AI is evolving so fast it eludes definition. The potential impact of the field is barely understood, even by those working in it. ‘Move-fast-and-break-things’ practitioners are deploying AI systems in autonomous vehicles, in courts, in medical diagnosis, and now even at the heart of the US federal government.

Few of the constraints that govern individual and corporate behaviour are being applied to the field. Large corporations are shaping the sector faster than governments can act. In a society where few have a useful understanding of the technology, neither market signals or social norms can steer how good AI systems are produced, and harms prevented. But experts and institutions are proposing systems that might professionalise the sector, establishing best practices and avoiding harms.

In this episode, Fordham Law School’s Chinmayi Sharma shares her proposal for a ‘Hippocratic Oath for AI’, that could bring the same professional duties to AI developers as are followed by surgeons. Peter Bannister, whose business Romilly Life Sciences supports the development of cutting edge medical devices, explains how his institution, the IET, is already shaping a professional approach to this technology. And Humboldt Prize-winning pharmacologist David Colquhoun shares a story from early in his career that illustrates the importance of ongoing monitoring of adverse effects of innovative products.

Guests

Chinmayi Sharma, Associate Professor, Fordham School of Law

Peter Bannister, MD, Romilly Life Sciences

David Colquhoun, Fellow of the Royal Society, Honorary Fellow, UCL

References

AI’s Hippocratic Oath, Chinmayi Sharma

Code, Lawrence Lessig

We Need a Building Code for Building Code, Carl Landwehr

Digital Empires, Anu Bradford

The Application of Artificial Intelligence in Functional Safety, The IET

Responsible Handover of AI, Sense about Science

DC’s Improbable Science, David Colquhoun

The post #316 What Can AI Engineers Learn From Medical Professionals? first appeared on Engineering Matters.