33
Examination of witnesses
Liz Kendall, Lord Vallance of Balham and Emran Mian.
Q1 The Chair: Good afternoon and welcome to this session of the Science and Technology Committee. I am very pleased to welcome our witnesses, the right honourable Liz Kendall, Secretary of State for the Department for Science, Innovation and Technology, Lord Vallance of Balham, Minister of State at the department, and Emran Mian, Permanent Secretary for the department. All three are very welcome. We have lots of questions, as you can imagine. Secretary of State, this is your first appearance before our committee, so we invite you to make a brief opening statement. We would like to know what your key priorities are for the role, what you are hoping to achieve in the post and perhaps what you feel you have achieved so far.
Liz Kendall: Thank you so much for inviting me and the team here today. I have been clear since I started in this job that the first priority for us is to use the huge power and strengths that this country has in science and technology to grow the economy. There is no route to stronger, better growth and no answer to how we compete and pay our way in the world without science and technology absolutely front and centre. We have huge strengths as a country, but we need to do even more to mobilise them to grow the economy. I am sure we will come on to discuss the really excellent report that this committee has done on these issues.
I am very proud that, in the last year, we have launched five AI growth zones, creating 15,000 jobs, many on former industrial sites—doing more to level up the country in the last year alone than we saw under many years of the previous Government. We are seeing our science and technology companies grow and succeed. There is much more we have to do, but that is my absolute first priority.
Secondly, I strongly believe that, if we are going to seize the opportunities that science and technology brings, we have to do two things. First, we have to show that it benefits people in all parts of the country from every background. Our focus on growth is very much along those lines; we have to show how the incredible potential in science and tech can be felt in communities across the country. Alongside those AI growth zones, I am very proud that we have launched the UK’s first tech town in Barnsley, where we will use the full power of government to show how AI and technology can improve a local economy, local public services and local skills and opportunities. The second thing we need to do, if people are going to feel confident to seize those opportunities, is keep children safe online. Just yesterday we launched what I believe is the world’s most comprehensive and ambitious consultation on how we keep children safe online and prepare them for the future.
My third priority is to make sure that government and our vital public services can use technology so that services fit around people, rather than making people fit into lots of different government or public services, and that we use that power particularly to help those from disadvantaged backgrounds. I am proud that one of the Prime Minister’s AI big bets is to use the power of AI to help the most disadvantaged children with new AI tutors in schools, to narrow the disadvantage gap between children from poor backgrounds and their better-off peers. That is just one example of how we can use technology to shape government and public services to deliver better outcomes for people. I think that many people feel worried about the future of AI and tech, but those who work in this sector see its power and potential. The role of government is to make sure that we leave no community behind and give people the skills and confidence they need to seize these opportunities, because change is inevitable but the consequences are not. The role of this Government is to make sure that we seize the benefits of this for all.
The Chair: Thank you. We have more questions on AI, which we will come to. DSIT is a relatively new department, which we welcome. When we have spoken in the past to Lord Vallance, he has talked about his desire to have a horizontal role for DSIT—in other words, being able to work across government. You have a lot of cross-government responsibilities in scientific expertise, digital transformation and so on. What mechanisms exist for you to work across government effectively? Are you optimistic that DSIT now has that horizontal role that Lord Vallance referred to?
Liz Kendall: I have always been a “glass half full” sort of woman. I am realistic, but if you are not optimistic about the potential to change things then I do not know what you would be doing in this job. We are making real progress, and I have certainly seen it. I saw it in my previous job, as Secretary of State for Work and Pensions, when we worked really closely with DSIT on trying to think through how we modernise and update our services—believe you me, that needs doing at DWP, where we have some very old legacy systems—and start to use tech and AI, for example, to really understand local labour markets and match the people who need jobs with the skills and jobs available for them, and how you could change all that.
We are also working especially well with the department of health. It is motoring on all cylinders in trying to modernise its systems. I am sure there is more we could do across other government departments, but I think we are showing how we can serve other departments and help them achieve their goals. In the end, I have always believed, as the suffragettes said, in “Deeds not words”. That is what we want. The proof of the pudding is in the eating. When we show how we can help other government departments improve their services, we build that trust and then we can make further progress.
Q2 The Chair: Thank you. We are going to switch the order now and focus on reforms to science funding. These questions are more directed at Lord Vallance. We are familiar with the concept of research funding being directed into three buckets—the pure, the applied, and the support for companies—but, as you are probably aware, several research councils have paused grants and announced cuts to basic science funding and there is considerable disquiet at the moment in the science community. We want to ask a bit more about that. We have had quite a number of letters from concerned scientists and the Royal Society saying that these cuts jeopardise the future of basic research, on which the UK has historically been very strong. Is it the Government’s policy to cut funding for curiosity-driven research—from bucket 1—in favour of research for the other two buckets? Is that the position?
Lord Vallance of Balham: That is the exact opposite of the position. I have been very clear that, of the three buckets of funding, the one that is most important to keep, maintain and grow, as the economy allows it to, is basic investigator-led, curiosity-driven research. That is what has happened. The funding for investigator-led research is going from £737 million per year previously up to £866 million by the end of this spending review. There have been no cuts in basic, curiosity-driven research. There is a very particular problem around STFC that I am sure we will come back to, but it is not the case that there have been cuts in any of the other areas.
I will just deal with the pauses. There is a pause that BBSRC has put in place for a few weeks, which is because it is moving to a system of no grant deadlines, where you can submit at any time. It is just transitioning to a new system where people can submit grants whenever they want to. EPSRC had a pause for a few weeks, which was to do with a particular programme. I cannot remember what it is called now, but it is an innovation programme which they have been running, which was seen as so successful that they wanted other disciplines to be able to join in, so they are amending that to make it open to other disciplines beyond EPSRC.
MRC had its last funding call in autumn last year. The results of that will come out in April, with funding being given out then, and the next round of funding will start in May, so there has been a gap there of a few months while they made that transition. One of the reasons they made that change was to try and make sure that when they launch the new call, which will be in May this year, they can make it clear what the totality of the funding is across buckets 1 and 2, so that people can see the whole picture before they put grants in. They are not cuts; they are pauses to get the system into the right place.
I do not think we should underestimate the degree of change that Ian Chapman is driving. Think about what UKRI has done. The research councils have been amazing over many years, and UKRI has tried to get into a new position based on what Paul Nurse envisaged. Ian is actually making that happen now in terms of interdisciplinary work and greater transparency of where money goes, and there are some quite painful trade-offs to make, to make sure that we get that right. He is trying to do something which is appropriate.
The Chair: In the spending review period and the sort of forecasts of how the three buckets will be allocated spending, there is a question of QR funding. Would it be right to say that QR funding is being assumed to principally support bucket 1?
Lord Vallance of Balham: No, it is going to support whatever the universities want it to support. It has just been allocated somewhere, because you had to put it somewhere.
The Chair: Allocated to?
Lord Vallance of Balham: It is currently bucket-1 allocated, and there is an increase in overall block grant to universities. It is going to be about £435 million over the period of the spending review, and that may be reallocated to other buckets, actually. Ian and the team—I think correctly—decided that trying to divide QR up in a complicated formula was bureaucratic. You just needed to get the money out.
