The Doctor Will See You...Online: Caroline Clarke on Building a New NHS Trust

Tom: Welcome to DrDoctor Will See You Now, the podcast where we cover the latest in tech, ops and movement across the NHS. I am delighted to have join me today Dame Caroline Clarke, most recently appointed - I think it's week one as chief...

Caroline: Week two, technically.

Tom: Okay, week two. So we've caught her at the end of week two, and it's been a fortnight, so hopefully we'll have a little bit of goss on the new NHS Online Hospital. Caroline, welcome.

Caroline: Thank you very much for having me.

Tom: Thank you so much for joining us. For people who don't know you, would you mind introducing yourself?

Caroline: Right. My name is Caroline. I am Welsh. I'm a North Londoner - I've lived in North London for a long time, which caused a few issues in my last job, because of course I was representing the NHS for the whole of London, and I disproportionately spent more time in South London than North London. Now, of course, that's all irrelevant because I'm doing a national job, so I have to put that to one side. I have a partner, Helen, who's a psychotherapist and acupuncturist, which is highly relevant to this actually in some ways, and a daughter, Esme, who is 16, and multiple animals.

Tom: Amazing.

Caroline: And I was chief exec at the Royal Free. I was a finance director before that, and have done most of my time in the NHS, with a little bit of time out of it.

Tom: So about 35 years in the NHS?

Caroline: That makes me sound very old, but yes, that's entirely correct. I'm 57.

Tom: So you've been around about as long as the first NHS website. That's mad.

Caroline: Apparently, according to my sources.

Tom: So, 35 years - from finance director to founder. What an arc, and what a story. I'm really interested to know how, as an accountant, you ended up leading one of the leading trusts in the country. Tell me about the journey in those first few years as an NHS leader.

Caroline: Yeah, so look, I wasn't a very good accountant. I did it because I thought it was a tool. I did an economics degree, went to university at the end of the '80s and had a lot of fun - you're not old enough to remember this, but it was fun. Music, dancing, all that stuff. And I came out as gay during that time as well. So I didn't get a brilliant degree, but I did learn a lot about what motivates people. I'm really interested in how systems work and how people work. I wasn't good enough to be an academic, so I thought, God, I need a trade. So I did accountancy - literally, because I'm not very practical and I'm quite numerate, so I thought I'd better go into it. I did CIPFA, which is a government accountancy qualification, and it's as much about policy and economics and what drives government thinking as it is about technical accounting. I got onto the NHS training scheme, mainly because the people who interviewed me were lovely and I thought they were gay, and I thought this would be the best place to come out - they weren't, I got that wrong, no gaydar then and now. But it was a nice time, the days they used to run assessment centres - you had to go to a place in Chingford where Norman Tebbit was the MP and everyone railed against him, it was an interesting time. I chose that over going into the City or being an auditor, because I felt I wouldn't be a terribly good auditor - it's very formulaic, and I wanted something a bit more creative and strategic. And like so many of us, at your core you care about people, don't you? It's all about what makes a human being tick. For me, it's all about improving things. I remember when we worked with the Institute for Health Improvement years ago at the Royal Free, when I was CFO, and I thought, this is so good, it's kind of social science, isn't it - using data to improve, and giving people control and power over their work environment. I thought, this is it, this is the thing. I still think that is the thing, or part of it. You need big thinking and structure for some things, but largely - I don't know about you - I want to go to work and feel like I'm in control. I don't mind having guardrails and accountabilities, but generally I want to be trusted to get on with stuff.

Tom: Yeah, and you can make things happen. But I love that you said "systems thinker," because that's what I've always noticed - the people who seem to have the most success in healthcare are people who fundamentally are people people. And I love a dance as well, and also didn't do as well as I should in my degree, because I spent far too much time having fun.

Caroline: Good, I knew you were all right.

Tom: Yeah, exactly. But it teaches you about people, right? And if you can bring together a genuine empathy for the people around you and what drives them, with systems thinking and enough structure, that's how you run any organisation, I think - particularly in healthcare.

Caroline: I know there's a quote from Dido Harding, where she said running a hospital is like running a supermarket, except everyone on the tills has got a PhD, is smarter than you, and thinks they know what the right answer is.

