ByteDance's Seedance 2.0, launched in pre-release on February 7th, marks a major step forward in AI video generation — producing multi-shot, 2K cinematic sequences with native audio, sound effects, and lip-sync in 8+ languages from simple text or image prompts in under 60 seconds. Its reference capability allows creators to replicate camera work and effects from existing footage, with Swiss consultancy CTOL rating it ahead of both OpenAI's Sora 2 and Google's Veo 3.1 in practical testing. However, the launch has been tempered by privacy concerns after a voice-cloning feature was urgently suspended, and the model's watermark-free output has raised wider questions about deepfake misuse.
Seedance 2.0 goes live as ByteDance’s AI videos ignite China market rally