ByteDance ‘strengthening safeguards’ after Seedance AI video controversy

South China Morning Post
by Vincent Chow
February 15, 2026
AI-Generated Deep Dive Summary
ByteDance ‘strengthening safeguards’ after Seedance AI video controversy
ByteDance, the Chinese short-video giant behind TikTok, has announced efforts to strengthen safeguards in its Seedance 2.0 video-generation tool following widespread criticism over alleged intellectual property (IP) violations. The controversy centers on claims that ByteDance used copyrighted content to train its artificial intelligence model, leading to the creation of videos featuring Hollywood celebrities, Disney characters, and comic book heroes. These generated videos have flooded the internet in recent days, sparking outrage from copyright holders, including Disney, which has demanded action and reportedly sent a cease-and-desist letter to ByteDance. The issue gained global attention when Disney addressed concerns over Seedance's AI capabilities during a shareholders' meeting. The company highlighted its efforts to protect its IP and stressed the importance of ethical AI use in content creation. In response, ByteDance stated it is reviewing its policies and implementing stricter measures to prevent unauthorized use of copyrighted material in its AI systems. This move comes amid increasing scrutiny over generative AI tools and their potential to infringe on creative works. The controversy raises significant questions about the responsibilities of tech companies in ensuring their AI models do not misuse or replicate copyrighted content. For creators, artists, and media giants like Disney, this issue is particularly critical as they seek to protect their intellectual property in an era of rapidly advancing AI technology. The outcome could set a precedent for how similar tools are developed and regulated globally. While ByteDance's commitment to strengthening safeguards addresses immediate concerns, the broader implications remain unclear. Critics argue that relying on事后 filters or restrictions may not fully resolve issues related to AI-trained models potentially reproducing copyrighted material. As generative AI continues to evolve,
Verticals
worldasia
Originally published on South China Morning Post on 2/15/2026