Tom: So there's nothing like running circles around them.

Caroline: Yeah, exactly, nothing at all.

Tom: All right, well, that was humble of her. Well done.

Caroline: Yeah, it's true, that is absolutely true. And I suppose there's something about - that's all important - but also this thing about relationships, and knowing how to relate to other people, up, down, across, all that stuff, is a really important part of what makes you work. And then this thing about knowing when to lead and when to follow. You can't lead all of it, and not everybody will listen to you. Having a little bit of humility - I know humility is very fashionable in leadership, but there is something about all that that's quite important.

Tom: There is. On that point, how did you find moving from a very successful chief exec of a trust that went through COVID, and so on, to becoming a regulator?

Caroline: Yeah, that's really interesting, isn't it - you should probably ask other people. I loved being at the Royal Free, I really liked being in the delivery organisation, I was really, really proud to lead it. It was the first ever free hospital, founded in 1828 by William Marsden - I've got a whole patter about that, but that's probably another podcast, you should get someone from the Royal Free to come and talk to you about it. Great hospital, great things happen there, great science happens there too. I wouldn't necessarily have left when I did, but opportunities arise in your career and you go for them. I remember thinking, okay, we've done some really good technological stuff here, we've digitised a lot of our pathways using our electronic health record, Cerner - we'd got really good clinical engagement to do that. And it felt like, if you can do that for a population of 2 million, maybe you can bring some of that thinking into a London population. So I was quite excited about post-COVID - could we bring more innovation into London, could we reduce the variation we see in care, all that stuff. Quite quickly you realise that, unlike hospitals, general practice and pharmacies, this middle tier is a bit ephemeral, there's a lot of looking up and looking in, and not looking out, and that stuff's quite hard. You can see poor old NHS England's in a bit of a spin at the moment, so that's still going on, and I think it happens every time the political administration changes - there's probably some good graphs there, Tom.

Tom: I'm sure there is.

Caroline: But what I really enjoyed - every job, you don't enjoy all of it, and I always think there's a third, a third, a third: a third you're going to love, a third is always going to be a bit tricky, it's the stuff you have to do, your licence to be there, and then there's this middle third you're just negotiating with. I spent too much time negotiating the middle third, but I loved working with people like me in London who lead organisations that really care about the people they serve. I loved getting into primary care and working with different stuff, visited a lot of pharmacies. I loved the people I worked with - there were some great people in NHS England, really cool people who genuinely care about what they do. And it's been quite tricky, all these reorganisations.

Tom: Tricky is an understatement.

Caroline: Yeah, I can only imagine. So that was difficult. And of course I feel really bad for the NHS crew in London, because almost the week Keir Starmer announced the merger, or abolition, of NHS England and the DH, we'd moved the London office from Wellington House down to Canary Wharf because we thought we'd run out of space. I thought, oh, we get first mover advantage - anyway, that wasn't so good, so sorry, everyone, about that. Although Canary Wharf, actually - 10 South Colonnade, which is government - the infrastructure is pretty good, and some people prefer Canary Wharf to Waterloo. Anyway, you make these decisions in good faith.

Tom: And it feels to me, from the outside in, that through all that change, you were a really stabilising force in London.

Caroline: It's nice of you to say that. Certainly - ask everybody else. I do feel a bit... I said I'd do three years, I remember saying in the interview, I think this is a three-year gig, and then you need to refresh and get some more energy in. I had no idea what was about to happen. I'm not sure I did the job the same way as the other regions, but for me it really is two things: it's regulation - when stuff goes wrong we need to attend to it, and we lean in and use that improvement thinking - it's all about how you help people improve. Have you got the capacity and capability to do it? Have you got the right leaders, the right tech? Are we investing in the right places? And really trying to be with organisations through that - that's the regulatory stuff. I hated being regulated when it was like being clubbed with a seal, because it just doesn't - you don't get the best out of people.

Tom: Doesn't help at all, does it?

Caroline: You do have to make some tough decisions, and there are plenty of people who've been on the end of that, and I hope we've done it with dignity and care, and that we haven't leant too far into the seal-clubbing space. But I'll go and ask some other people about it.

