TAIPEI, July 20, 2023. On June 27, the Judicial Yuan and the Taiwan Constitutional Court (TCC) welcomed the visit of Professor Dr. Aziz Huq (Frank and Bernice J. Greenberg Professor of Law at the Law School of the University of Chicago), who was accompanied by Professor Dr. Ching-Yi Liu (Director and Professor of the Graduate Institute of National Development, National Taiwan University) during this visit. On behalf of the Judicial Yuan and the TCC, President of the Judicial Yuan, Chief Justice Tzong-Li HSU, expressed sincere welcome to Dr. Huq and Dr. Liu upon their meeting. The meeting was accompanied by Vice President Jeong-Duen TSAI, Justice Jiun-Yi LIN, Justice Jui-Ming HUANG, Justice Sheng-Lin JAN, Justice Ming-Yan SHIEH, Deputy Secretary General of the Judicial Yuan Mr. Lin-Lun HUANG, Director General of the TCC Mr. Chen-Chou HSU, Director General of the Department of Information Technology Mr. Wu-Chih LAI, and Dr. Chien-Chih LIN (translator of the event; Associate Research Professor at the Institutum Iurisprudentiae, Academia Sinica.) 

    During his visit, Dr. Aziz Huq lectured on "Due Process for AI Decision Tools," with a special focus on the normative arguments for the fundamental rights of human decisions concerning the application of AI. Dr. Huq started his speech with an introduction to current major legal sources toward AI in the world, which are Article 22 of the General Data Protection Regulation (GDPR) of the EU and the AI Bill of Rights of the US. Both said regulations recognize the right to a human decision, which is the right not to be subject to a decision based solely on automated means. Dr. Huq discussed three common reasons providing the normative ground of said right, namely: "Accuracy," "Dignity," and "Fit and compatibility." Dr. Huq argued that the first and the second reasons, accuracy and dignity, are not suffice in supporting the normative ground for the right of a human decision. In terms of accuracy, it is hard to argue that AI provides less accurate decisions than human beings. In terms of dignity, considering the fields where the application of AI decisions have been common nowadays, e.g. online medical service, elderly care, and distribution of social services, which are relevant to the handling of sensitive data and the rights of vulnerable groups, it is hard to argue that applying AI in other areas would definitely violate human dignity. Dr. Huq concluded his speech by arguing the only plausible reason for the normative foundation of the right of a human decision is the "fit and compatibility" when putting the limitations of AI decisions into consideration. He elaborated that training data for AI apparently influence AI’s accuracy in making an evaluation. In certain circumstances, a trained individual can easily make evaluations with better quality. Although it is easy to point out abstractly the areas where AI decisions are limited, it is hard to give a specific description of areas where the application of AI is "fit and compatible."

    After the speech, Justice Sheng-Lin JAN and Justice Ming-Yan SHIEH joined the discussion by expressing their views on the possibility and limitations of AI application in adjudications, even constitutional judgments. Dr. Huq noted that the interpretation and application of law involve high value judgments, making AI hard to replace judges.