Ms Emma Frost1
1University Of Wollongong, Australia
Biography:
Bio to come
Abstract:
There is a rapidly increasing number of healthcare applications for artificial intelligence (AI). Alongside this, a new field of research is investigating public support for healthcare AI (HCAI). These studies often report that people do not want AI to replace human physicians, and that people prefer ‘human-in-the-loop’ systems where physicians work with AI tools. Few studies, however, have examined how people think these ‘human-in-the-loop’ systems should work.
We conducted eight dialogue groups with 47 Australians to identify the reasons behind their judgements on the use of AI in healthcare. We aimed to identify why and under what circumstances publics want to retain humans in the loop in AI-assisted healthcare systems.
We found that most participants were conditionally supportive of the use of HCAI and observed strengths and weaknesses of both humans and AI tools. They felt that AI could improve the healthcare system by making healthcare more efficient, precise, and equitable. They saw AI as something that could enable social and technological progress and lead to the detection of currently undetectable risks. At the same time, participants wished to retain human input in healthcare systems. They felt that healthcare workers prioritised patient interests, were intuitive, acted with prudence, and provided an opportunity for embodied interaction. As AI systems are increasingly used in healthcare, decision-makers should consider how to work to the strengths of both AI and human healthcare workers to improve healthcare systems.