Tom: Well, you should. But it doesn't seem that way.

Caroline: I didn't mean to. I think the thing for me is actually the reform agenda, the transformation agenda, and particularly now given what we have in terms of technology - and you know this, because you run one of these organisations, it's phenomenal and incredible - you have to have the politicians and the planning authorities, like NHS England, totally lined up to that. One of the most joyful things I got to do was get chief execs and chairs in a room and show them what's happening, and you can see everybody lean forward. I remember showing the chairs in London a version of ambient voice technology - a couple of years ago, when Penny was the ICB chair - and she got up and went round to have a look, and it was just... and you think, right, we're going to do something with that. We went on to do a pilot with one of the firms. My regret is that we haven't put it into every single organisation, but we really should.

Tom: What do you think's held that back?

Caroline: It's a spready thing, isn't it. Some of it's money, some of it's people not knowing - part of the job is showing people what's possible. Some of it's people being unclear about regulation. AVT is a good example: are we regulated, are we not, is it a device or not a device? It's been really confusing for the market - some of you say we're going to be regulated, and some of you don't. I think that's now coming to a conclusion.

Tom: Yes, I think so, I think I heard Lawrence be quite firm on that.

Caroline: Yeah, and I think it's probably good that we've had the debate, actually.

Tom: Yeah, but that debate will probably run and run for various different technologies, won't it. We need to get better at having the debate.

Caroline: I think that's my learning from all that. You may not have seen this, but one of the nicest things I got to do was work with Chris Streather, our CMO in London, and others, on a Life Sciences and Innovation Strategy - which is all about: in order to really radically improve patient care, you need a much better relationship with industry. We want to put more people through trials, more people through these emerging technologies as they get regulated and safe, and so what are the things we need to do, as the middle tier, to help.

Tom: Yeah, that's interesting. So things like...

Caroline: Well, yesterday I saw Lawrence and the MHRA launch the AI Commission. And in London, the MHRA have created a sandbox - other sandboxes are available if you're the MHRA now, but we had our own special one.

Tom: Is this part of the Innovation Zone?

Caroline: Yeah, the Health Innovation Zone. So if there are tech companies listening to this, I think the call is still open - we're trying to get around 10 products to test and see whether we can accelerate.

Tom: And if anyone's listening, who do they talk to about that?

Caroline: Just Google "MHRA London Sandbox," it's on the website.

Tom: We'll add a link somewhere.

Caroline: Somewhere. Thank you. But then there are other reasons things don't spread. There's not enough money - seed funding, working capital. Generally, the technologies we're going to be interested in are things that will ultimately reduce your operating expenditure and make you more productive, so you need something at the beginning to get through the change, and then your opex comes down - and we haven't been very good at doing that. So there are loads of different conversations in London about whether we could use our scale to work with the British Business Bank, or some of those social finance people, to provide a slug of capital for people to come in and use. So money's one thing. And then there are health passports: if you've got through the innovation governance people at Guy's and St Thomas', it should be good enough for anyone else in London, but it turns out it's not, because they're all statutory organisations - we've got to make that better. NHS England is doing some work on it nationally, and we're trying to go faster in London - there are some really good people, and our health innovation networks are all working on that. So that's a very cool thing. And there are a few other things we're doing to make procurement a bit easier.

Tom: I tell you what, those passports will do more than just ease the commercial side, because the other thing I often observe is there's more connectivity between organisations than is leveraged nationally. I can only speak for DrDoctor, but we have half the hospitals in London using us in one way, shape or form. We could connect all that data together and do all sorts of interesting things with load balancing and transfers of care - but we don't, because the problem isn't technical, it's IG.

Caroline: So using those passports in innovative ways would be really interesting. One of the other things we've done across London is create a secure data environment - primary and secondary care records all in one place, which took a lot of effort from a lot of people to get those primary care records in. Now we're trying to figure out how you give that value, how you actually let people access that data. But you're right, other forms of data are available, and we need to get into that, and work it all together.

Tom: Yeah, for sure. One of the really interesting things I observe, having gone on a startup-to-scale-up journey myself, is that I think the NHS is actually really, really good at incubating new ideas - it's phenomenal at that.