The Chair: We understand that. Lord Duncan, you want to come in?
Q3 Lord Duncan of Springbank: It is just a very quick point. We have had a lot of letters from learned bodies. You say that their concerns are groundless—that they have misunderstood. If that is the case, is that perhaps a failure on your part to explain to them? We would not normally expect quite such a wealth of concerned correspondence—and we have had a lot.
Lord Vallance of Balham: Yes. We all have. Ian Chapman was clear when he appeared in front of the Commons Select Committee that he thinks the communications from UKRI have been poor. We can all agree that has not been done well at all and there is a lot of work to do to get that in the right place. The bit that is right, is important and is being looked at very carefully is this specific question around STFC. STFC is unusual in research councils because it has a very large infrastructure pot, and it also funds particle physics and astronomy. Therefore, there has always been this tension between the research projects and the facilities. When it was first set up, that was probably a sensible thing to do, because most of the big facilities were to do with particle physics and astronomy. That is no longer the case. There are other big facilities as well.
There is something that needs to be resolved there, but what I have been clear about with UKRI—and I will be clear about it here—is the thing that cannot suffer as we try and work out how to get this funding right across the facilities. The basic, curiosity-driven, investigator-led research in that bucket needs to be protected. It needs to be sorted out. Because of the way that was handled, where a letter went out saying, “Can you model”—I cannot remember what it was—“30, 60% and 70% cuts?”, not surprisingly people thought, “Well, that is pretty gruesome. We cannot do that”. That is what has driven a lot of the concern. Then the pauses have made people think, “Oh, maybe there are other cuts coming”. But if you actually look at the numbers, they have all increased. It is really worth reflecting that we got the biggest ever settlement for R&D in a Budget—a spending review—that was extremely tight. We got an uplift right the way across the board in all of the areas.
There is work to be done to articulate this properly—and, frankly, to make it meaningful. I know what it is like; I was an academic for long enough. You care a lot about whether your grant funding is going to be continuous, whether you are going to have gaps in it, whether you can support the people coming through. UKRI needs to make sure that communication is right, because the money is there to support people.
Q4 Baroness Willis of Summertown: Can I just follow up on that one, particularly on the STFC? We are hearing that there is a 30% reduction—the budget itself has not changed, but there is a shifting in the budget for STFC, and there is also the removing of the Drayson partitions, which happened in October 2025. The ring-fencing for the blue skies has gone from that structure. Is that understanding correct? Is that what has happened? Is that why we are seeing this pressure on the blue-skies part of that research budget?
Lord Vallance of Balham: STFC has always had this business of doing infrastructure and research.
Baroness Willis of Summertown: But it was partitioned, was it not? It is ring-fenced.
Lord Vallance of Balham: No, there was no hard partition in that. It has always been tensioned against the two things. In fact, it has been quite favourable for the research side for a very long period. My number 1 priority is to protect and grow curiosity-driven research, and I am determined to do exactly that—to protect and grow it. But the real issue at the moment that they have got in STFC is the overspend on facilities. If you look at the international spend, it has gone up by about 20% at a time when domestic spend has gone up at about 11% over a period of six or seven years. That has put big pressure on the overall system. In previous years, the overspend in STFC has then been absorbed by the other research councils, so there has been a strange picture where other research councils have actually ended up having to give money into the system to cover that. We need to fix that. We need a sustainable, proper, well thought-through, structured way to fund the infrastructure. I am very determined that UKRI must find a way to look after so-called PPAN—particle physics and astronomy. All the funding gets tensioned against something. It either gets tensioned against the rest of physics, which might be one way to do it, or you tension it against EPSRC or something. But they are looking at that at the moment.
Baroness Willis of Summertown: Will that 30% reduction be looked at again?
Lord Vallance of Balham: There has been no statement about a 30% reduction. The statement has been, “Could you please look at what those reductions would mean for the community, and what that would translate to in terms of impact?” One thing I am confident of is that in Michele Dougherty we have an outstanding world-class scientist—Astronomer Royal, FRS. She understands this community. It is right that the heads of the research councils should make these decisions. They will work out where the best way to tension this is. In that particular example, there is the bigger question of “What do you tension it against in terms of other disciplines?”, rather than just putting it all in one place.
Baroness Willis of Summertown: Could I just come back to that? If there is going to be a reduction, who will make the decision as to where those cuts are made? Where will that sit?
Lord Vallance of Balham: In the classic Haldane principle, the decision of exactly where the money is used will be made by a research council with expert advice. They have got an expert advisory panel. If it goes outside STFC, that is one for UKRI to look at and say, “What do we do across the patch?” In a way, what is happening now with the buckets is to try and get more of a cross-council view of things, less of a rigid-discipline view where you cannot do some of the things that UKRI was set up to. It will be a decision made by those with the expert knowledge, in consultation with scientific communities.
Q5 Lord Drayson: Before I ask my question, I should declare my interests as a science entrepreneur, and in particular as chairman of the AI company LociLabs and the wireless energy company Freevolt Technologies. Patrick—Minister—you mentioned there are painful trade-offs to make, but the Government have declared their intent to, if you like, take more control over how science spending is made and to align it with their industrial strategy. Do you accept that that is what is taking place—that the Government are effectively taking more control over what types of research get funded?
Lord Vallance of Balham: They are very deliberately doing it in the three buckets, which is entirely appropriate. I go back to Haldane; I am sure you will be familiar that he not only gave the principle about where decisions on individual projects are made but described general and specific research. General research—research necessary to increase knowledge—should be driven totally by academic curiosity. Specific research is for things that departments and government need research on. If we want a sustainable funding system in this country of £38.6 billion, we will not get that if we just say: “Well, you know what? It’s just a big bucket of money and it goes wherever you want it to”.
We said that the curiosity-driven bit should be the same as it always has been and grow as the economy allows it to grow. For the other parts—the Innovate part; the bits for companies—we should have a much clearer view on what private sector leverage we should get from that bucket and be clear about how that bucket delivers growth for the UK. The middle bucket is about societal and government priorities, things that matter to more applied science to deliver benefit. That has always been things that UKRI funds. There is no difference in the types of things being funded; the difference is in the ability to see where the money is going. The industrial strategy bit does not apply to bucket 1.
Lord Drayson: But you will accept, I think, that the reorganisation that UKRI is bringing in—you have mentioned its looking to facilitate the removal of duplication and have cross-cutting thematic research—means that the complexity of the decision-making process is becoming more opaque. This is driving some of the concern that the science community is clearly feeling. What can you say to reassure the community that the way in which decisions are going to be taken about what research projects are funded will conform, as you say, with the Haldane principle?
Lord Vallance of Balham: There is great clarity that £14.5 billion is going to curiosity-driven research, £8.3 billion to the middle bucket of societal and government priorities, and £7.4 billion to companies and the Innovate side of things. It is very clear how much is going into each area. In terms of how that money is spent in the areas, buckets 2 and 3 have an industrial strategy bias—things that are relevant to what the country says it needs to do—and the decisions on individual projects, which are really in buckets 1 and 2, will completely follow the Haldane principle. The experts will decide where the money goes, as it has always been.
