ABSTRACT
Compliance checking is integral within the Architecture, Engineering, and Construction sector to ensure the safety, stability, reliability, and usability of building designs. However, a significant challenge arises from the fact that building codes are mainly written in textual documents intended for human interpretation, typically by domain experts. Translating this unstructured natural language into machine-readable, semantically rich representations enables the creation of structured regulatory data that can be exchanged, interpreted, and executed by computational systems. Recent advances in Artificial Intelligence (AI), particularly in Natural Language Processing (NLP) and Large Language Models (LLMs), offer promising approaches to address this challenge. This chapter presents methods for leveraging NLP and LLMs to interpret regulatory texts and generate structured, machine-processable regulations. Recognising the sensitive nature associated with regulatory compliance, this chapter also discusses strategies for applying these AI techniques in accordance with responsible AI principles to ensure accuracy, transparency, and trustworthiness in regulatory interpretation.
