ByteDance upgrades its AI video creation model
What's the story
Chinese tech giant ByteDance has unveiled the latest version of its artificial intelligence (AI) video generator model, Seedance 2.5. The announcement was made at a conference in Beijing on Tuesday. This new model is an upgrade over its predecessor, Seedance 2.0, and promises to offer even more realistic imagery in AI video generation.
Advanced features
Seedance 2.5 allows up to 50 reference pieces
The new model, Seedance 2.5, can create 30-second long videos in 4K resolution from a single prompt. It also allows users to provide up to 50 reference pieces such as images, videos, or audio files. This is a major improvement over its predecessor which only allowed for 12 references. The increase in the number of references gives users more control over the video creation process.
Market position
ByteDance's presence in AI video generation
ByteDance has been a major player in the AI video generation space, creating models that rival those of OpenAI and Google. The company, which was once the majority owner of TikTok, is said to be launching this new model in China next month. However, no timeline for international release has been given yet.
Legal concerns
Legal issues surrounding Seedance's rollout
The launch of the new model could bring legal issues, especially considering the controversy surrounding Seedance 2.0's US rollout. The earlier version was delayed under Hollywood's pressure to stop using copyrighted works that appeared to be used for training the model. If Seedance 2.5 is significantly better than its predecessor, it might face similar backlash if it doesn't address these legal and copyright issues effectively.