YouTube Shorts adds support for DeepMind’s Veo 2 Generative AI Model

YouTube Enhances Dream Screen with DeepMind’s Veo 2 Integration

YouTube has unveiled an exciting update to its Dream Screen feature for Shorts, incorporating Google DeepMind’s advanced video generation model, Veo 2. This model, which was introduced late last year, aims to compete with OpenAI’s Sora, providing creators with enhanced capabilities.

What is Dream Screen?

For those unfamiliar, Dream Screen enables users to effortlessly add AI-generated images and videos as backgrounds in their YouTube Shorts, using simply a text prompt. Initially launched as an experimental feature in June 2022, it became publicly available in November of the same year.

Key Features of Veo 2

The recent upgrade to Veo 2 significantly accelerates the video creation process while improving the level of detail and realism in the outputs. This powerful model can generate high-definition videos across diverse subjects and styles, boasting resolutions up to 4K and extending several minutes in length. Notably, it can also grasp human movement dynamics and real-world physics.

Create and Customize Your Shorts

In addition to generating backgrounds, Veo 2 allows users to produce standalone video clips suitable for inclusion in Shorts. To do this, users simply need to open the Shorts camera, tap ‘Add’ to access the media picker, and then select ‘Create’ at the top.

Following this, users can outline their concept with a text prompt, choose an image from the provided suggestions, and specify the desired length of the video clip. To ensure transparency, YouTube employs SynthID watermarks and clear labels, indicating that these videos are AI-generated.

Official Statements and Rollout Plans

YouTube stated in a recent blog post: “Together with Dream Screen, Veo 2 generates state-of-the-art, high-quality videos in a wide range of subjects and styles that better match your vision. Plus, we’ve also made improvements so Dream Screen now generates videos faster than before.”

The rollout of these features is currently underway for users in the United States, Canada, Australia, and New Zealand, with plans to expand availability to additional regions in the near future.

Source & Images