CHINA Trends and Developments Contributed by: Dr James Luo and Angie Guo, Lawjay Partners
Balancing trademark values in algorithm-driven commerce The proliferation of AI shopping agents has recalibrat- ed the relationship between trademark identification and algorithmic recommendations, creating a new dynamic in which brand distinctiveness risks being overshadowed by automated decision-making. While these systems enhance market efficiency by matching consumer needs with relevant products, their opaque operational logic may compromise trademark auton- omy. This imbalance manifests in practices such as the unauthorised association of generic products with established brands through “dupe” marketing, which fosters consumer confusion, and the systematic pri- oritisation of selected brands through algorithmic tuning, which undermines fair competition by limiting legitimate trademark exposure. Cross-border challenges in algorithmic trademark protection With the globalisation of AI shopping agents, trade- mark protection faces dual challenges of jurisdiction and applicable law. Countries differ in recognising whether algorithmic recommendations constitute “trademark use” – some require explicit visual display, while China acknowledges trademark-associated behaviours in algorithmic data indexing as qualifying use. This divergence may lead to: • cross-border agent platforms being deemed law- ful in Country A while constituting infringement in Country B due to lack of trademark authorisation; and • difficulties in determining actual infringement loca- tions when violations occur in virtual algorithmic spaces, complicating rights enforcement. International trademark organisations are promoting harmonisation of “algorithmic trademark use” stand- ards, such as incorporating digital intermediation sce- narios into the revised Singapore Treaty on the Law of Trademarks.
Synergistic Approaches: Technological Governance and Legal Regulation
Effective trademark protection in AI shopping agent contexts requires deep integration of technical meas- ures and legal frameworks: • technological dimension: implement “Trademark Digital Filing Systems” embedding registered trademark data into algorithmic databases, ena- bling agents to verify trademark legitimacy through API interfaces and reduce infringing recommenda- tions at source; and • legal dimension: clarify liability boundaries for algo- rithmic intermediation in Trademark Law revisions, such as requiring “agent platforms to establish trademark authorisation review mechanisms and implement prominent risk warnings for unauthor- ised brand recommendations”; • industry dimension: develop industry standards for trademark use by AI shopping agents through collaboration among e-commerce platforms, rights holders, and technical service providers, unifying standards for algorithmic recommendations and infringement-determination procedures. As algorithmic technology evolves, trademark pro- tection will shift from “infringement suppression” to “infringement prevention,” using technical govern- ance to pre-emptively avoid trademark confusion risks in algorithmic recommendations – ultimately achieving tripartite benefits for brand rights, platform develop- ment, and consumer interests. AI-Driven Trademark Enforcement: Integrated Monitoring Framework The limitations of traditional manual monitoring in addressing sophisticated cross-platform infringe- ment have catalysed the adoption of AI technologies in trademark protection. This transformation manifests through three core applications that collectively estab- lish a comprehensive enforcement ecosystem. Official registration monitoring utilises AI systems to conduct global surveillance of trademark filings by extracting distinctive visual features and continuously scanning major trademark offices, including CNIPA, USPTO, EUIPO, and the Madrid System. By imple- menting customisable similarity thresholds typically
212 CHAMBERS.COM
Powered by FlippingBook