Digital health
The AI revolution: from promise to practice
Recent research from System C, involving nearly 300 health and social care professionals, reveals a striking consensus: over 93% of healthcare workers believe that well-designed AI will help them provide better care. Dr. Jon Shaw provides an insight into the key challenges and issues.
The NHS stands at a critical juncture. Despite ministers celebrating early signs of recovery through increased activity data following the publication of the NHS 10-Year Plan, recent analysis reveals that progress has already begun to stall. Structural reforms continue to dominate the policy agenda, yet frontline staff and patients alike experience the same fragmented services and overwhelming administrative burden that have plagued the system for years. Reform alone, it seems, rarely transforms outcomes without tools that genuinely enable integration at the point of care. But something fundamental has shifted in the past 12 months. Artificial intelligence has moved from boardroom speculation to clinical reality, and with it comes a subtle but significant transformation in both public and professional attitudes. The government’s Artificial Intelligence Sector Study 2024 1
confirmed that AI adoption
has more than doubled across sectors in a single year, driven largely by advances in generative AI. In healthcare specifically, progress has been uneven but visible - from predictive analytics in emergency departments and diagnostic imaging, to ambient voice pilots that capture clinical conversations in real time. The question is no longer whether AI will play
a role in healthcare, but how we ensure that role genuinely serves the needs of clinicians and patients, rather than adding to existing burdens.
The problem-first imperative Recent research from System C involving nearly 300 health and social care professionals reveals a striking consensus: over 93% of healthcare workers believe that well-designed AI will help them provide better care. Yet this optimism comes with a crucial caveat. Healthcare professionals are not calling for AI at any cost. They want technology that is purpose-built, ethically grounded, seamlessly integrated and designed to complement human expertise rather than replace it. The core message from the frontline is
56
www.clinicalservicesjournal.com I May 2026
that AI must be a solution to a problem, not a technology in search of one. For too long, healthcare has been subjected
to well-intentioned digital initiatives that were technically impressive but operationally impractical - systems that required duplicate data entry; applications that didn’t communicate with existing platforms; and innovations designed by people who, as one clinician in the report observed, “may have no clinical grasp right now” of contemporary workflows. The most impactful applications of AI in healthcare are those that focus on very specific, real-world challenges faced by practitioners daily. According to the research, the challenge that matters most is clear - 86% of healthcare professionals identify AI’s most valuable role as streamlining repetitive tasks like documentation and referrals, freeing up more time for patient- focused activities. This vision of AI working quietly in the
background to handle routine administrative processes, while clinicians retain full oversight and decision-making authority, represents the sweet spot that healthcare professionals are seeking.
The non-negotiable of ethics and governance If enthusiasm for AI’s potential is widespread, so too is concern about its risks. An overwhelming 97% of healthcare professionals worry that AI designed without compassion and ethical principles risks undermining human judgement. This near-unanimous concern highlights that the stakes are uniquely high because decisions guided by AI can have immediate and profound consequences for patients. AI doesn’t think, it predicts - and at times, it
hallucinates, generating plausible but factually incorrect information. This characteristic represents a significant barrier that must be overcome before widespread clinical adoption becomes possible. The solution, however, is not to abandon AI, but to ensure that human involvement remains non-negotiable. This leads directly to questions of safety
and regulation. The NHS is developing a sophisticated, multi-layered approach to AI governance. AI tools that have direct impact on clinical decision-making are increasingly being classified as medical devices, subjecting them to rigorous standards of clinical validation and
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60
