One aspect of AI that is rapidly developing is video generation, and Google has developed its Veo 3 model to focus on that. The tech giant has just added a new trick to Gemini and Veo, and that's the power to generate videos, eight seconds long, from a still image. It's not a new concept, but this is arguably the best execution of it yet.

Starting today, Google is bringing image-to-video generation to the Gemini app. The feature comes courtesy of the company's Veo 3 model, which Google began rolling out more broadly to AI Pro users last week after it was initially only available to AI Ultra subscribers.

Instead of typing out every detail about the objects you want in a video, the new tool lets you upload a photo and just add instructions for how you want them to move.