As DeepSeek’s breakthroughs in the field of AI have led to a plunge in U.S. technology stocks, debates over the differences between Eastern and Western intellectual property strategies have intensified.
OpenAI, relying on U.S. protections for software copyrights and trade secrets, accuses DeepSeek of misappropriating intellectual property. DeepSeek, however, operates under China’s continuously developing intellectual property regime and may argue that its conduct falls within the scope of “fair use” or legally permissible data scraping.
This divergence in legal interpretation is exacerbating the trans-Pacific tug-of-war over intellectual property. As emerging Chinese AI enterprises and U.S. competitors continue to vie for market dominance, this tug-of-war is only likely to intensify. How, then, should the AI industry seek a common path forward to achieve greater synergy and mutual benefit?
The Impact of the U.S. and Chinese Intellectual Property Frameworks on the AI Race
The differing perspectives of the United States and China on intellectual property protection have led to fundamental divergences between their respective legal systems.
The United States has consistently adopted a “light-touch” regulatory approach, prioritizing innovation and market growth. This approach is typically manifested in reliance on industry self-regulation, with voluntary guidelines and frameworks at its core.
Although agencies such as the Federal Trade Commission (FTC) actively combat unfair and deceptive practices related to AI—for example, the FTC took action against five AI companies last year—the overall regulatory environment remains relatively relaxed, providing space for the rapid development and deployment of AI technologies.
By contrast, China has adopted a more proactive and interventionist approach to AI regulation, placing greater emphasis on state supervision and guidance. The Chinese government regards AI as a key strategic technology for achieving economic objectives and has been actively promoting its development and application. This proactive stance is also reflected in China’s evolving intellectual property system, which includes specific regulatory guidance on generative AI, data scraping, and other AI-related activities.
Although this legal framework is still being refined, it offers numerous potential advantages for innovators. For example, China’s legal system recognizes the importance of agreements in determining the ownership of copyright in AI-generated works. By emphasizing the execution of formal contracts, innovators can clarify in advance the ownership and use rights of intellectual property, thereby obtaining legal certainty and protection.
In addition, Chinese courts have taken a proactive approach in interpreting the Copyright Law and applying it to the assessment of the copyright attributes of AI-generated outputs. In particular, cases such as Li v. Liu (the 2023 “Spring Breeze Brings Tenderness” case) and Lin Chen v. Hangzhou Gaosi Co. and Changshu Qinhong Co. (the 2024 “Banxin” case) have provided valuable guidance and precedents for the copyright protection of AI-generated outputs, offering direction and reference for innovators in the complex and rapidly evolving AI field.
This fundamental difference in approaches to intellectual property regulation in the AI field lies at the core of the trans-Pacific tug-of-war over intellectual property, and it also creates numerous challenges for emerging enterprises from the East and West when they attempt to enter each other’s markets.
Challenges in Cross-Border Enforcement of Intellectual Property Rights
Enforcing intellectual property rights across borders in the AI field is a daunting task. Differences in legal frameworks among countries such as the United States and China make this challenge even more complex.
These differences involve not only divergent interpretations of specific intellectual property rights, but also broader issues such as data privacy, competition law, and even the fundamental legal philosophies underlying intellectual property protection.
One of the greatest difficulties lies in the differing standards for determining infringement. For example, when examining data scraping conduct under China’s Anti-Unfair Competition Law, multiple factors must be considered comprehensively, including whether a competitive relationship exists, the nature of the scraped data, and the potential impact on market competition. Such assessments are often subtle and complex.
Furthermore, in the recently published Chinese case of Shanghai Xinchuanhua Cultural Development Co. v. Hangzhou Shuimu Intelligent Technology Co. (the 2024 Hangzhou Ultraman case), the court held that, when assessing whether the data input and training of an AI model constitute copyright infringement, a relatively relaxed and inclusive standard should be adopted, thereby making “fair use” a viable defense. This approach stands in stark contrast to that of the United States, which typically adopts a stricter stance and often regards data scraping as an infringement of copyright or trade secrets.
In addition, evidentiary issues further complicate cross-border intellectual property disputes. Collecting evidence of infringement in different jurisdictions often presents both legal and procedural challenges. Differences in procedures for evidence exchange (for example, the presence or absence of discovery), data privacy rules, and even language barriers may all impede the collection and admissibility of evidence.
The complexity of jurisdictional issues is equally significant. In cross-border intellectual property disputes, determining which court has jurisdiction is often contentious. This is particularly true in cases involving online infringement, where infringing acts may span multiple jurisdictions.
This lack of uniform standards further contributes to uncertainty and inconsistency in the cross-border enforcement of intellectual property rights. This naturally raises the question: is it possible to establish a unified global framework for AI-related intellectual property protection?
International AI Agreements or Industry Standards
Given that the United States and the United Kingdom failed to sign the Paris Summit declaration on inclusive and sustainable development of artificial intelligence, it is evident that the establishment of truly international standards on a global scale will still take time.
However, to address the challenges of intellectual property protection and enforcement and to promote the sound development of the global AI industry, it is essential to advance the conclusion of international agreements and to formulate common standards for AI-related intellectual property.
These initiatives include harmonizing the definitions of key terms such as “fair use” and “data scraping,” establishing clear guidelines and procedures for cross-border enforcement of intellectual property rights, creating effective dispute resolution mechanisms, and, through bilateral and multilateral AI governance, reaching consensus on ethical norms, safety measures, and transparency requirements for advanced AI technologies.
Industry-led best practices and standards can also play an important role in bridging the gaps between different legal frameworks. Cooperation between AI companies in the United States and China can help countries build consensus on intellectual property protection, thereby fostering a more cooperative and innovative environment that benefits all industry participants.
Resolving the trans-Pacific tug-of-war over intellectual property in the AI industry requires the joint efforts of governments, enterprises, the legal community, and international organizations, and this is clearly no easy task. Yet, by striving to establish clear international standards and frameworks, we can ensure fair and effective protection of cross-border intellectual property rights while unlocking the potential of AI for the benefit of the world.
