The OECD.AI Policy Navigator

Our policy navigator is a living repository from more than 80 jurisdictions and organisations. Use the filters to browse initiatives and find what you are looking for.

Building regulations copilot (BR Copilot)


Added by:   OECD analyst
Added on:   16 Nov 2026
Updated by:   OECD analyst
Updated on:   16 Nov 2026

1. Integrate dispersed laws and regulations into an AI-based database to enable fast and accurate legal searches. 2. Use generative AI technology to allow users to easily ask questions and receive answers, while providing checklists based on regulations to minimise errors and omissions. 3. Ensure consistency in regulatory interpretation, strengthen administrative transparency, and shorten processing times.

Name in original language

생성형 AI 기반 건축법령 행정 민원 코파일럿

Initiative overview

The Architecture Regulation Copilot is an AI-enabled regulatory information platform that consolidates and structures dispersed building-related laws, administrative rules, and local ordinances in Korea. It supports users in searching, understanding, and applying building regulations in a clear and systematic manner. The system structures legal data at the clause level and provides integrated regulatory information through natural language processing (NLP)-based query search and keyword search functions. It also includes address-based search features that allow users to identify applicable land-use regulations, permitted building uses, and development restrictions.

Through real-time linkage with the Ministry of Government Legislation, the platform ensures up-to-date legal information and enables integrated access to acts, enforcement decrees, enforcement rules, notices, and local ordinances. In addition, it enables responses based on public complaint cases, official Q&A from the Ministry of Land, Infrastructure and Transport, and judicial precedents to support regulatory interpretation. The initiative establishes a digital regulatory support system that enables regulators, permit authorities, professionals, and the general public to systematically access and use building regulations.

Results, outcomes and impacts

An integrated clause-level regulatory database with NLP Q&A and address-based inference lets a single query return acts, decrees, rules, notices, ordinances, precedents, and administrative interpretations. Real-time updates keep regulations current, and automated similarity detection flags overlapping or conflicting provisions in advance. These features reduce omissions, improve consistency in administrative decisions, speed permit reviews, increase operational throughput, and support better compliance outcomes. Future impact will be measured by quantitative indicators: reduced search time, fewer repetitive inquiries, and shorter administrative processing durations.

Other relevant details

Challenges and lessons learned: The main challenge was structuring dispersed building regulations at the clause level and standardising them into formats suitable for AI learning and inference. Overlapping or conflicting provisions and frequent legal amendments posed significant risks to data consistency and currency. To mitigate these risks, real-time linkage with the Ministry of Government Legislation was established, along with version control and data validation procedures to ensure reliability. A key lesson learned is that successful AI adoption in the public sector depends less on technological sophistication and more on data standardisation, institutional co-ordination, accountability, and transparency. Sustainable implementation requires inter-agency collaboration, periodic model review, and clear ethical guidelines. When these conditions are met, AI can simultaneously enhance the reliability and efficiency of regulatory governance.

About the policy initiative


Category:

  • AI policy initiatives, programmes and projects

Initiative type:

  • AI use cases/projects in the public sector

Status:

  • Active

Start Year:

  • 2024

Target Sectors:


OECD AI Principles:

—

Other relevant urls:

—