Caroline: Yeah, anyone who's worked in a hospital will know a consultant who's come up with an idea, or a nurse who's pushed something, or whatever it might be.

Tom: Where it has a gap, I think, is when you have a scaled product, or organisation, or methodology, and at some point it feels like a switch flips, and you go from being accelerated, part of the journey, to - oh, you're a supplier - and a wall comes up.

Caroline: That's exactly right.

Tom: So what do we do about that?

Caroline: Yeah, what do we do about that. Well, I'm just going to throw in my other issue - when I was at the Royal Free, there was this Friday phenomenon, which I christened "snake oil." Some really lovely young doctor would come into my room and say, I've given up some of my clinical practice and I've developed an app.

Tom: Great.

Caroline: What's the app for? And they describe a problem - I had no idea we had that problem. From their perspective it was this problem, but if you took a multi-perspective view, it wasn't. So how do I know that the thing being developed is the thing I want? Matching solutions to problems - I think there's something in us being much better at that kind of soft market-testing, where you say, these are the sorts of things I'm interested in. I've got to think about this in my new job, because we're clearly going to have to partner with lots of different people to get the scale and speed we need. I haven't got any people at the moment - this is a future thing - but you need a commercial market-test function to do that early, pre-procurement work, and then you write a much better tender.

Tom: Yes.

Caroline: You're much better at then asking suppliers to respond.

Tom: Yeah, and probably it makes the market smaller.

Caroline: I think I read the other day that London's got something like 2,500 startup and health tech firms - an implausible number, I think it was a slide designed to attract international investors to London, so that might be a bit toppy.

Tom: I can believe it.

Caroline: But I was thinking, gosh, if we do this really well, that number will halve. There's something about... and of course VCs already do some of this, by just not funding stuff and killing it.

Tom: Which they do. And on that point - I'm keen to manage this online, but you've mentioned funding, so I'm just going to ask: should the NHS fund more of this stuff itself? Should it be taking ownership, and should it be taking equity?

Caroline: I have a commercial bias, so I'd probably say yes. What are we trying to do? We're trying to create really safe products and services that patients trust, and the NHS brand is really, really valuable, and we mustn't destroy that. Then it's how you go about doing that - do you do it through owning more, which some systems do? You could argue health innovation networks were designed to do some of that, you could argue some of the ecosystem's already there - definitely some NHS trusts and foundation trusts have incubators and so on. Some brilliant ones in London, and some of the stuff places like Great Ormond Street and the Marsden are doing is really impressive. And other hospitals are available, obviously - I don't have favourites, outside London other hospitals exist.

Tom: Manchester's fantastic.

Caroline: Manchester HIN's fantastic, actually, they're doing really good work there. So love you, Manchester, and Leeds, and Yorkshire, and everywhere.

Tom: Manchester would be where I lived, if I didn't live where I live. I love it.

Caroline: Yeah, me too, great city. But they are genuinely doing really good stuff, and we look to them for leadership on some of this, I think. I don't know if it's "own" - I think some of the NHS already does own some of it, and then there's a scaling problem. I think the HINs were originally conceived with the potential to scale. In London, we've tried to get the three HINs to work really collaboratively, and they've been brilliant, actually - they're like a throuple, they're really nice.

Tom: Can be political, but ultimately the outcomes are good.

Caroline: They've been great. And trying to get them to think about how we scale across London - we touched on the Health Innovation Zone, but the work beneath it is very much sponsored by the Health Innovation Networks and is all about spread and scale. So I think that's right. There is this funding issue we haven't quite solved though: what is our relationship with funds, with VCs and equity, and how do we resolve that as a public service?

Tom: It is really tricky, isn't it. It's not as simple as, say, some American healthcare systems, which act very commercially and will take a stake.

Caroline: I don't think it is that simple. It's political, but for me it's also really about trust. If I'm a patient, I don't want to see this funded by anyone other than me, the taxpayer. We ran a really good private practice at the Royal Free, and even that was tricky - we had no relationship with any of the big private providers, it was our own doctors and nurses working there, and even that was contentious.

