Gemini now lets you make 8-second videos with just a picture

Google, one of the largest tech companies in the world, has engineered its Gemini AI to be one of the most advanced in existence. Now the company has launched Veo 3 11 – built to create lifelike videos from scratch. Here’s how it works.

Users can choose from either creating a verbal prompt themselves, or using an existing image they have. The first option works just like an AI image generation tool. You type down what exactly you want to appear and the Gemini Veo 3 tool will deliver the output.

On the other hand, using an existing image opens a realm of possibilities. For example, uploading a photo of a box and letting the AI tool surprise you can result in unparalleled creativity.

We’ve only seen the examples so far, but the samples were far more realistic than we were expecting. With resulting videos being as long as eight seconds, this is definitely an advancement in AI generation tech.

The feature is currently limited to Google AI Pro and Ultra users. The featureset is limited as seen in the table below:

Based on the website, it also works on mobile. This accessibility is likely to make it a go-to for all video-generation needs moving forward. To prevent abuse and misuse, Google is taking stringent measures to ensure that all resulting outputs are in line with fair policies.

What do you think?