On September 21, 2026, the 19th BESETO Law Conference was successfully held at Seoul National University School of Law. An important platform for academic exchanges among Peking University Law School, Seoul National University School of Law, and the University of Tokyo Faculty of Law, the conference has been hosted by the three institutions in rotation since 2007. Under the theme “Legal Issues of Artificial Intelligence,” this year’s conference brought together more than ten scholars from the three law schools for in-depth discussions on tort liability, civil litigation, copyright, international law teaching, and AI legislation in the age of artificial intelligence.

The opening ceremony was chaired by Prof. Dongeun JOH, Vice Dean of Seoul National University School of Law. In his welcome address, Prof. Hyeok Joon RHO, Dean of the School, recalled the longstanding tradition of the BESETO Conference and warmly welcomed colleagues who had traveled from Peking University and the University of Tokyo. Prof. Guo Li, Dean of Peking University Law School, reflected on the strong institutional ties linking Peking University with Seoul National University and the University of Tokyo. He affirmed the conference’s important role as a platform for intellectual exchange and talent development across the three institutions and expressed the hope that participants would engage fully and deepen their dialogue. Prof. Shusuke KAKIUCHI, Vice Dean of the University of Tokyo Faculty of Law, recalled the insightful discussions on the “risk society” at the previous conference hosted by Peking University. He observed that this year’s focus on AI raised new questions across different fields of law and prompted reflection on the legal profession and legal scholarship themselves. He looked forward to comparisons among the three jurisdictions that would reveal both commonalities and differences.

The first session, chaired by Prof. Guo Li of Peking University Law School, focused on tort liability in the age of AI.

Prof. Youkang KO of Seoul National University presented “Ordinary Causation for Extraordinary Machines: Against Causal Exceptionalism in Generative AI Liability.” Taking the U.S. case Raine v. OpenAI as its starting point, the presentation examined four reasons advanced by scholars for viewing AI as a challenge to causation: the difficulty of tracing the causes of malfunctions (the black box problem), the stochastic nature of outputs, intervening human conduct, and the involvement of multiple actors along the value chain. He argued that existing legal doctrines could address all four issues.

Prof. Taro NAKAHARA of the University of Tokyo delivered a presentation entitled “Artificial Intelligence and Tort Liability,” analyzing Japan’s Guidance on the Interpretation and Application of Civil Liability in the Utilization and Application of AI (Version 1.0), issued in April 2026. Based on Article 709 of the Japanese Civil Code, the Guidance distinguishes between AI that assists humans and AI that replaces them. Drawing on the Guidance’s hypothetical cases involving delivery route optimization, visual inspection, assistance with legal services, transaction screening, image generation, and customer service agents, he examined how this distinction operates and offered his observations.

The second session, chaired by Prof. Akira KAMO of the University of Tokyo, addressed AI in judicial proceedings and copyright protection.

Prof. Shusuke KAKIUCHI of the University of Tokyo presented “AI in Civil Litigation: How and to What Extent May Courts Use AI?” He outlined recent developments in the digitalization of civil justice in Japan and assessed the permissibility of AI use at four stages: gathering information, processing information supplied by the parties, forming judicial convictions, and drafting judgments. He argued that relatively flexible use could be permitted in tasks such as legal research, whereas AI should not be allowed to review all case materials, predict outcomes, and draft the reasoning for judicial decisions.

Assistant Professor Bian Renjun of Peking University presented “Recognizable Human Input, Copyrightable AIGC.” The presentation explored two questions: whether AI-generated content can receive copyright protection, and whether the public can discern the “human hand” in the finished product. She proposed shifting the focus from the “resulting work” to the “creative process” and taking the recognizability of human input in the finished product into account when determining copyright protection.

The third session, chaired by Prof. Dongeun JOH of Seoul National University, focused on international law teaching and AI legislation in the age of AI.

Assistant Professor Zhang Kangle of Peking University delivered a presentation entitled “Of the Various Nations: Teaching International Law in the Age of Artificial Intelligence.” He argued that international law textbooks are, in essence, a mediating technology with three characteristics: their national standpoint can be identified, they have identifiable authors, and they can be studied. Large language models, as a new intermediary, make all three characteristics disappear simultaneously. Taking China’s efforts to develop the rule of law in foreign-related affairs as an example, he described institutional initiatives in talent development, language corpora, translation, and the placement of personnel. He emphasized the importance of teaching students to question and probe AI-generated answers.

Prof. Sangchul PARK of Seoul National University presented “Emulating Brussels but Recalibrating in Seoul: Challenges in South Korea’s AI Framework Act.” He outlined the divide in global AI regulation between context-specific, sectoral frameworks and unified, horizontal frameworks, explaining why South Korea’s AI Framework Act, which entered into force in January 2026, initially followed the model of the EU AI Act. He summarized lessons from South Korea’s experience of trial and error for other countries pursuing horizontal legislation and discussed recent legislative developments in South Korea, including amendments to personal information protection legislation and special legislation on AI data centers.

At the closing ceremony, representatives of the three institutions delivered concluding remarks. They agreed that the questions raised by AI were both profound and challenging, and that the BESETO Conference provided a valuable platform for the exchange of ideas. They looked forward to continued cooperation among the three law schools to advance legal scholarship in East Asia.