YouTube will let you turn photos into videos with AI – then upload them as Shorts

Your phone’s camera roll is likely full of static images. But YouTube is now introducing a way to make them move with a new suite of AI tools. Some of these are designed to let creators animate their photos and generate video effects without needing complex software.

Photo to video in no time

The main addition is a new photo-to-video tool. It works by using Google’s Veo video generation model to add motion to still images. This could apply subtle movement to a landscape photo or animate a group shot. The process involves selecting a photo from your camera roll, choosing a suggested animation, and tapping a button to create the clip.

The feature is being rolled out at no cost to creators in the US, Canada, Australia, and New Zealand. There’s no word on an Indian rollout at the moment, though that’s likely in the cards. Google also states that an integration with its newer Veo model is planned for later this summer.

ALSO READ: Google Search’s AI Mode is now on your homepage

YouTube is also launching new generative effects. These allow users to turn doodles into images or apply video effects to selfies, such as an animation that makes them appear to be swimming. These AI functions are collected in a new section called the AI Playground, which is accessible through a sparkles icon in the Shorts camera. The hub contains the creation tools, along with templates and pre-written prompts designed to streamline the creation process.

The update aligns with Google’s broader strategy of integrating AI across its products. Google Photos is also getting a similar photo-to-video function, plus a Remix tool for altering images into different artistic styles like anime and sketch.

It’s worth mentioning that the Shorts and Photos tools are more limited than the capabilities within the paid Gemini app. That’s hardly surprising considering they use pre-written prompts instead of allowing custom text descriptions.

Of course, as is the case with anything AI-generated, there’s always the possibility of misuse. For this, Google is embedding its invisible SynthID digital watermark into all such content for easier detection.