Tom: It's interesting, isn't it - it's like the porters would say, well, I'm not taking that practice, I'm not walking through that door.

Caroline: Yeah, difficult. I think we have to get into it, it's a question of how we do it at scale. But the other thing to say is, of course, all government delivery - well, not all of it, but health - we've had PFIs for years, we've had the primary care LIFT partnerships, and all our supply chains are pretty commercial, and I think that's okay.

Tom: I'm very proud of the fact that DrDoctor is a privately held business, but we only work with the NHS. We don't work with any private providers, we don't try to work with any other countries, because I'm here because I love the NHS. And we became a B Corp to try and reflect that.

Caroline: But people do still worry, which I think is interesting. Actually, I think it's much more about what the relationship is like with the trust - how are you setting up the outcomes, how are you making sure things like regulation and data protection are there, to make sure everybody's doing the right thing, and it's done well. They should be enablers to scale rather than blockers, I think.

Tom: Totally agree, totally agree. Let's go to NHS Online. So, going live from 2027, 8.5 million virtual appointments planned in the first three years - which I made about four times the volume of an average trust, if I've done my maths correctly. You can call me out if I'm wrong.

Caroline: That's probably about right. Yeah, actually, that's good, well done.

Tom: How are you feeling about it? Ten days in, no team - is it a freeing feeling, or do you feel like you're standing on the edge of a cliff?

Caroline: Both. Life's full of contradictions, and it's great - if you don't have anxiety, then something's wrong, right? So that's good. There is a team - a team of people working as a programme, coders, engineers, people building the base tech, many of them came from NHS England or affiliated suppliers, and they're great. It's really exciting meeting people from a completely different generation from me. Like I said before we started, I'm 57 - to do something like this feels really... NHS leaders say it's humbling and an honour, but actually it's really thrilling too, it's absolutely amazing. I feel very, very lucky to have been given this opportunity, and I promise I'll do my best to make it work. I feel like it's mine.

Tom: To be honest, it's radiating from you, it's like a full tiger mum.

Caroline: I was joking with you, like, you would never give me an executive job because your firm is probably full of young people like you.

Tom: It is full of young people, but we'd have you, Caroline.

Caroline: Yeah, but you'd put me in an advisory role and stick me around the back. But it feels genuinely - and it's because I think we're an NHS organisation - this is about trust, it's about public and political sentiment, it's about getting clinicians and people running health organisations in the NHS to trust us. And then it's about having commercial partnerships so we can go further and faster. So it's all those things. How am I feeling? It's a good, big job, and I'm feeling really energised by it, very excited, and very aware of the scale we have to achieve. It's a startup-scale-up challenge.

Tom: And it's one, to be honest, I'm a little envious of - it sounds really exciting, I've got a bit of FOMO.

Caroline: Good, well that's great, because we need people like you to have FOMO so you'll come and help us. This can sound quite grandiose and hubristic, but this is a really good shot at making the NHS properly modern, properly transformed. As currently constituted, we're sort of the elective delivery arm of the app, in a way, and actually what we need to do is be a doctor in your pocket. Even as a middle-aged woman, I want more and more of my services digital. So I want us to design the whole ecosystem, and the tech has to be able to cope with all that.

Tom: Yeah, that's a big conversation to have.

Caroline: And it's not just about us, it's about how the app team and all those interconnected services work.

Tom: But it's a good conversation to have. I actually think your experience running London may be more valuable than your experience running the Free, because I think it's a coordination role.

Caroline: Yeah, actually, yeah. And it's getting all those different tech platforms - we won't go into the tech platforms, but there are quite a lot of them, nationally and locally - to work together, leveraging them. It's getting the people to work together.

Tom: Were you... I was surprised and impressed by the boldness of it being a trust of its own.

Caroline: Yeah, lots of people have asked me about that. Honestly, the answer is: we're going to be taking clinical risk, so we need to be properly regulated. The CQC will regulate an NHS trust more easily than something else, and NHS England went through quite a few options to get to "NHS trust." In explaining it to my old buddies in the health delivery world, it's going to look a bit more like - I'm going to have a chief nurse, probably not focused on infection control, I'm guessing.

