Governing artificial intelligence safety in healthcare: developing a learning architecture for regulation, investigation and improvement
This projects aims to analyse how artificial intelligence (AI) safety should be regulated and governed in healthcare and how those systems of governance should be able to adapt over time to accommodate future developments.
Duration: Awarded 2018
Funder: Wellcome Trust
Partners:
CHILL investigators:
Professor Carl Macrae
Research summary
Background:
Artificial intelligence (AI) has enormous potential to improve the safety of care that patients receive by, for example, improving diagnostic accuracy and treatment planning. But AI also has the potential to introduce a range of new risks to patient safety, such as making decisions that are difficult to understand or check. To maximise the benefits of AI in healthcare it is critical that the patient safety risks it poses are robustly regulated and governed.
Design and methods:
I will analyse how AI safety should be regulated and governed in healthcare and how those systems of governance should be able to adapt over time to accommodate future developments. The types of patient safety risks posed by AI will be analysed, along with the key regulatory functions needed to effectively govern those risks.
A series of workshops will bring together key stakeholders to help develop this proposed governance framework and plan future research.
