ByteDance’s New AI Video Model Can Make 30-Second Clips From a Single Prompt
ByteDance’s New AI Video Model Can Make 30-Second Clips From a Single Prompt
https://www.cnet.com/tech/services-and-software/bytedance-introduces-new-seedance-2-5-video-model/
Publish Date: 2026-06-23 15:20:00
Source Domain: www.cnet.com
- ByteDance unveiled Seedance 2.5, an advanced AI video generator model, during a Beijing conference, showcasing improved capabilities for realistic video creation.
- The latest model allows users to provide up to 50 references (images, videos, or audio) for generating better control over 30-second, 4K videos from a single prompt, doubling the previous model’s reference limit of 12.
- The ability to generate high-quality synthetic video raises concerns about deepfake content; thus, watermarking is needed to identify synthetic or false material.
- ByteDance, known for its impressive AI models, plans to release the new version in China next month, though no specific timeframe was given for other countries.
- The release could face similar backlash as its predecessor if it doesn’t address legal and copyright issues, particularly following delays in the US last year due to concerns about the usage of copyrighted materials.
