Seedance 2.0 claims to produce “movie content” with “seamless video enhancement and natural language control.”
ByteDance, TikTok’s parent company, released a pre-release version of a new AI video model called Seedance 2.0 over the weekend. Positive reaction to the announcement boosted the Chinese AI company’s stock price.
Seedance 2.0 markets itself as a “true” multimodal AI creator that allows users to combine images, video, audio, and text to generate “cinematic content” with “accurate referencing capabilities, seamless video enhancements, and natural language control.” The model is currently available to some users of Jimeng AI, ByteDance’s AI video platform.
The new Seedance model can export at 2K with 30pc faster generation than the previous version 1.5 (company website).
Switzerland-based consulting firm CTOL calls it “the most advanced AI video generation model available” that “outperforms OpenAI’s Sora 2 and Google’s Veo 3.1 in practical tests.”
Publisher COL Group Co hit the daily price ceiling of 20%, while Shanghai Film Co and games and entertainment company Perfect World Co rose 10%, according to data compiled by Bloomberg today. Meanwhile, the Shanghai Shenzhen CSI 300 index is up 1.63% at the time of publication.
Consumer AI video generators have come a long way in just a short period of time. As is often the case, blurred fingers, overly smooth and unrealistic skin, and unexplained changes from frame to frame in AI videos are all very difficult to notice.
Competing AI video generator Sora 2 from OpenAI produces watermarked results (though there’s no shortage of tutorials on how to remove watermarks). Google’s Veo 3.1 comes with a metadata watermark called Syth ID, but Seedance boasts that its results are “completely watermark-free.”
The proliferation and accessibility of advanced AI tools has opened the door to a new wave of AI deepfakes, with xAI’s Grok and others at the center of the problem.
Last month, the EU launched a new investigation into X to examine whether the Elon Musk-owned social media site adequately assessed and mitigated the risks posed by Grok, an AI chatbot within its platform that includes image editing capabilities.
Users of the social media site quickly used the tool to encourage the people (usually women and children) featured in the images and videos to undress. According to the New York Times, millions of such pieces of content were generated on X.
Don’t miss out on the knowledge you need to succeed. Please sign up for daily briefsSilicon Republic’s digest of science and technology news you need to know.
