Artificial intelligence (AI) refers to the capacity of a computer to perform operations analogous to learning and decision making in humans. Machine learning is one advanced application of AI concerned with developing computer programs that automatically improve with experience. Although it has existed for more than 50 yr, there has been recent interest in the potential for applying machine learning and big data (in combination) to clinical medicine, including anaesthesia and critical care. 1 This would have important implications for the practice of anaesthesia and, even sooner, for the training of anaesthetists. Machine learning and medical procedures Of the many aspects of medicine that might benefit from machine learning, its application to improving the performance and outcome of medical procedures represents a special case. First, medical procedures are the source of a significant proportion (13e45%) of medical errors. 2e4 Given that medical errors occur commonly and account for much death and disability 5,6 , improving procedural healthcare is a high priority. Second, advances in robotics, navigation systems, sensorised instruments, and image-guided interventions provide ready-made sources of data that could be used to 'train' a machine or decision support software. Third, the paradigm shift to competence-based training in medicine has resulted in the codification of observable behaviours that make up the procedure into specific steps and errors, which define proficient performance. The An example: ultrasound-guided peripheral nerve block Consider ultrasound-guided peripheral nerve block as a possible subject for application of machine learning. Video of