YouTube Shorts now integrates Google’s Veo 2, enabling AI-generated video clips with improved realism and customization. The feature includes SynthID watermarking and is available with expansion plans in the U.S., Canada, Australia, and New Zealand.
YouTube Shorts is expanding its AI-powered video creation tools by integrating Google DeepMind’s Veo 2, a generative AI model that allows users to create high-definition video clips from text prompts. This upgrade builds on Dream Screen, a YouTube feature that previously only supported AI-generated backgrounds.
Advanced AI Capabilities Offer Greater Creative Control
With Veo 2, creators can now generate standalone video segments that integrate seamlessly into their Shorts. The AI model provides a more realistic understanding of human movement and real-world physics, resulting in smoother, high-quality outputs. Users can also customize styles, lenses, cinematic effects, and durations, allowing greater creative control without advanced editing skills.
“Veo 2 understands real-world physics and human movement better, making its output more detailed and realistic,” said Dina Berrada, YouTube’s Director of Product. “You can even specify a style, lens, or cinematic effect, making Dream Screen an easy and fun way to express yourself.”
The feature is accessible through the Shorts camera, where users can open the media picker, tap “Create,” enter a text prompt, and choose visual elements before generating a clip. These clips can then be used within a larger Short or as standalone AI-generated content.
To ensure transparency, YouTube is implementing Google’s SynthID watermarking technology, which invisibly tags AI-generated content alongside visible labels marking synthetic videos. While these measures aim to prevent misinformation, concerns remain about AI’s role in deepfakes and misleading content.
he Veo 2 integration is now available to creators in the U.S., Canada, Australia, and New Zealand, with plans for a broader rollout in the future.