Lord Drayson: What about the balance between the themes? For example, the industrial strategy theme around AI cuts across many areas of scientific research. How will visibility be generated on how the balance is struck for those decisions?
Lord Vallance of Balham: I should say that there is a fourth pot of money for infrastructure, skills and things. AI goes across all the buckets, and some of the AI funding comes out of the £8.4 billion in that. I think Ian’s aim—we should remember that he has not been in post that long—is to try to get the UKRI systems in a place where you can really see where the money is going, why it is going to certain places, what the priorities are and how they are being supported, and what the system is for allocation of funding. It is an uncomfortable period and I absolutely get the anxieties that are there, but the aim is to make this much more transparent. The challenge is that everyone now wants to map it back on to what it was before—which I think he has done, to show how it is changing. The intention is absolutely to make this as transparent as possible. One of the primary aims is to make this transparent, and I think he needs to get to that. I will say again that the comms have been terrible—absolutely no question about that.
Lord Drayson: We hear you and I think we agree. Secretary of State, following up on the STFC question, this is not a new problem. Geopolitical shocks which lead to changes in energy prices and exchange rates disproportionately affect the big facilities. We saw this back in the financial crash of 2007-08. That is when we put in those protections to ensure that the other budgets were not hit. I think this requires the Treasury to accept that, for long-term research, this type of fundamental facility requires long-term funding. When short-term shocks happen, as we are seeing in the Middle East with energy prices right now, the Government need to be able to recognise the long-term funding requirements for the science budget to protect these facilities. Would you like to comment on that?
Liz Kendall: I think the fact that, during an extremely difficult spending review, as Patrick said, we have the biggest investment into research and development of any Government ever is an indication from the Prime Minister and the Chancellor that we understand that proper investment in science and research and development is critical for the future of the country and the future of growth. I cannot put it any more simply than that. The Chancellor set out in her spring statement today that, in her next Mansion House speech, one of her top three priorities will be to look at how AI, science and tech is at the heart of our growth strategy for the future. You will see the commitment on that.
Of course there are pressures on everybody’s budgets, particularly in this area, because of energy costs. You know through your long experience that that is not a new thing. We are here again, but our commitment to long-term funding of these areas is absolutely there. Secretaries of State in other departments would look with envy at the long-term increasing investment we have in science and technology. I am sure there is more that we can do, and we will, but I think that is crystal-clear.
Q6 Lord Willis of Knaresborough: It is encouraging to hear what you have said about investment in industry as we go forward. What I cannot understand, Lord Vallance, is why the Institute of Physics is so critical of what you are doing. Why is it being so out of form?
Lord Vallance of Balham: It is worth saying that Michele Dougherty, the head of the STFC, was the head of Institute of Physics and has stepped down to take on this role. That is the organisation that she has just come from.
Lord Willis of Knaresborough: So it is her fault?
Lord Vallance of Balham: No, it is not her fault at all. I am just saying that she knows this area extremely well and is obviously determined to make sure that it is successful. I think this question of the particle physics and astronomy partition in the STFC is the root of the concern. I think the concern is very legitimate, because of the tension between the increasing facilities cost. That is what she is trying to work through at the moment and come up with solutions.
Lord Willis of Knaresborough: I am hugely concerned about what is happening in our universities and particularly in our education system, because unless we get that right over the next few years, no matter how much we spend elsewhere, we will fail. Getting the Institute of Physics on board as a commitment is important to us.
Q7 Lord Stern of Brentford: Thank you very much for being with us and for getting the significant increase in the R&D budget. It is really important. My question is about understanding what the buckets mean. Perhaps there is a danger of replacing disciplinary rigidity with bucket rigidity. Many people around this table are researchers; that is what we do and where we have spent our life. Most of the interesting stuff goes backwards and forwards between the buckets. I worry that by insisting that this over here is blue sky and this over here is applied, you risk leaving out or not concentrating enough on the most interesting things, which is the to and fro. Do you worry about that rigidity? Are you doing something to avoid it, locking you into strange decisions?
Lord Vallance of Balham: It is absolutely one of the risks that, wherever you divide something, you are going to have that problem. We have been divided on nine research councils quite rigidly by discipline. The plea from everybody was: “We need to be able to do more together”.
I will take, as an example, AI, which does not fit in any one of those but is going to change every single aspect of academic research. That needs something that is cross-cutting, able to support all the disciplines and able to encourage what is going to be an extraordinary area of interdisciplinary research. This system allows that to happen much more easily than if things are segmented nine ways.
But I completely agree that, to take the example, if £14.5 billion for curiosity-driven research becomes absolutely fixed in stone and you cannot increase it, if you need to, by viring between the buckets, then that is a problem. But I am sure that is what UKRI will do. It will look at how to make that work across buckets, and it is going to put in systems.
The rest of it, the £8.4 billion, is precisely there to enable investment that supports things across all the buckets. There are systems to try to avoid that. But I have worked in enough organisations to know that, whichever way you cut something, you will end up with interfaces. That is what we need to watch out for.
Lord Stern of Brentford: In this case, it is not simply: “You can cut here and cut there”. The logic of our whole way of doing research is the pure and applied mix, and that is what we want. It is not about saying, “Whichever way you do it, there’s a problem”. This is at the heart of the story of what really productive research looks like.
Lord Vallance of Balham: It is, as is interdisciplinary work. That is my point—that is also at the heart of it. That was very difficult sometimes, in the old system, and a perpetual worry, as you will know, that you could not mix the arts, humanities and biology and you could not get a grant to do that. This is an attempt to try to get that right.
I do not think there is any desire to try and make this a rigid system. The point about bucket 1—I want to be really clear about this—is that it is the bucket in which things happen that no Government would necessarily think about, have any idea whether it is going to be of any value to anyone or be able to give any direction to that. The right answer to the question “Can you tell me the impact of your research?” might well be, “I’ve got no idea”. But in 10 years’ time, this is going to be incredibly important, and we need to allow that to happen.
You are right that moving across into development, company formation and commercialisation needs to be very easy across this path, because that is ultimately how the eggs that get laid in bucket 1 turn out to be the thing that grows in buckets 2 and 3.
Liz Kendall: Chair, may I just make a tiny point here, which is that when Sir Ian was in front of the committee down our neck of the woods, he faced similar questions. I myself, as a new Secretary of State, have asked to look at questions such as: how do you compare? What did we fund in the past and up to now? How difficult is it? He committed to producing what he called high-level mapping about what has happened in the changes. As soon as that is available, we will make sure that this committee has that information too.
The Chair: Thank you. We are just going to spend a little bit longer on this very important subject of UKRI funding.
Q8 Lord Patel: Thank you all for coming to talk to us today. I have two very brief questions. Listening to the conversation that has been going on, I am sure, when you were thinking out this new way of R&D that will drive our economy, that you made an assessment as to what the gains will be and that there will, no doubt, be some downsides. So what was that assessment?