Tom: You never know.

Caroline: I'll have a chief medical officer and all the other roles, but with a disproportionate emphasis on tech and the delivery of a really good architecture, and probably skewed towards a more commercial mindset. So it will look different, but the clinical governance, the safety and the public trust stuff should look recognisable to people.

Tom: Yeah, I think that's healthy. I've got to ask - online hospital, are you going to have a physical office?

Caroline: Well, I've got to sit somewhere, right? At the moment, it's weird - it was only christened on 1 June, so the team are still... they've got a place in Leeds and a place in Waterloo.

Tom: Nice.

Caroline: But there's a kind of - what's the opposite of separation anxiety? I think the NHSE team think we should probably go somewhere else, so we need to figure out where. We need to be able to put our own diagrams on the wall and all that stuff, I think that's important, a sense of identity.

Tom: Oh yeah.

Caroline: And it needs to feel a bit more - look, I'm one of those old-fashioned, visible leaders, I want to see people, and I want to induct them into the office and make them feel it's a really nice place to work. They don't have to come every day, but it should be a cool place to be, all that stuff.

Tom: I'm the same on that - I'm one of those annoying people who's in on a Monday morning. So, what if I worked for DrDoctor - how many days a week would I have to work? Well, we don't mandate any days in the office at all, because I don't believe in mandating. But we've got a really nice space on Bermondsey Street where we've made enough room for people to work, with quiet zones, and we encourage people to come in and collaborate with their colleagues. Once a quarter we do a big thing - everyone gets together for two days and we do an overnight, it's very optional, but what we find is most people spend quite a lot of time together.

Caroline: What I love most is going and spending time together on a hospital site - when we go and do a diagnostic, spend three or four days there, that's where you really get the bonding, because you're solving problems together.

Tom: You're doing things together, and you do what you say you're going to do, and then people trust you.

Caroline: Yeah, exactly, I think that's right. Lots of my friends in organisations like yours have that kind of working pattern, so we've got to think about all that - it works really nicely. And trust - I'm really big on people having lives. I think you can do four hours of really productive, creative work in a day; you can do more work, but less of that hard-brain stuff. So I'm big on: if you want to pick your kids up, pick your kids up, that's cool, because we trust you on your output.

Tom: I think that's an opportunity for the online hospital - if I've got my stats right, 60% of consultants said they'd work for you.

Caroline: Spot on.

Tom: Which is amazing. I think if you can give people that work-life balance in a clinical practice, you'll be overwhelmed with people who want to come in.

Caroline: Yeah, no, I agree. What I said in my interview was I'd really love to create a feeling of faculty - this shouldn't just be the place you go to retire, or get your money for your next holiday. This needs to feel like a place, particularly for clinicians, where you're going to learn how to provide modern, digital care, and then take some of that back into your organisation, and feel like you're actually creating the future, and learn from other countries doing this.

Tom: There's so much in there.

Caroline: And again, that's kind of IHI thinking, I don't think it's a particularly original thought. I've realised, as I've got older, very little of what I say or think is original - it's really quite depressing. And then I heard this stat the other night: I was at an event about geopolitics and AI, and someone put up a slide that basically said, what does AI do to your thinking - a 47% reduction in critical thinking and cognitive ability.

Tom: Yeah, that's what kind of...

Caroline: And that's for individuals - for organisations, something like an 80% reduction. So you have to be a bit careful with that stuff. I'll be open about it - we've really struggled with it internally, because we've really pushed AI, and it's made a huge difference to how quickly you can produce code, and to the breadth of things you can cover, because it can do a lot of the synthesis. But what we're seeing is people sharing things that haven't been as critically appraised as they'd expect, so we're kind of winding it back. It's great as a thinking partner though - if you get writer's block, it's the best.

Tom: It's really helpful.

Caroline: Yeah. So, confession time - sorry, Lawrence Tallon - I did a little presentation for you yesterday, and realised the night before that I hadn't written it. So, to my credit, I went ChatGPT...

Tom: Claude.

Caroline: ...and Copilot. But they all came out anodyne - maybe it was how I fed it, and I didn't structure it properly. I read it back just before, and thought, I can't say any of that. But I think it was all right in the end.

