second teaser with native sound effects fundamentally speeds up early-stage prototyping.

Is it worth upgrading to Google AI Pro or Ultra for? If you actively produce visual content, absolutely—getting integrated audio alongside video saves you from juggling multiple separate sound-design apps. For casual experimenters, locking this behind top-tier subscriptions keeps the barrier relatively high.

In the broader race against Runway, Pika, and OpenAI's Sora, Google is making a very smart move by prioritizing native audio synthesis and workflow convenience inside a general chat app. The biggest risk ahead remains deepfake mitigation: animating realistic human faces from arbitrary photos poses severe impersonation risks, and invisible watermarks like SynthID will face their toughest real-world stress test yet.