Name in original language
行政通則法AI利活用ガイドライン
Initiative overview
Japan faces an unprecedentedly rapid decline in its working-age population, creating a significant challenge in ensuring sufficient staffing for administrative functions. Leveraging AI to streamline operations and efficiently handle complex and advanced administrative tasks has become inevitable.
On the other hand, risks associated with AI such as mis/disinformation, hallucinations, and black-box operations have been highlighted. Given that administrative services and procedures offer limited choice to citizens (unlike the private sector, citizens cannot select alternative providers), any implementation of AI in government must prevent situations where it may compromise citizens' rights and interests or undermine administrative fairness and transparency.
Under these considerations, an expert panel convened by the Ministry of Internal Affairs and Communications examines the legal challenges surrounding AI utilization from the viewpoint of the legal system that provides for matters such as ensuring the proper conduct of administrative procedures, post-decision redress rules, and the right to request information disclosure. Specifically, the committee aims to carefully outline usage guidelines and key considerations through a soft law approach, accounting for diverse AI application scenarios. Clarifying both legally impermissible forms of AI utilisation and clearly non-problematic use cases is considered conducive to promoting smooth and appropriate utilisation within permissible boundaries while addressing associated risks.
In the future, there is a possibility that AI technology could experience revolutionary advancements beyond current projections, leading to qualitative and quantitative expansion of AI applications in government administration while gaining widespread public acceptance. In such cases, entirely new legal challenges may emerge that may differ fundamentally from current assumptions. From this perspective, it would be necessary to prepare by identifying potential future legal challenges and exploring corresponding responses, keeping in mind possible use cases that assume this qualitative and quantitative rapid expansion of AI utilization.
Furthermore, in Japan, hard law and soft law have been established across various fields such as personal information protection, information security, copyright, and education. AI-related measures based on these are to be established under the AI Act that came into effect in 2025. Comprehensive and integrated measures will be promoted under the AI Strategy Headquarters, which oversees and coordinates initiatives across ministries and agencies at a high level.
