By Professor Nicos Savva Professor of Management Science at London Business School
Speaking at the recent SXSW London festival, former British Prime Minister Sir Tony Blair said the UK should embrace a future of AI doctors and nurses, or risk being left behind in the biggest upheaval since the Industrial Revolution. He went on to say that fears about artificial intelligence should be outweighed by the “absolutely transformative” impact it could have on public services like healthcare and education by saving time and money – “When I stand back and look at what AI is doing, I think we’re in the foothills of the most transformative revolution since the Industrial Revolution of the 19th century”.
This bold claim by the former Prime Minister follows hot on the heels of news coming out of Saudi Arabia, regarding the world’s first AI-powered doctor’s clinic, where Shanghai-based Synyi AI’s “Dr. Hua” is already diagnosing and prescribing treatment for respiratory ailments. It paints a vivid picture of a future where artificial intelligence autonomously cares for our health in a way that was considered science fiction only a few years ago. But while such developments are thought provoking, I firmly believe that the true, impactful role of AI in the short to medium term is not to replace human clinicians, but to powerfully augment them.
There are compelling reasons for this perspective, extending beyond the obvious concern of potential error and the severe consequences should one occur in a medical context. Equally important is the intricate ecosystem of healthcare itself. Clinicians are not merely service providers; they are key stakeholders deeply invested in their roles, particularly in the most critical and decision-intensive aspects of patient care. They will legitimately cite safety concerns when contemplating the ceding of such control, but it would be naïve to ignore the financial incentives and professional autonomy that also shape their approach.
Therefore, the integration of AI in healthcare has, quite rightly, begun in auxiliary roles. We’ve already seen AI excel in tasks such as note-keeping, drafting letters and managing paperwork – administrative burdens that often detract from direct patient interaction. This evolution will steadily progress to passive monitoring, for instance, checking prescription drug dosages, identifying potential interactions and flagging contraindications. From there, AI will transition into more active advisory capacities, suggesting diagnoses, recommending follow-up investigations, and outlining possible treatment options. Crucially, however, the final, nuanced decision will remain firmly in the hands of a highly trained human expert.
This measured arc of AI integration presents a profound “win-win-win” scenario. For physicians, it offers a pathway to significantly increased job satisfaction and reduced burnout. I am yet to meet a physician who genuinely enjoys the endless paperwork; many would willingly forgo a portion of their salary to alleviate this burden. By offloading these tasks to AI, doctors can reclaim precious time, focusing their energy on what truly matters: direct patient engagement and complex problem-solving.
Patients too stand to benefit immensely. Imagine a consultation where your doctor makes genuine eye contact, actively listens and engages in a conversation, rather than constantly typing or staring at their screen. This enhanced human connection, facilitated by AI managing the background administrative load, promises a more empathetic and effective healthcare experience. For healthcare systems, the advantages are equally compelling.
Reimbursement often hinges on the quality and accuracy of medical notes, an area where AI can deliver substantial improvements. By enhancing the precision and completeness of documentation, AI can streamline processes and bolster financial stability.
AI’s Quiet Revolution in Healthcare
Slowly but surely, however, the conversation will shift from augmentation to replacement. This transition is likely to occur first in less critical areas, such as routine follow-up visits or standard checks. More significantly, it may take hold in resource-constrained environments where the alternative to AI is not a highly skilled human professional, but rather, nothing at all. Some industry observers are worried that this could lead to a two-tiered healthcare system, where human physicians attend to the affluent, while the less privileged are left with an inferior, AI-driven substitute for the care they truly need. This, they argue, will exacerbate existing health inequalities.
Personally, I believe they are right to worry about AI exacerbating existing health inequalities, but I think they might have the story backwards. As AI continues its relentless improvement, it is plausible that at some point, perhaps sooner than many anticipate – it will surpass human physicians across all dimensions, including the delicate art of bedside manner and empathy. When this happens, perhaps the opposite scenario will unfold: the affluent world will be treated by the superior Dr. AI, while the less privileged may find themselves priced out of access to these expensive, patent-protected AI systems, and instead have to contend with the comparatively inferior human alternative. It’s a provocative thought, but one we must seriously consider as we navigate the extraordinary potential and profound ethical implications of AI in healthcare.
Professor Nicos Savva is a Professor of Management Science at London Business School and an expert in data science, using it to solve operational problems and help large organisations develop data-science capabilities. His research at LBS focuses on healthcare management, including hospital operations, regional organization of care, assessing performance, measuring health inequity, and innovation. Professor Savva’s work has appeared in leading journals such as Management Science, Manufacturing & Service Operations Management and Nature Biotechnology.
