Seedance 2.0, an AI video generation model developed by ByteDance, has recently gone viral in China but is now facing industry scrutiny over technical controversies. The model reportedly demonstrated the ability to generate highly accurate personal voice characteristics using only facial images, even without user authorization.
While the feature showed strong technical capabilities, it raised serious privacy and ethical concerns and has since been urgently suspended.
ByteDance’s Seedance 2.0 leverages a dual-branch diffusion transformer architecture to generate video visuals and native audio simultaneously. Users can input text or upload images, and within 60 seconds, the model produces multi-shot sequences at 2K resolution.
Its standout feature is the ability to generate coherent multi-scene content from a single prompt, automatically parsing narrative logic while maintaining consistent character representation.
In a recent test, Pan Tianhong, founder of tech media outlet MediaStorm, discovered that uploading a personal facial photo caused the model to produce audio nearly identical to his real voice — without using any voice samples or authorized data.
The revelation has quickly raised public concerns over AI-driven identity forgery. If misused, the technology could enable new forms of abuse, including deepfake news, scams, and reputational attacks.
On Monday, operators of the Jimeng platform (the Chinese-market app name of Seedance 2.0) announced, “To maintain a healthy and sustainable creative environment, we are urgently updating the model and will immediately suspend the feature that allows real photos or videos to be used as reference subjects.”
At the same time, ByteDance’s Jimeng and Doubao apps introduced a live verification step, requiring users to record their own image and voice before creating a digital avatar. ByteDance stressed that these adjustments are designed to uphold the bottom line of responsibility while balancing innovation with regulatory compliance.
Seedance 2.0 remains in its internal testing phase, with ByteDance taking proactive steps such as limiting certain features and strengthening content review to mitigate potential risks. The move shows the company’s commitment to ethical AI development.
As the technology advances, striking the right balance between innovation, data compliance, and copyright protection becomes not just a corporate responsibility but a global challenge for the AI industry.
Industry observers remain optimistic about Seedance 2.0’s potential in AI-driven short dramas and animated series, but they emphasize that any real-world deployment must be grounded in safety, control, and accountability.
