A Practitioner’s Response to Lord Darzi’s Vision
Earlier this year at the Alan Turing Institute, Lord Darzi of Denham delivered what may become one of the most defining healthcare addresses of the decade. In a keynote followed by an engaging conversation with Professor Cathy Sudlow (see video below), he laid bare both the challenges and the opportunities facing the NHS as it confronts the age of artificial intelligence.
This wasn’t a blue-sky speech full of distant promises. It was grounded, urgent, and brimming with real-world examples — from breast cancer screening in Sweden to antimicrobial resistance research with Google DeepMind. And yet, its impact came not just from what was said, but from who was saying it.
Lord Darzi is not a theorist. He is a surgeon, policymaker, technologist, and public servant with decades in the trenches of the NHS. When he speaks about AI, he speaks not from hype or fear, but from hard-earned understanding.
And as someone working with NHS Emergency Services in the areas of telematics, real-time monitoring, and intelligent asset systems, I couldn’t help but see his vision not only as necessary — but, frankly, overdue.
From Data Deluge to Insight: What Darzi Found
Before diving into his perspective on AI, it’s worth reflecting on his Independent Review of the NHS, commissioned earlier this year by Health Secretary Wes Streeting. Produced in a mere eight weeks (a feat Lord Darzi himself admits was “madness”), it drew from over 23 years of data across more than 60 million patient records.
The results were sobering:
- 100,000 adolescents with mental health conditions waiting over a year for an outpatient appointment.
- 100,000 children under the age of two spending more than six hours in A&E.
- Despite hiring 20% more staff, NHS productivity has collapsed — not from laziness or lack of effort, but from a decade-long underinvestment in digital and physical infrastructure.
The narrative is clear: you can’t expect 21st-century outcomes from a system running on 20th-century foundations.
Darzi’s Three Pillars of Change — Powered by AI
Lord Darzi articulates three critical “left shifts” to transform healthcare:
- Treatment to Prevention
- Hospital to Community
- Analogue to Digital
At the heart of all three? Artificial Intelligence.
Let’s unpack these, particularly through the lens of those of us delivering real-time services in front-line care.
From Reactive Treatment to Proactive Prevention
Healthcare today is still overwhelmingly reactive. You wait until you’re ill, you show up, you get seen — often far too late.
Darzi makes a compelling case that AI can flip this model on its head. With access to genomic, biometric, and behavioural data — much of which the NHS already collects — we can:
- Stratify populations by risk,
- Detect disease signals before symptoms arise, and
- Intervene early to prevent deterioration.
As someone who works on telematics-enabled monitoring for emergency services, I see this every day: predictive models that forecast ambulance wear-and-tear; alerts triggered by abnormal telemetry before a critical failure occurs.
Why not apply the same logic to people?
We already have the ability to monitor vital signs, movement, medication adherence, and more via wearable devices. When combined with AI, this enables true pre-emptive medicine. Not just telling a clinician “this patient is ill,” but saying “this patient will be ill in 48 hours — unless we act now.”
From Hospital to Community
Here, Darzi hits a long-standing frustration: Despite years of ambition to decentralise care, more NHS staff are now in hospitals than ever before.
How do we reverse that? By giving community clinicians the tools they need. AI-powered diagnostics, portable imaging, and remote patient monitoring can turn GPs, paramedics, and community nurses into diagnostic powerhouses. The potential for technology to support care in the home, not the ward, is massive.
But these tools must be embedded in workflows — not tacked on. We’ve seen this in emergency fleet logistics: a dashboard is useless unless it integrates seamlessly with dispatch systems, maintenance scheduling, and resourcing.
Similarly, we must move away from fragmented pilots and focus on integrated platforms that actually support staff. The AI can be clever — but if it’s not connected to the clinician’s everyday ecosystem, it’s irrelevant.
From Analogue to Digital
Darzi’s frustration with the NHS’s sluggish digital transformation is clear. He compares it to the introduction of robotic surgery — technologically sound, but delayed for years due to adoption challenges, cost fears, and cultural resistance.
He’s right. We’ve seen the same in telematics: systems proven to improve fleet safety sit unused because of “change fatigue” or lack of leadership support.
AI must be framed not as replacement, but as augmentation.
Darzi’s example of mammography AI, which matched or exceeded the accuracy of human radiologists, should be national news. Instead, it’s buried in trials. We should be scaling it — now. Not next year. Not in 2028. Now.
Adoption, Not Innovation, Is the Bottleneck
This, in my view, is where Lord Darzi is most on point — and where his experience as both surgeon and policymaker comes through.
The UK is not short on AI capability. It’s short on delivery capability. Or more accurately: delivery will.
As someone in the business of deploying systems to NHS Trusts, ambulance services, and community response teams, I’ve seen firsthand:
- Procurement delays that stall innovation,
- Regulatory processes that fail to keep up, and
- Cultural resistance from staff rightly worried about “yet another system.”
Darzi warns of “pilotitis” — the tendency to run endless small-scale trials with no intention (or ability) to scale. That’s real. And it’s a waste of time, money, and morale.
Instead, we need a new recipe:
- National-level commissioning models that reward real outcomes.
- Clear pathways to scale for successful innovations.
- Active involvement of clinicians and patients from day one.
- Adaptive regulatory frameworks (yes, NICE included) that focus not just on today’s safety but on tomorrow’s impact.
Engaging the Public: From Consent to Co-Design
A final, vital point: public trust is everything. Darzi argues that the public understand AI more than we give them credit for. I agree. In my work with NHS asset monitoring, people routinely accept the presence of GPS, RFID, and real-time tracking — because they see the benefit.
We must take the same approach to AI: show the value, be transparent about data use, and engage the public not just for consent, but for co-design.
His example of rethinking stroke care in London — where 32 decentralised centres were consolidated into five specialist hubs with public support — is a masterclass in change leadership.
What Next? A Practitioner’s Thoughts
The NHS’s motto could once again be free at the point of use — but only if we make it intelligent at the point of care.
If we’re serious about reimagining healthcare around AI, we must:
- Invest in digital infrastructure — not just shiny apps, but durable platforms, clean data pipelines, and secure connectivity.
- Prioritise adoption — fund change management, user training, and integration support as seriously as the tech itself.
- Create feedback loops — from emergency services to primary care, every AI deployment should generate measurable insight, not just reporting metrics.
- Streamline procurement — make it possible for Trusts to buy, test, and scale AI-driven tools in months, not years.
- Empower clinicians and SMEs — the next DeepMind may be hiding in a radiologist’s side project or a paramedic’s prototype. Let’s fund them, not frustrate them.
The NHS doesn’t need another committee. It needs courage.
Darzi is right: AI is not the solution. It is a means to an end. But it is the most powerful one we’ve got. And in a system under strain, we cannot afford to let culture, caution, or bureaucracy become the blockers.
This is the moment. We have the data. We have the tools. We have the opportunity to lead not just in using AI in healthcare, but in redefining what healthcare even is.
If we get this right, the NHS won’t just survive — it will evolve.
Let’s make sure AI is not just another missed opportunity.
Let’s make it the force that brings healthcare home, puts patients first, and gives clinicians back the time to care.