Tom: Which is your favourite to talk to?

Caroline: Oh, Claude, easily. How clever are they to name it Claude?

Tom: I know. Have you got kids?

Caroline: I have, yeah.

Tom: Are they little?

Caroline: They're small, yeah.

Tom: Right, so I'm going to recommend a children's book for you, called "Claude and Sir Bobbily Sock."

Caroline: Amazing, love it! There's a whole series of them, and they're just magnificent. But of course, in my family now, anything that's called Claude is like, right, brilliant, we're in - there's nothing else.

Tom: It's great, honestly, you'll thank me for that another time.

Caroline: I'm going to look it up. We went for Claude as well. There's a growing health presence in the UK from some of the AI labs, by which I mean nothing that NHS England or my organisation is doing particularly, but quite a few of my mates who are GPs all seem to be working for them.

Tom: Oh, really?

Caroline: Yeah, it's a hell of a gig if you can get it - editing prompts and content and just making sure it's right, it's quite interesting. Something's going on. I have a theory that we'll end up with an NHS sovereign LLM.

Tom: Yeah, I think that's where we're going to end up. If we can get to the point where we run that on our NHS data, we're not paying a tax.

Caroline: My hope is for that too. I think that is exactly where we need to go, and we need to be brave - this is quite political, isn't it, you can see the French government have been a bit bolder on sovereignty. But you can also see, when you get into conversations with government people, they're thinking about this. We've obviously got an AI minister now.

Tom: I spend too much of my time thinking about this, if I'm honest - it's the nerdy corner of my brain.

Caroline: I think it's really important, we have to, and if we don't, I think we're a bit stuck.

Tom: Yeah, totally, and not just from a "how do you operate our services" point of view, but a cost point of view too - we can control costs, and, anyway, we could go down that rabbit hole. Do you want to talk about data centres?

Caroline: We can go there as well if you want.

Tom: Yeah, why not.

Caroline: Well, it's just this - I heard Rory Stewart and Alastair Campbell having this debate, Alastair Campbell saying, my daughter hates data centres, you can't have them, everyone's marching against them, look what's happened in America. And actually I have a bit of that at home with my partner and my kid, it's a bit of a live debate. And then Rory saying, well, yeah, but you can't ignore it, you can't not do this. So you've got these quite entrenched views.

Tom: I think there's an analogy to the conversation about nuclear power. I studied engineering...

Caroline: How utterly unsurprising.

Tom: No, not because of what you're doing, not because of where you are - sorry, edit that bit out.

Caroline: No, I'm pro-nuclear. I'm very green, I vote for the Green Party, and I think we have a huge climate challenge we're ignoring, but I am pro-nuclear, and that's a bit of an unusual stance. I think what happened in the '70s around nuclear was there were some challenging things, and because of that, we ended up not investing in the cheapest, cleanest form of energy. I think the data centre debate is similar in some ways, because it's a physical manifestation of people's concerns around AI, rather than the data centres themselves actually being the problem. In fact, they create job opportunities - in areas I think we should be building data centres, in parts of the north, for example, where we used to build things. I think there's a real opportunity there for the UK, and we can take our position as a kind of intermediary between the US and Europe, as a really safe place for people to store their data. So if I was around your kitchen table, I'd be saying it's not the data centres that are the bogeyman, it's the way that data is utilised, and we should think about that.

Tom: Yeah, the other thing that gets thrown in is the water consumption of data centres.

Caroline: And actually then it comes down to the abuse of AI and the sovereignty of data and all that stuff.

Tom: Yeah, it always gets there. And if there's a glass of wine, the debate's a mess.

Caroline: They're the best ones. I'm actually genuinely not sure what I think about it. I totally get what you say about the nuclear analogy, I think that's a good one.

Tom: I have met greens like you, because of course it's about clean energy, right?

Caroline: Yeah - no, I'm feeling quite thoughtful about it. I don't know, we could just switch the microphones off and talk about it for hours, I'm sure.

DrDoctor LinkedIn

Tom Whicher LinkedIn

Caroline Clarke LinkedIn

NHSE director to run NHS Online