As OpenAI shut down its video generation app Sora, ByteDance will add a new AI video generation model, Dreamina Seedance 2.0, to CapCut, TechCrunch reported on March 26.

The Dreamina Seedance 2.0 model can automatically generate, edit and sync video and audio using only text, images and reference videos.

Dreamina Seedance 2.0 will be rolled out sequentially in Brazil, Indonesia, Malaysia, Mexico, the Philippines, Thailand and Vietnam, and is set to expand to more markets. A global release is being delayed due to controversy over copyright infringement, and in China it will be provided through ByteDance's Jianying app, TechCrunch reported.

The model can generate high-quality videos using text alone and can be used to produce a range of content, including cooking recipes, fitness tutorials and product introductions. It supports videos up to 15 seconds long and 6 aspect ratios, and will also be applied to AI video editing functions and video creation tools.

ByteDance said it added a feature to block generation of videos containing the faces of real people to protect copyrights, and also strengthened security measures to prevent unauthorized use of intellectual property. It plans to embed an invisible watermark in generated content to support responses to copyright infringement.