Lord Vallance of Balham: One of the gains is greater clarity on where money is going, which becomes increasingly important as spending reviews are always tight and difficult. I have been in and around government for long enough to know that that question is getting more acute. Where does this money go?
One of the unintended consequences of creating UKRI is that it is a big pot of money. When it was nine smallish pots of money, it was not interesting to people; this is a very interesting sum of money and you need to say what has happened to it. One gain was clarity.
Secondly, and more importantly, was to make sure that we really value and drive the type of research that gives answers to societal and government problems. We do that in a disciplined way, in which we put timelines on outcomes, are clear about what the benefits will be, and get pull-through from that to use in government, the public sector, companies and company formation.
We see those measures as ways to get a more valued and better approach to that bit of the funding, which will deliver societal and economic benefit. We view the first bucket as protecting that against what I have seen in companies and see as a risk in government, which is somebody looking at the £14.5 billion and saying, “Well, it wouldn’t really matter if we didn’t do that for a while”. It matters enormously because once you lose that, you lose it for a very long time, and it is that work that ultimately creates wealth in 10 or 20 or 30 years’ time, even though I cannot tell you which bits of it are going to create wealth.
Q9 Lord Patel: My second question is: you made a fairly clear case about the anxieties we heard from the science community on the reductions to basic science—curiosity-driven science—but there is another side to the curiosity-driven science which is the other policies that will make it possible for universities to be able to carry out the basic science. The implications are there that, if you cut the visas for international students, the universities cannot recruit those students, which has a knock-on effect on their finances and means that they are not able to support blue-sky research. The same applies to international students for research purposes. Do you have any comments on that?
Lord Vallance of Balham: I think you are talking about the income that comes from overseas students, which became a big source for lots of universities, as you know. Part of the reason for financial difficulties at the moment is that the numbers are dropping for all sorts of reasons, including visas and so on, and that has put pressure on universities. That is a big issue that has been recognised.
The Post-16 Education and Skills White Paper, published late last year by the DfE, with our input, dealt with a number of measures, in consultation with the sector, on what can be done to get universities on to a more stable footing financially. This goes back to the earlier question. The universities are crucial for everything we do and we need to make sure that we do not end up with a university sector that is unable to perform.
Some of the changes there will make a significant difference, including in how universities, particularly the smaller ones, might work together better; how you can share facilities and equipment; and how you can look at courses that go across universities rather than sticking within just one. There are a series of measures in the White Paper that will be important, including, of course, how we remain open and attractive to international students from around the world.
The Chair: We want to switch topics now to AI and the future of work. Secretary of State, these questions are, at least to start with, principally for you.
Q10 Lord Ranger of Northwood: Secretary of State, thank you for coming. AI—exciting times but also worrying times, especially when we think of the jobs market. I know you have recently announced the new AI and the Future of Work Unit in your department, which is an interesting initiative and will be welcomed by a lot of people. Just to get under the skin of it a little, you mentioned to Bloomberg the impact on entry-level jobs for graduates in areas such as finance, business and law is significant at the moment due to AI, with one in six people now unemployed. I would also throw creatives in there, because I think that every area of creativity is also being affected by AI immediately. With statements such that, workers are quite rightly worried, as are new graduates. DSIT’s analysis is saying that the UK is quite significantly exposed to potential AI jobs automation. What are your ambitions for this new unit and the support it will give? There is a question around its immediate, medium and long-term objectives. We see the immediate challenge with potential graduates and people in the market being displaced as we speak, but how are you looking to address that medium level of transition and what is the longer-term vision of work? It looks like it is all going to change.
Liz Kendall: This is hugely important to me personally and to the Government. We do not believe the Government should be a bystander in the jobs transition, which there will inevitably be, as there is with any technological change. However, it is especially so with one as powerful as AI, a general-purpose technology which, despite its huge advantages, is possibly still in the foothills of its development.
I think that AI will create jobs. We are already seeing that in our AI growth zones, with 15,000 jobs in the last year and £40 billion of inward investment. It will change jobs, freeing up people’s time from more routine tasks to focus on the things that matter to them, whether that is growing their business or, if you are a teacher, as I have seen in my constituency, using AI tools to do lesson planning and getting the equivalent of 30 days a year back so you can spend more time with kids in the classroom and with your own kids at home.
But some jobs will go, and we need to support people through that. We do not just think we will let people cope on their own. That is why we have set out a big group of things. I will come on to the Future of Work Unit. We have announced free AI skills for all; we have a new range of skills courses that anybody from all walks of life can take, from 20 minutes to a whole week. I encourage members of the committee to give it a go. I did one myself and was disappointed that I did not get 10 out of 10 in my test at the end, but I am going to redo it.
Our goal is to upskill 10 million people by 2030. That is a third of the workforce. We have already done 1 million. We are working with partners in the private sector, the trade union sector and our public services. That is really important. We are also giving £5 million in each of our AI growth zones to ensure that local businesses can adopt and take up this technology. Our goal to make Britain the fastest-adopting AI country in the G7 is really important, because the best protection for businesses and jobs is to make them more productive and more competitive.
The third thing, which is extremely important, is that as a Government we have a plan to support people through this. The Future of Work Unit will gather the best available evidence in this country and internationally about the impact on jobs and bring together the full force of government—the Treasury, the DWP, the Department for Education, DSIT and the department for business—to see what more we need to do, whether on skills or business adoption, so that we have a plan to help the country through these difficult times. We want to see AI create jobs, but we know that we have to support workers through that. I want to level with people about that.
Lord Ranger of Northwood: On that point about the Future of Work Unit, it is good to hear that it will be looking at those things internationally and using the power of government, as you say, but, in terms of engagement with businesses and what they are looking for, how will it curate its offering and inform its thinking? How will it engage with business?
Liz Kendall: We will be working with all relevant organisations—the CBI, the FSB and the British Chambers of Commerce as well as trade unions and the TUC. The Future of Work Unit will have a representative from the CBI and one from a trade union, as well as the involvement of academics. I have always believed in partnership; we need to do this together if we are going to seize the benefits.
In the upskilling programme of 10 million people, we want to make sure that a minimum of 2 million of them are in SMEs. I know from my own constituency in the great city of Leicester that SMEs are the backbone of our economy. Sage, the most successful tech company at the minute in Britain on some measures, often sells its AI products to small businesses so that they can use AI to free up their time from doing accountancy, HR or other functions.
Let me come back to where you started. Because of our service economy, I believe we have both the greatest challenges and the greatest opportunities from AI. There is more that we need to do to seize that. In the end, we know that the more people use AI and see the benefits, the less fearful they may become, the more productive their businesses become and the more they can grow. That is the key. The best way to protect jobs is to have competitive businesses that are growing. There is a lot more we need to do to support businesses to do that.
Q11 Baroness Nicholson of Winterbourne: Thank you very much, Secretary of State, for that clear statement. Earlier history in the United Kingdom shows that, where we have these enormous spurts of growth in something technical, jobs cannot be protected as they are because the work itself has changed. Can you explain how you will be looking into the jobs that are going and how to train people for something completely new—not AI, because AI jobs will be taken by other people? It is always about what to do with this enormous number of people who lose their known employment, based on their earlier training, and government finding other pockets of unemployment for them and training them up to that. Is that something you are thinking about?
