YouTube has rolled out a major update to Dream Screen, its AI-powered tool for generating custom backgrounds in Shorts. This upgrade, powered by Google DeepMind’s latest video model, Veo 2, enhances the creative potential of YouTube Shorts.
According to Dina Berrada, Director of Product for Generative AI Creation at YouTube, Veo 2 introduces standalone video generation, allowing creators to produce independent video clips and seamlessly integrate them into their Shorts.
This update is particularly valuable for content creators who need specific scenes but lack the right footage. By simply entering a text prompt, users can generate custom video clips that align with their creative vision. The AI model improves video quality by offering better rendering speed, more realistic physics, and more accurate human movement. Additionally, creators can personalize their content by selecting from a range of styles, cinematic effects, and lenses, making the final output more visually engaging.
Accessing these features within YouTube Shorts is a straightforward process. Users can open the Shorts camera, select the Green Screen option, and then choose Dream Screen. They can enter a prompt, pick an image, and set it as a video background. Alternatively, they can tap the “Add” button, access the media picker, select “Create,” enter a prompt, choose an image, and set the desired video length before finalizing the video.
To ensure transparency, YouTube applies SynthID watermarks to AI-generated content, helping viewers distinguish between AI-created and human-made videos. These features are currently available in the United States, Canada, Australia, and New Zealand, with plans for a global rollout in the near future.