YouTube Shorts is set to undergo a significant transformation with the integration of Google DeepMind’s Veo 2, the latest version of its text-to-video model. This new feature allows creators to generate standalone AI video clips, enhancing their Shorts content.
Last year, YouTube introduced Dream Screen, enabling creators to create AI-powered backgrounds using text prompts. With Veo 2, creators can now generate complete video clips, offering them more creative freedom and the ability to tell unique stories.
According to YouTube's official blog, Veo 2 supports a wide range of video styles and subjects and enhances speed and realism. Users can also customise their videos with different styles, cinematic effects, and lens options.
YouTube’s Director of Product, Dina Berrada, emphasised that Veo 2 improves the understanding of real-world physics and human movement, making AI-generated content more natural and lifelike. She noted that the platform now generates videos faster, reducing wait times for creators.
To ensure transparency, YouTube will utilise DeepMind’s SynthID, a watermarking tool that embeds digital markers into AI-created videos. AI-generated videos will also be clearly labelled to help viewers identify them.
Creators can generate both video backgrounds and standalone clips using Veo 2. To create an AI background, users need to open the Shorts camera, select Green Screen, tap Dream Screen, and enter a text prompt. For AI video clips, they must open the Shorts camera, tap Add to open the media picker, tap Create, and enter a text prompt with the desired length.
The new features powered by Veo 2 are currently available for creators in the U.S., Canada, Australia, and New Zealand, with plans for expansion to other regions in the future. YouTube is enthusiastic about providing creators with more AI-powered tools to bring their ideas to life.