Liz Kendall: I am not a scientist. I studied history at university. I have looked at the first Industrial Revolution—the fears about it and what happened initially, and then the decisions and power of government and organisations that shaped those changes to better meet the needs of workers in terms of pay, protections and conditions. I believe in the power of government to act to support people. I believe in progress and I am a progressive politician who thinks you do not just leave people to cope on their own. I believe the decisions that government can take can make a difference.
We already know about the impact of AI on some jobs, but it will create jobs that we do not even know about. Skills are at the heart of this. We already know that, if you have AI skills, you are more likely to be employed and to have a better-paid job. That is what the initial evidence is showing. I have no doubt that some jobs will go, but others will be created that we cannot even imagine.
On my first trip to Silicon Valley, I decided not to go to Stanford and Berkeley—I will go there at some point—but to San José State University. It has 40,000 students. They are the workforce of California. The principal said to me: “We are in the belly of the beast. If I do not train and get my young people to understand what AI will mean and give them the right skills to get the right jobs, I will fail”. She does not just have AI courses; every single faculty has been asked to think through the implications for teachers, nurses, business administrators, engineers and scientists, right across the board. She is doing for her students what we need to do for the country, because we need to be prepared for the changes that AI can bring in every walk of life, because that is the best way. She understands that some very big companies are gaining all the benefits. No—she wants the benefits for her students, the workforce of California. That is what I want for this country too.
Baroness Nicholson of Winterbourne: Secretary of State, thank you very much, but the Government cannot control AI; indeed, at the moment its owners do not control it. It is an animal growing of itself that will automatically bring in a huge amount of new thinking and jobs. However, the Government are perhaps uniquely positioned to help and catch the people left behind. Will this be a top priority for you?
Liz Kendall: Absolutely. Our AI growth zone in the north-east is literally on a former industrial site. This is true in north Wales, south Wales and Lanarkshire. Our first AI growth zone in Scotland is in the place that was the beating heart of the Industrial Revolution in Scotland. We are saying: “You will lead this technological revolution”. We have to show how this benefits the people and places who most need hope that there is a better future for them and their kids—good jobs, prosperity and something to be proud of. That is the priority of this Government.
Q12 Baroness Jones of Whitchurch: Secretary of State, you have said that the future of work is a key priority for the department. There is a huge degree of urgency around all this. One in six young people is unemployed and graduate vacancies are dropping by something like 70%. It is an urgent issue, so can you explain a bit more what difference you expect the Future of Work Unit to make? Government departments always have units. What will be so special about this one? Will it not only do some research but drive change on an urgent basis in the way that we expect and hope the Government to do?
Liz Kendall: You are right to stress the urgency and be sceptical—what is some internal unit actually going to do? That would be my starting point too. But the truth is that we need a cross-government plan that urgently includes the Department for Work and Pensions. We are developing a youth guarantee so that every young person will be earning or learning, because we do not accept that 1 million young people in this country are not in education, employment or training. How will new AI skills be part of what we are putting in place for young people for the future? On the Department for Education, we have already looked at the curriculum, how we need to change GCSEs and what more we need to do on apprenticeships. Apprenticeship starts for young people have dropped by 40%. We do not accept that. We need to turn that around with skills for the future.
We have announced that Barnsley is our first UK tech town. We are going to use the full force of government to go in and do what people in Barnsley want, whether that is new work programmes or looking at how we transform public services and support businesses. We will pilot approaches there too, because the leadership in Barnsley is determined to make sure that we build on their existing strengths in tech and AI, but we will show how it will work for people there.
We do want research and evidence. We should have the best possible research. But, above all, we want action. Unless we bring government together, we will never deliver that, because people do not see their needs in siloed little boxes. We will bring everybody together. How will we make sure that tech and AI and the youth guarantee work? How will we pilot things in Barnsley to show how we will tackle those deep-rooted inequalities and challenges facing young people? How do we make sure our education system is fit for the future? It is not just AI being done over here. AI will affect everything and we are determined to make sure it delivers for people.
Baroness Jones of Whitchurch: All that cross-governmental work you are describing is great, and I am sure that that is the right approach, but can I take you back to what DSIT will do with the Future of Work Unit? How many people are in this unit? Will it do only research or something much more proactive? Can you give us a bit more confidence that the department has a proper plan for action and that it will make significant change?
Emran Mian: If I may, I will come in on this. Sometimes, the mistake the Government make is to burrow down and try to answer questions such as this on their own, thinking that civil servants working on their own will get to the answer. That is rarely the best approach to any complex issue, but it is certainly not to this, because the technology is moving so quickly.
Here we have very deliberately made the choice, following the Secretary of State’s speech, to do it out in the open. We have been taking that approach across AI policy. We published, for example, a consultation paper on our AI Growth Lab where we very openly asked AI companies looking to grow in the UK, “What are the areas where regulation may be getting in the way of your growth? What are the areas where you would encourage us to look at regulation in a different way?” We have tried to do the same thing here too. We have not tried to work out exactly what we think the best approach is to managing the impact of AI on the future of work. We will do it out in the open, with academics, think tanks, the CBI, the TUC and others across government.
It is going to be a big priority for us. We are growing the team at the moment and will ask colleagues from other parts of government to join us in this endeavour. We are hoping to use the Growth and Living Standards Committee that the Government have announced as a way of bringing together senior politicians and Ministers on this so that we can have a common plan across government.
The Chair: We will now move to the third area: economic growth from UK science and technology. The questions are probably more for you, Lord Vallance. You will know that this committee has recently done an inquiry on this subject, but we have some more questions around that.
Q13 Lord Stern of Brentford: Obviously, we come at the growth story from the point of view of science and technology, but that is set in the context of the UK having the lowest investment rate in the G7. We are off probably five percentage points or so compared with 30 or 35 years ago. We are looking particularly at the innovation side, but a lot of innovation is embodied in investment. Investment drives innovation and innovation drives investment. Your response to our earlier report on scaling up S&T companies, which we found interesting, was that a new cross-departmental approach to supporting scaling up would come in the middle of this year. Could you tell us how that cross-departmental approach will help unlock innovation and investment and, particularly, scale up S&T companies?
Lord Vallance of Balham: Scaling up, as you well know, requires not just funding but regulation, procurement, pull, skills, training and clear unlocking of private sector capital. Those levers do not all sit in DSIT. That comes back to the horizontal role discussed earlier. I hope that what you saw in the Budget was a very joined-up approach to scaling up, including things such as getting shareholder options for people in start-up companies and expanding them so they can grow to a bigger stage and still have effective share options for people working there—including maintaining the R&D tax credits, linked to a statement about procurement. There would be more advanced market commitments and ways of using that as a pull-through mechanism, which is more valuable than a grant at the beginning; at that stage, a company which gets a contract from government will get massive investment.
