A 15-second AI-generated video depicting Tom Cruise and Brad Pitt engaged in a high-stakes fight has sent ripples of alarm through Hollywood, prompting one prominent screenwriter to declare, “It’s likely over for us.” The video, created using the Seedance 2.0 AI model developed by Chinese tech company ByteDance, demonstrates a level of realism that has industry professionals fearing the potential displacement of human creatives.
The clip, which circulated widely on X (formerly Twitter), shows the two actors battling on a rooftop. Irish filmmaker Ruairi Robinson created the video using a simple two-line prompt, according to his posts. The ease with which such a convincing simulation can be produced is at the heart of the growing anxiety.
Rhett Reese, screenwriter of the Deadpool franchise, expressed his concerns on X, stating that a single individual with sufficient talent – comparing it to the level of Christopher Nolan – could soon create films indistinguishable from those currently produced by major studios. He added that he was “terrified” by the encroachment of AI into creative fields, acknowledging the potential loss of careers, including his own.
The rapid advancement of AI video generation has triggered a swift response from Hollywood. Paramount and Disney have reportedly sent cease-and-desist letters to ByteDance, accusing the company of intellectual property infringement, while the Motion Picture Association and SAG-AFTRA have condemned the unauthorized use of copyrighted works and actor likenesses. ByteDance responded with a statement indicating it would implement safeguards to protect intellectual property, though details remain scarce.
Seedance 2.0 is the latest in a series of increasingly sophisticated AI models released by Chinese technology companies this year, intensifying competition in the AI industry. China’s government has prioritized advanced technology as a key component of its national development strategy, showcasing humanoid robots performing complex martial arts maneuvers during this week’s Lunar New Year celebration.
The emergence of this technology is fueling a sense of technological rivalry between the US and China, reminiscent of the 20th-century “Space Race” with the Soviet Union. Ramesh Srinivasan, a professor at the University of California, Los Angeles, described a “nationalist fervor” surrounding which country will “win” the race to dominate AI development.
The AI model has been praised for its ability to generate short scenes with polished characters and motion editing control at a lower cost than traditional filmmaking. However, concerns extend beyond copyright infringement. A Chinese tech blogger demonstrated that Seedance 2.0 could generate realistic audio of his voice based solely on an image, raising significant privacy concerns. ByteDance subsequently rolled back this feature and introduced verification requirements for users creating digital avatars.
Rogier Creemers, an assistant professor at Leiden University specializing in Chinese tech policy, noted that the speed of AI development in China is outpacing the country’s ability to assess and mitigate potential negative impacts. He likened the situation to building a faster car, which increases both speed and the risk of a crash.
In response to the outcry, ByteDance stated its commitment to respecting intellectual property rights and strengthening safeguards against unauthorized use. Popular Chinese app RedNote has similarly restricted AI-generated content lacking proper labeling. These actions coincide with a tightening of regulations for AI content within China itself.
China’s regulation of AI is among the most stringent globally, owing to its established censorship apparatus. The Cyberspace Administration of China recently announced a crackdown on unlabeled AI-generated content, penalizing over 13,000 accounts and removing hundreds of thousands of posts. However, enforcement remains uneven, with competing platforms incentivized to prioritize user engagement over strict content control, according to ChinaTalk newsletter writer Nick Corvino.
Analysts suggest China is navigating a delicate balance between fostering domestic AI development and maintaining control over its application. While some argue that stringent regulations may hinder innovation, others point to the success of companies like DeepSeek, which have achieved competitive results with limited resources by employing alternative training methods.
The situation is further complicated by potential data access agreements. Disney recently reached a deal with OpenAI, granting its Sora video generation model access to trademarked characters like Mickey and Minnie Mouse, potentially giving Sora a competitive edge. The question remains whether such agreements will ultimately refine AI models or simply exacerbate existing inequalities.
