Positioning Australia as a Leader in Digital Economy Regulation – Artificial Intelligence and Automated Decision Making: Response to the Issues Paper

This submission responds to the Australian Government’s issues paper on regulating artificial intelligence (AI) and automated decision making (ADM), focusing on their use in healthcare and in health and medical research. Research Australia describes current applications, including smartwatch heart monitors, fall monitors, closed-loop insulin pumps, AI-enhanced imaging and prescribing decision support, and emerging risks: models that do not generalise to new populations, such as a sepsis prediction tool that achieved 63% accuracy against a reported 76–83%, alert volumes that shift with case mix, standalone clinical software that escapes device regulation, and automation bias. It maintains that because health practitioners, products and settings are already closely regulated, existing bodies are best placed to regulate AI and ADM within a national safety framework with common principles. It identifies roles for the Therapeutic Goods Administration through its software as a medical device reforms, the Australian Commission on Safety and Quality in Health Care through standards such as the digital mental health standards, the National Health and Medical Research Council (NHMRC) for clinical guidelines and the National Statement on Ethical Conduct in Human Research, the Australian Health Practitioner Regulation Agency and accreditation bodies for practitioner competencies, professional colleges, and human research ethics committees. It recommends these bodies be resourced to regulate AI and ADM in their own areas without duplication, applying the Australian Government’s AI Ethics Framework and the OECD/G20 AI principles.