I think you can already see that that Budget was not a segmented departmental series of things. It was a joined-up approach to scale-up. That group has been meeting on a regular basis. There is now big cross-government work on procurement of innovation, which is one of the key things we need to fix to get this right. The Regulatory Innovation Office is working right across government to look at regulatory barriers. I think this is now working. The Science and Technology Committee has now been reformed as the Digital and Technology Committee with that as part of its mandate, and it will be prime ministerially chaired.
Lord Stern of Brentford: As you know, I spent a bit of my life in Her Majesty’s Treasury. How enthusiastic and active is the Treasury in this process?
Lord Vallance of Balham: Very. On the procurement project, Spencer Livermore, the Financial Secretary to the Treasury, and I meet very regularly, together with Varun Chandra from No. 10 and Alex Depledge from the Treasury. It is a very joined-up approach and there is a lot of enthusiasm for this. The Chancellor is very keen on this as a mechanism to make sure we get things moving. There is lots of enthusiasm, but I am not naive enough to think that that will spread through every part of the Treasury.
Lord Stern of Brentford: Thank you. I understand what you just said. Finally, we started the discussion on this issue on the conditions for the industry and the activity, which was the right way to start, but there are also questions about supporting individual companies. In the bad old days, they sneered about picking winners, as opposed to creating the conditions for winners to come through and supporting them when they did. How will you think about supporting and picking the companies as you see them starting to emerge?
Lord Vallance of Balham: Innovate UK, under Tom Adeyoola’s leadership, is looking at how to focus things and to give bigger cheque sizes to certain companies, rather than having a small amount going all over the place. There is an element of beginning to make choices, even at Innovate.
You will know well that, at the start of this, with lots of start-ups, you do not want to make too many choices, because you do not know which ones are going to make it. As you get to scale up, you have to make some choices, and you need to accept that some of them will fail.
One of the things that I have been pushing on—and, I have to say, I have been pushing at an open door with both the National Audit Office and the chair of the Public Accounts Committee—is whether we can do more to create a culture where if you pick, let us say, five companies in a particular area in the UK that you want to back and three of them fall over, that is not a failure; that is the price of success for innovation. One of the cultural things that we need to work on is how we get a culture that accepts that there are failures during the processes you go through, and you do not have, therefore, a conservative risk reduction in terms of where you back things. You will start to see choices being made, particularly at that later stage, when you need to make choices and start using a procurement pool to try to get customers for these companies.
Liz Kendall: The sense of ambition and urgency in your report is correct. While we are taking action on many, if not all, of the areas you identified, I am absolutely sure that there is more that we can do.
On the issue of using the power of public procurement, we are beginning to make some significant progress, particularly around defence issues. There is more that we can and should do. You talked about Innovate UK, the British Business Bank and the National Wealth Fund merger. It is at least much more seamlessly joined up with a single front door. Companies say that they do not want to keep going through this time and time again. We are beginning to do this, whether through our advanced market commitments in chips or our new sovereign AI unit, where the Government will say, “We will put in this amount of money. If your technology reaches this scale, you can use that to leverage in the private sector”, so that we grow our own companies and we do not lose them all to people with deeper pockets. We have to think innovatively, because this is an incredible place to start up a business. It is a cliché, on the scale-up challenge, but we have to do that. There are real opportunities for us as a country.
The question for government, in this fast-moving world—and we have discussed this quite a lot—is how to be nimble. However, we are also accountable for public money. We absolutely have to take those risks, but we will be the ones in the Chamber dealing with questions about why investments were made or not. We have to do that properly, but we have to balance that with the ability to move swiftly, because that is what the world is doing. Currently, we are too slow.
Lord Stern of Brentford: It is the courage of the Minister and the Permanent Secretary, when push comes to shove, to allow that failure.
Emran Mian: To allow that failure and ensure that the people supporting Ministers and making these decisions have the right expertise. That is why it is important that we have people who have expertise in making investments in companies, which is different to making investments in fundamental research—in Innovate UK and the British Business Bank. Crucially, for us, in terms of the function we will run in the department of the sovereign AI unit, that is why have brought in James Wise, who is a partner at Balderton Capital and has extensive experience of early stage venture capital investing, to help advise us, and why we are beginning to recruit a team which has experience in doing exactly this kind of work.
The Chair: We really welcome what has been said about risk aversion and the idea that innovation and risk go hand in hand. There is no question about that.
Liz Kendall: When that happens, if you could all pile into the Commons and support the risk-taking failure, I would be very grateful.
Q14 Lord Drayson: Secretary of State, it was wonderful to hear your recognition of the need to move faster on this agenda. The points you have already made about the option of technology, such as AI, speaks to the growing concern that our report highlighted—and it is in the country—about the United Kingdom becoming a vassal state of the United States through so much of our technology being dependent on the US tech stack and us losing so many companies just at the stage when they have an effect. I congratulate you on being one of the few politicians able to stand up to Elon Musk and win. Can you speak to how you can accelerate this process—for example, in the Mansion House reforms, which have been talked about for a number of years, but the money is not flowing from our pension funds into our UK companies?
Liz Kendall: I know that very well, because I was the Secretary of State who introduced the Pension Schemes Bill into Parliament, which will provide the backstop to the voluntary commitments that have made. That needs to move faster. We have been pretty clear about our intent. We have these voluntary commitments. The legislation is there to make sure that those are fulfilled. It is a big part of what the Chancellor is doing.
We need to have other mechanisms to be much quicker. We need to look at individuals, which is what we are doing through our global talent taskforce. The advanced market commitments in our sovereign AI unit are going to be extremely important. There is the often long, but very important—in terms of the scale—finance of public procurement. I have already mentioned that we have got some big commitments from the Ministry of Defence to direct a certain proportion of its funding to innovative UK companies. We have increased defence spending to 2.5% of GDP. The Prime Minister has said that he wants to go further. We have some very innovative companies.
We need to push from all these different angles. The time will come—I said this in my Bloomberg speech—for people who are ambitious for AI and science, and who want to push the boundaries and do it responsibly. We are the country to come to. The Government need to pull all the levers possible to keep and attract people here.
Lord Drayson: Could you give us some sense of your feeling towards the ability of the UK to work more closely with countries in Europe to create an alternative to, say, the US tech stack?
Liz Kendall: I absolutely want to work, as the Prime Minister has said, more closely with our European allies. It is not only Europe; there are other, so-called middle powers, as Mark Carney has said, where we can build a strong relationship with the technological developments they have.
I strongly believe that, as AI becomes the driver of economic and hard power, we need to be very clear-sighted about the race that Britain is in. I am not naive. I know how strong and powerful America is and what China is doing, but there are races that we can win.
I often use the analogy of the Olympics. If you are old enough to remember 1996 in Atlanta, we were dreadful. But then we took a decision. We said that we were going to back our strengths and pile the resources there, and then we began winning. I strongly believe that we are world-leading in some places. I do not know if it is my competitive nature, but to people who say, “You should be grateful to come third”, nobody runs their fastest race to come second let alone third; you run your fastest race to win. That is what I intend to do.
