ByteDance Seedance 1.5 Pro: AI Video with Native Audio
The previous generation of ByteDance's AI video model with speech and sound effects.
Read Article →
On February 12, 2026, ByteDance officially launched Seedance 2.0, a next-generation AI video model that generates cinematic, multi-shot videos up to 15 seconds long. The model is available to Chinese users through ByteDance’s Jianying app and is expected to roll out globally through CapCut.
Within hours of launch, users began creating viral videos featuring Hollywood intellectual property - triggering one of the most significant copyright confrontations in AI video history.
According to Reuters, ByteDance described the system as designed for “professional film, e-commerce, and advertising productions” capable of processing text, images, audio, and video simultaneously.
Accepts text, up to 9 images, 3 video clips, and 3 audio clips simultaneously
Camera movement, lighting, VFX, composition, and motion direction
Generates multi-shot videos with native sound and visual sync
15-second clips with improved physics, motion stability, and realism
Unlike most text-to-video tools that rely primarily on text prompts, Seedance 2.0 accepts four input types within a single generation pipeline. Users can combine natural language instructions with reference images, video clips, and audio to guide the output.
According to BusinessToday, the model emphasizes professional-level controllability - users can command camera movement, lighting, visual effects, and composition like a film director rather than relying solely on text prompts.
| Feature | Seedance 1.5 Pro | Seedance 2.0 |
|---|---|---|
| Max Video Length | 12 seconds | 15 seconds |
| Resolution | 720p | Higher (exact TBD) |
| Input Types | Text, Image | Text, Image, Video, Audio |
| Multi-Shot | No | Yes, cinematic multi-shot |
| Audio Sync | Basic speech | Advanced synchronized audio |
| Film-Style Controls | Limited | Camera, lighting, VFX direction |
| Availability | ImagineArt | Jianying (China), CapCut (global, coming soon) |
The jump from 1.5 Pro to 2.0 represents a generational leap. Where the previous version focused on affordable audio-visual generation at lower resolutions, Seedance 2.0 targets professional-quality cinematic output with granular creative controls.
See how Seedance, Kling, Sora, Runway, and others stack up for your projects.
View Full Comparison →The speed of the backlash was unprecedented. Within a day of launch, major Hollywood organizations issued formal statements:
Users immediately tested Seedance 2.0 with Hollywood IP, creating videos that spread across X/Twitter and Weibo:
Motion Picture Association CEO Charles Rivkin issued a statement demanding ByteDance “immediately cease its infringing activity”:
“In a single day, the Chinese AI service Seedance 2.0 has engaged in unauthorized use of U.S. copyrighted works on a massive scale. By launching a service that operates without meaningful safeguards against infringement, ByteDance is disregarding well-established copyright law.”
Disney sent a cease-and-desist letter accusing ByteDance of a “virtual smash-and-grab of Disney’s IP,” after videos featuring Spider-Man, Darth Vader, and Grogu surfaced. Axios reported Disney claimed ByteDance was “hijacking Disney’s characters by reproducing, distributing, and creating derivative works.”
SAG-AFTRA said it “stands with the studios in condemning the blatant infringement enabled by Bytedance’s new AI video model.”
The Human Artistry Campaign, backed by Hollywood unions and trade groups, called Seedance 2.0 “an attack on every creator around the world.”
The reactions from individual creators ranged from alarm to measured perspective:
Disney isn’t opposed to all AI partnerships. While it sent a cease-and-desist to ByteDance (and previously to Google over similar issues), Disney has also signed a three-year licensing deal with OpenAI - demonstrating that the industry distinguishes between unauthorized use and licensed collaboration.
In China, Seedance 2.0 has been framed as a national achievement comparable to the DeepSeek breakthrough.
Chinese state-backed newspaper Global Times wrote: “Early last year, the release of DeepSeek-R1 sparked heated debate in the U.S. tech community over a ‘Sputnik moment.’ This year, the continued breakout success of Seedance 2.0 has given rise to a wave of admiration for China within Silicon Valley.”
Hashtags related to Seedance 2.0 generated tens of millions of clicks on Weibo. Elon Musk amplified the buzz by commenting “It’s happening fast” on a post praising the model.
Meanwhile, other Chinese AI firms launched competing models around the same time. CNBC reported that Alibaba released RynnBrain for robotics and Kuaishou launched Kling 3.0, signaling an AI video arms race within China.
Seedance 2.0 raises the stakes on multiple fronts:
The Seedance 2.0 controversy highlights a critical distinction that creators should understand:
This isn’t the first copyright confrontation - and it won’t be the last. But the scale and speed of Seedance 2.0’s IP-infringing outputs may accelerate:
HeyGen lets you create professional AI avatar videos for marketing and training with commercial licensing included.
Try HeyGen Free →Seedance 2.0 is ByteDance's latest AI video generation model, launched February 12, 2026. It creates cinematic, multi-shot videos up to 15 seconds long from text, images, video clips, and audio inputs. It's integrated into ByteDance's Jianying app in China and is expected to launch globally through CapCut.
Users immediately began creating videos featuring copyrighted Hollywood characters and real actors' likenesses without authorization. The Motion Picture Association, Disney, SAG-AFTRA, and the Human Artistry Campaign all condemned ByteDance for launching without meaningful copyright safeguards.
As of February 2026, Seedance 2.0 is available to Chinese users through ByteDance's Jianying app. ByteDance has announced it will be available globally through the CapCut app, but a specific global launch date has not been confirmed.
Both generate high-quality AI video, but they differ in approach. Sora 2 produces clips up to 60 seconds with OpenAI's content safety guardrails and licensing deals (including with Disney). Seedance 2.0 generates 15-second multi-shot clips with multimodal input (text, images, video, audio) but has faced criticism for lacking copyright protections.
Creating AI-generated videos using copyrighted characters or real people's likenesses without permission raises serious legal concerns. Studios like Disney are actively pursuing legal action against unauthorized use. Licensed AI tools with proper content agreements are the safest option for commercial use.
Disney sent a cease-and-desist letter to ByteDance after users created videos featuring Disney characters including Spider-Man, Darth Vader, and Grogu (Baby Yoda). Disney accused ByteDance of 'hijacking' its characters through unauthorized reproduction and distribution.