The government’s ten-year health plan puts technology at the heart of modernizing the NHS, identifying AI, data, wearables, genomics and robotics as critical enablers.
The language is deliberately bold: the plan frames these as “big bets,” an unusual term for a service that has traditionally prized predictability. However, the accumulated pressures of growing demand, labor constraints and aging infrastructure make this all but unpredictable.
In early March 2026, the New Statesman held a roundtable in partnership with Medtronic to explore what it will take to turn this ambition into practice. Policy makers, clinicians, patient representatives and industry leaders gathered to discuss how digital technologies can be developed and used to improve outcomes, experience and efficiency – for patients and the people who care for them.
The discussion was held under Chatham House rules, which means that individuals and organizations are not named. It has been summarized for length and clarity.
The discussion started with a provocation. An industry speaker argued that there is a growing gap between the technology people use in their everyday lives and what is available on the NHS frontline.
Physician workplaces are still shaped by slow entrants, fragmented systems and outdated processes. “We have incredible people in our public services in the NHS,” they said, “but often they are doing their jobs in spite of technology, not because of it.”
The examples were many. One policymaker noted that the debate around health technology too often focuses on what it does for the patient, without considering how it supports the clinician. They gave the example of appointment booking systems that remain extremely rigid, a problem that ultimately falls back on patients.
Another pointed out that some of the most basic infrastructure is not yet in place: good Wi-Fi is not universally available in GP surgeries, for example. Some participants argued that before the NHS can meaningfully embrace AI, it needs to get its digital foundations right first.
However, the foundations are not all missing. The UK’s centralized health system holds an incredible volume of data, which many participants claimed was a real asset by international standards. One participant noted that GPs are already using AI-driven tools to triage patients through online questionnaires and that the appetite for innovation at the local level is real.
However, a major hurdle is connectivity. Clinicians still cannot access most of the information held elsewhere in the system that would make their work more efficient. “There’s a real bottleneck that we just have to get rid of. We have to have a unified system,” noted one doctor.
This echoed across the table. One industry representative remarked: “Although the NHS is incredibly data-rich, I would argue that it is knowledge-poor,” explaining that fragmented systems and poor usability mean that the potential for learning at scale has never been realised.
Doctors themselves identify the poor usability of current tools as the main barrier to greater use of AI – not a lack of ambition, but a lack of workable systems. Building what one person described as genuine “learning health systems” will require not just better data, but better infrastructure to use it.
If NHS data is to become the engine of transformation, the room was clear that trust must also be at the forefront of the discussion. One participant warned that health care data can often carry bias—whether based on ethnicity, gender, or geography—and that automating flawed processes can risk entrenching those biases. “Once we automate those biases, it’s going to be even harder to back off, to say that’s how you came to this conclusion,” they said. The call was for rigorous work to understand and address bias before systems are deployed at scale.
To build this trust, there was broad agreement that digital transformation should be with patients, not with them, and with clinicians, not around them. “If technology is not co-produced with patients and physicians within the system, it will never be adopted,” noted one participant. “It will never shrink. It won’t even fit in the front door.”
Several voices emphasized that patient groups and frontline staff should be meaningfully involved in the design and governance of new technologies, not consulted as an afterthought.
Another recurring theme was the challenge of scale. “There is no shortage of innovation happening in the NHS,” said an industry representative, but the system’s capacity to adopt new technologies across individual trusts or regions remains very limited. One participant responded that the health service has many pilots that never progress – “pockets of excellence that fail to become standard practice”.
Ambition, they said, should shift from running individual pilots to creating replicable models that can be adopted across the system. Without that change, resources are spent on small-scale gains that never reach a “critical mass” of patients.
Part of the solution, some attendees argued, lies in how partnerships between the public and private sectors are structured. The NHS already relies on private technology, so the question is not whether to partner, but how to do it better. One participant called for clearer expectations about what “responsible and ethical profitability” looks like, so that companies entering this space understand the guardrails from the start.
If the process, patient involvement and benefit can be clearly demonstrated, the model can work. “We need to make sure that if we spend money, it’s really well spent,” noted one participant.
One of the most surprising parts of the roundtable was when the discussion turned to the patients themselves. Some attendees challenged the assumption that the public is resistant to change or unrealistic in its expectations. “I think sometimes people are afraid to talk to patients because they think patients are going to say, ‘I want red roses every time I’m in the hospital.’ They don’t. They know how the system works,” noted one doctor. Patients want to be real partners, understand the risks in healthcare and AI, and have honest conversations about what is and isn’t possible.
There is already practical evidence of what this might look like. Work is underway, with consent, to use the data locally to identify people at risk of falls and other frailty-related health issues.
When the goal is clear and the benefit tangible, trust follows, the table is heard. “The public will have confidence if we work with them and start with that very foundation: ‘What do you need to live your life to the fullest?’ said one participant. Another added that if systems are built around the needs of the people who use them — patients and doctors alike — adoption will naturally follow.
Ultimately, the participants were all cautiously optimistic. The combination of political will, technological capabilities and clinical needs creates a real window of opportunity. But that window won’t stay open indefinitely.
The ten-year plan must translate ambition into delivery – not through another generation of isolated pilots, but through a sustained commitment to building reliable and usable digital systems at scale.
The challenge, as one participant put it, is to bring the same security-first discipline that drives other high-stakes industries to healthcare’s digital transformation. What is needed is a common direction of travel: one that aligns policymakers’ ambitions with frontline realities; that builds the patient’s trust from the start; and that treats digital transformation not as a speculative bet, but as essential infrastructure for a future-proof health service