Q15 Lord Berkeley: You have given us very interesting information. One of the issues we put in our report was a recommendation to set up a new cross-ministerial council to co-ordinate policy for scaling up technology companies. The Government response basically says that the existing Digital and Technology Cabinet Committee will suffice. What has that committee been doing? We do not seem to have heard much about it. Has it met? How often does it meet? What is its agenda, and what is its output? At the moment, we have a feeling that this has been shoved into the long grass. It probably will be.
Liz Kendall: In the Commons just now, the Chancellor said that, as part of her next Mansion House speech, she will set out how AI and technology will be a top three priority for growth in this country. Absolutely, having cross-government committees to drive forward work is essential. I also believe in strong leadership from the Prime Minister down. We have our AI action plan. We have been working on this, not only through the plans that we have put forward. In the end, we have to follow the money, and the investment that we are putting into R&D is just one indicator of how important this issue is. We know that without growth we cannot improve people’s living standards or fund our vital public services, and we know that science and technology is the absolute key to that.
Lord Berkeley: Can I comment briefly—
The Chair: Very briefly.
Liz Kendall: I have been giving long answers; it is not the long questions. I apologise.
Lord Berkeley: —on what we have heard this morning and all the letters that people have received. Is there a slight lack of communication between government—on these things, with all the changes coming—and those who are frightened for the future, maybe rightly or maybe wrongly? Is it something that you would be able to look at? Will you be able to make sure that the communication between government and all these people writing letters is responded to and let them get on with their research, rather than them writing the letters?
Liz Kendall: We always get lots of letters but, as Patrick said, UKRI has itself accepted that it needs to sort out its communication. I am sure that Ian Chapman will be laser-focused on that in the coming weeks.
Q16 Lord Patel: My question is about UK investment in AI. There has been some criticism that we are falling behind, not just in terms of government investment but the Government’s ability to get private sector investment, particularly in areas of chip design and areas where we lead—we may not number one, but we are certainly number two—in quantum computing. Currently, we have made a lot of investment. You mentioned that and all the data centres that we are developing, but where are we talking about narrow AI? What I am talking about is related to quantum computing and the ability of the next generation of general intelligence AI. We are not investing, but we have a quantum computing ability that is probably second only to the United States. We are not taking advantage and developing capacity in the next generation of compute power, because we are not investing, although we have some leading scientists who could take the quantum computing to the next level.
Lord Vallance, I know that you had an opportunity, with a very short question in the Chamber. We were told that there would be announcements about possible investment by the Government, but maybe I have to wait until the Mansion House speech.
Liz Kendall: I am sure that Patrick will want to say something on quantum. I have absolutely no doubt that we are at the forefront here. I want to make sure that those companies stay in the UK and that we do not lose them as we might have lost some of our AI companies. There is more to come on this.
Our venture capital market is the third-largest globally, behind only China and the US. We are leading in Europe. UK science and tech companies got more VC funding in the last year than France, Germany and Switzerland combined, and we produced 16 new so-called unicorn companies in 2025, second only to the US. It is not all doom and gloom, but we need to do more.
It is very interesting that some of the work at the very early stages is looking at really innovative forms of compute. ARIA will be doing some quite interesting work on that. We have a £100 million advance market commitment into chip companies. There is more that we could be doing around quantum. I do not know if Patrick wanted to say any more than that.
Lord Vallance of Balham: I admire your persistence and recognise your impatience. You are having another go at this. All I can say is there will be some announcements coming up on quantum. The Government have put a lot of money over the past decade—a billion or so—into getting quantum in a good place. A billion will go in over the next four years. We have the National Quantum Computing Centre, which is one-of-a-kind in the world, with six or seven different types of quantum computers all in one place. We have lots of companies now. It is a very good system that now needs to be taken to the next level. That next level means getting companies that can truly scale and can produce quantum computers at a significant level. To assuage your impatience, you might even hear something on one little bit of it tonight.
Q17 Baroness Willis of Summertown: Minister, thank you. I have really enjoyed your response and your energy about backing our leading sectors. I could not agree more, and it is very good to hear that being stated. Of course, one of our leading sectors—I would argue, and I hope that Lord Vallance will agree—is the life sciences sector, which seems to be absolutely critical. We seem to be so strong in that, and it is critical for prospects for growth. What measures are the Government now taking to make the UK attractive for the life sciences companies? We keep hearing about these cancelled investments by global pharmaceutical companies in the UK. Have we seen a change in our prospects following the UK-US pharma deal? This is absolutely fundamental to where we need to be going, and I am not hearing very much about what we are doing about it.
Lord Vallance of Balham: The Life Sciences Sector Plan has been published. It is closely linked to the 10 Year Health Plan, which is important. It links both the R&D side and the commercial side through into the NHS.
The deals that were not done get a lot more noise than the deals that are done. It is worth reflecting that UCB has just announced £500 million for its R&D relocation. BioNTech has put a billion in, Moderna has put a billion in, Convatec has put £500 million into its centre in Manchester, and Prologis is putting over £3 billion toward creating a science park around life sciences, including in Cambridge. There is a lot going on, and there has been an uptick in investment and interest. I do not for a second underplay the importance of the problems that we had last year around this.
The US pharma deal is part of making the UK an attractive place and will help to get medicines out to patients as quickly as we can. In a way, it goes back to the very origins of NICE. The idea behind NICE was not to stop lots of things but to make sure that we spent the money on the medicines that have the biggest impact and then to make sure they are ubiquitously available across the NHS. Over the subsequent years, the “ubiquitously available” bit tailed off somewhat. That is one of the things that happened as a result of the deal. We are now back to making sure that we will get the best medicines early. We will make sure that they are out across the NHS. That will stimulate the industry in this country. We have a thriving life sciences biotech sector. It is very strong. We need to scale those companies. It comes back to the scaling point, which you raised in your in your report.
We are very lucky. This is important in terms of scaling. The fact that we have got two very big global companies—AstraZeneca and GSK—is crucial, because they train most of the people who end up doing something in the biotechs, just as DeepMind in AI has been responsible for training almost all the people who end up creating the AI start-ups.
There is a lot going on. I will just try to pull that together in one place. The cancer plan that has just been announced is a brilliant example of how the Life Sciences Sector Plan and the 10 Year Health Plan come together to really improve the outlook for patients, in diagnosis, prevention, treatment and vaccines.
Baroness Willis of Summertown: All these things are music to my ears and are great. On AstraZeneca, how can we be confident that we are creating an environment where it wishes to remain in the UK?
Lord Vallance of Balham: The real issue with AstraZeneca last year was around pricing and access of medicines. It loves the environment from an R&D perspective. It loves the skills it can get here. It loves the biotech scene. It was extremely worried that its medicines were not being pulled through into use in the UK, and that is partly why the pharma deal is so important—to make sure that we get that sorted out.
The Chair: Lord Drayson has a question on ARIA.
Q18 Lord Drayson: This speaks to the IP point, not just related to ARIA, but the way in which the UK, as a leading science nation, has often made available technology to other countries. They then commercialise that technology, and that does not lead to benefits back in the UK. We heard from ARIA that it is not securing control over intellectual property or economic return from the support that it gives. Secretary of State, could you comment on whether or not you believe its “throwing the bread on the waters” approach to the rest of the world is consistent with the UK’s strategic interest?
