Manual scoring of polysomnograms (PSG) is time-consuming and subjective. Artificial intelligence (AI) offers promising opportunities to automate and assist this process. This article first explains the basics of machine learning in sleep analysis, including supervised learning and the importance of adequate data separation. It then outlines the key data quality and diversity requirements for practical application in the sleep laboratory. The importance of appropriate scoring metrics and the importance of examining classification quality in different patient populations are highlighted. Based on these requirements, an overview of current approaches to automated apnoea/hypopnoea detection is given. Despite advances in automated detection, there are still no approaches that meet the requirements to replace manual assessment. This article is intended to help evaluate current approaches and formulate requirements for future AI algorithms in sleep medicine.