The rapid advancement of artificial intelligence (AI) has touched nearly every facet of modern life, including copyright management. As this technology evolves, it raises fundamental questions about its interaction with copyright law and the systems designed to protect creators. While collective copyright management organizations (CMOs) play a significant role in managing and licensing rights for creators, broader discussions at the Copyright Collective Management Summit 2024 reflected the multifaceted efforts needed to address AI's disruptive potential.
Recently held in Beijing by the China Audio-Video Copyright Association (CAVCA), the summit brought together global experts from copyright organizations, entertainment industries and legal academia to tackle multiple pressing issues, including the development and refinement of collective management systems, the opportunities and challenges generative AI brings to copyright management, and the implementation of copyright-related laws. The event stood out for its inclusivity, offering a platform for voices from diverse sectors to explore collaborative approaches.
Central to the AI-related debates were two key questions: how to regulate AI's use of copyrighted materials during training and whether AI-generated content qualifies for copyright protection.
The expanding role of collective management
Collective copyright management refers to the centralized system where organizations, known as CMOs, manage and license copyright on behalf of creators. These organizations act as intermediaries, streamlining licensing for large-scale usage and ensuring fair compensation for rights holders.
"We act as a bridge," explained Zhou Yaping, director general of CAVCA. "Without collective management, many copyright laws would remain unenforceable in practice. AI developments only underscore this necessity."
Discussions at the summit underscored the importance of CMOs in adapting to technological changes. With the increasing reliance on AI and digital platforms, CMOs have the potential to simplify licensing for AI training datasets and enhance rights management through advanced tools like blockchain. This adaptability positions CMOs as critical players in balancing innovation with protection.
AI's disruption of copyright
The summit highlighted two major disruptions AI brings to copyright. First, AI systems require vast amounts of data for training, often sourced from copyrighted materials. This raises questions about consent and fair remuneration for creators whose works are used in datasets. Abbas Lightwalla, global legal policy director at the International Federation of the Phonographic Industry, emphasized the urgency of strong copyright rules to address unauthorized voice cloning and deepfakes. He suggested that generative AI service providers be mandated to disclose training content, ensuring transparency and respect for copyright holders' rights.
Second, the copyrightability of AI-generated content remains contentious. National Judicial Expert and World Intellectual Property Organization arbitrator Chen Jinchuan explored the legal nature of AI-generated works, arguing that current laws do not recognize AI users as authors under the input-output framework. He further noted that AI-generated content often fails to meet fair use criteria or pass the three-step test mandated by international treaties. To address these challenges, Chen advocated for compensation schemes or statutory licenses managed by collective management organizations.
China's recent legal reforms aim to address these challenges. Provisions now require explicit permissions and fair compensation for using copyrighted works in training datasets. Zhang Hongbo, executive vice president of the China Written Works Copyright Society, highlighted the importance of defining statutory licensing conditions and ensuring rights holders' access to compensation, even when identities are unknown or untraceable. He noted that such measures must balance protecting creators' rights with fostering innovation to drive high-quality industry development.
Collaborative solutions and global lessons
The summit featured a roundtable discussion titled "AI Copyright Issues and Collective Management," where industry experts from various sectors shared their perspectives on the intersection of generative AI and copyright protection. The discussion included prominent figures, such as Guo Biao, president of the Greater China Region and chief representative in China for the International Federation of the Phonographic Industry (IFPI); Zou Xiaoman, co-founder and chief operating officer of the "Kanjian" platform; and Fan Zhihui, founder of media matrix "Music Herald." The roundtable was moderated by Zhu Yanzheng, deputy secretary-general of the China Music Copyright Association (CMCA).
The panelists agreed that AI technology presents both opportunities and challenges for the copyright sector. They emphasized that while the creative content industry must embrace new technologies, legislative protections are essential. As CMOs bridge the gap between vast numbers of rights holders and users, they are uniquely positioned to navigate these complex issues and play a pivotal role in managing AI's impact on copyright.
Globally, the European Union's opt-out mechanism for training datasets and Japan's lenient policies offer contrasting approaches. Experts at the summit suggested that China, with its robust CMO infrastructure and legal reforms, could lead by example. The combination of centralized licensing and government oversight provides a framework that could inspire international harmonization.
Challenges and the road ahead
Despite these advancements, significant challenges remain. Transparency in AI's use of copyrighted materials continues to be a pressing issue. Additionally, ambiguity surrounding the ownership of AI-generated content creates legal grey areas that need resolution. Multiple panelists emphasized the urgent need for clearer definitions and standardized practices during the summit.
The summit concluded with a call for continued collaboration across sectors. By integrating insights from global practices and fostering innovation, China's approach to collective copyright management could serve as both a model and a catalyst for addressing the complexities of AI in the copyright landscape.
"This is not about resisting technology," said Zhou, "It's about shaping its future responsibly."