YouTube has announced the integration of new generative AI features for creating videos in Shorts, using the Veo 2 model from Google. This model now operates within the experimental Dream Screen feature, which allows users to generate individual video clips based on text descriptions. Previously, Dream Screen could already generate images and videos for Shorts content, but they could only be used as virtual backgrounds.
To use the new clip generation feature, users need to open the Shorts camera, select the media library, and tap the “Create” option at the top. Then, they can enter a text prompt describing the desired video and choose the style, lens, cinematic effect, and video length. This feature is available to users in the US, Canada, Australia, and New Zealand, with plans for further expansion.
The integration of Veo 2 enables Dream Screen to generate videos faster with more “detailed and realistic” results, taking into account real-world physics and human movement. All created content will be marked with both visual labels and invisible SynthID watermarks from Google, which identify that the video was created or modified using AI.
These new capabilities can significantly simplify the content creation process for creators, giving them more tools to bring their ideas to life without the need for complex equipment or special skills.