Liz Kendall: I always want to get the best for Britain. That is my job, but I also understand that science is inherently international in its nature. We previously talked specifically about AI, which I am particularly focused on, and the sovereign capabilities that we have, because it is such an important general-purpose technology. I always want to make sure that we get the balance right. It is already seeing some UK start-ups, which it is supporting. Some of its initial ideas are now already getting out into practice. I understand others’ concerns about it. I am always happy to look at what more we can do to make sure that we keep the very best in Britain.
Lord Drayson: That is good to hear, thank you.
The Chair: We are aware that you have a fairly hard stop on timing. Are you able to go a little bit beyond 3.30 pm?
Liz Kendall: Not much, I am so sorry.
The Chair: We will be as brief as we can. I just want to switch to the final topic: AI governance and its use in government.
Q19 Lord Stern of Brentford: We are short of time, so we will understand any brevity of response. On the question, AI governance is often all about the downside or the difficult things that can happen and how we avoid them. Any thoughts that you have on that would be valuable. Could you also say something on steering AI to where it is most valuable, which is not necessarily interrupting me when I am doing a Google search but, for example, how we run our cities, reform our energy systems, discover new materials and so on? As well as briefly saying something on the regulation—the downside and the harms—could you say something on how we steer it to the potentially enormously productive areas of system management, discovery and so on?
Liz Kendall: I will say something on the former, and Emran could say something on the latter. Politicians do not usually pay tribute to their Opposition, but one of the important things that the last Government did, and Rishi Sunak in particular, was set up the AI Security Institute. The work that it is doing, with some of the best people in the world, looks at the safety and risks of the models that are being developed. I am hugely impressed with it and the way it is working with companies to identify those potential risks and to work with them to put in place mitigations. We are world-leading in that.
It reminds me of what happened back in the day with NICE. We developed something that was innovative, new and world-leading, and then it helped to spread its ideas and ways of working across the world. AISI is very much in that category. The people there are extremely committed, because—this is the thing I have found most interesting since joining the department—the people who are most ambitious for the technology are often the most responsible, because they know that it is only going to work if people feel that the guardrails are in place. That was my sense of the people who worked there. I do not know if you want to say anything on that or on the other issues, Emran.
Emran Mian: I can add a bit on the opportunities, if that is helpful. It is in three buckets, to go back to the buckets metaphor. The first bucket is on how we are using AI in public services. There is a big drive on that. For example, at the moment, we are testing an AI chat interface for GOV.UK. It is testing very well, and we are hoping to be able to roll that out across the GOV.UK domain very shortly. It will mean that when citizens are interacting with GOV.UK, they will not have to rely on search or finding the bit of information they are looking for in the vast library of resources that are on GOV.UK, but they will be able to interact with it using a natural language chat interface.
Earlier, the Secretary of State mentioned the work that we are doing to bring AI tutors into schools to tackle educational disadvantage. There is a massive focus on how we can use AI in public services to deliver tangible benefits for citizens.
The second bucket I would pick out is where we are focusing on existing areas of UK economic strength and the deployment of AI to support those areas of strength, whether that is on actual services, life sciences or other areas. That is where our sovereign AI effort is important. You mentioned advanced materials. There are a number of very promising AI start-ups in the UK, which are working with AI and advanced materials. We are talking to a lot of them, and we are interested in how we can support them, both in terms of procurement and early-stage funding.
The third bucket I would pick out is AI adoption across the wider economy, where—in addition to providing AI skills to a much wider range of the population—at the moment, we are observing lower-than-expected levels of AI adoption by businesses all across the economy. This is an issue that lots of countries are facing; it is not just the UK. We have now set out plans to have an AI adoption champion in each of the industrial strategy sectors, so we will have local advocates—people who are experts in particular parts of the economy—to rally companies together and work on that. That is definitely an area on which we should be looking to do more.
Lord Stern of Brentford: Is that the case in the big systems as it is in energy transport? That is not done by an individual country; that has to be done across—
Liz Kendall: That is partly why Barnsley became the UK’s first tech town. The leader of the council invited us to come, because they want to look at every angle—how the NHS might use AI more to free up doctors’ time as well as the research side; how in schools, the AI tutors can provide personalised support to children from disadvantaged areas; how to support businesses to take it up; and looking at what the council can do. That is the very reason why—to show that it works for real-life solutions. In the end, this has got to be useful to people, so that it benefits their lives, not just the lives of the few people who own these huge companies.
Lord Vallance of Balham: I will just add one thing with my DESNZ hat on. We have just started a piece of work, led by Lucy Yu, on AI in the energy system. It is exactly that point—to look at it as a complete system of where AI can change that. That should read out this summer.
Lord Stern of Brentford: Lucy?
Lord Vallance of Balham: Yu.
Lord Stern of Brentford: Oh yes, okay.
Q20 Baroness Jones of Whitchurch: I just have a very quick question. At the AI Security Institute, at one point it was said that it would be put on a statutory footing. Can you give us an update on your thinking on whether that is your intention—to make the regulation of the frontier sector statutory? If not, if it finds a problem—something that is really dangerous—what mechanisms does it have to intervene to stop that spreading further?
Emran Mian: At the moment we are relying on—and it has been very successful so far, since the advent of the institute—collaboration with companies that are producing frontier models, both pre-deployment and post-release. We have seen those companies respond really well to the feedback they get from the experts at the AI Security Institute and make changes to strengthen the guardrails on the models. A lot of that conversation has to happen in private, because it involves classified security material or commercially sensitive information.
Alongside that, we are very mindful of the fact that we should give as much information about the work that the AI Security Institute is doing as possible. A couple of weeks ago, the AI Security Institute published its latest report on the trends it was seeing in frontier AI. It was very open about the developing capabilities and the opportunities that come from that and, crucially, in laying out what some of the challenges and risks are. The report, for example, picked out the rapidly improving performance of frontier AI models on carrying out cyber tasks. That can be very beneficial in strengthening cyber security defences. It can also create opportunities for misuse.
We will continue to use our regulatory frameworks and check where those need to be strengthened to cover for these risks. We are taking legislation on cybersecurity through Parliament. The Secretary of State has outlined our plans to strengthen the online safety regime to pick up where we have identified gaps that now need to be tackled because of the development of AI.
Liz Kendall: Specifically, I am worried about AI chatbots for kids. We know that some AI chatbots are covered by the Online Safety Act if they use live search or if they share user-to-user content but, if it is just you and the chatbot, they are not. We will legislate in the Crime and Policing Bill to close the gap on illegal content for everybody—adults and children. We are taking powers in the children’s Bill through which we can implement the results of our big consultation on children’s lives online. We are looking at one of those because we have seen examples where children develop personal relationships in a very harmful way.
As Lord Drayson mentioned, on Grok, we were clear that what it was doing was not only against British values; it was against British law. We said that Ofcom would have the full backing of the Government to use its full powers in the Act. If we identify gaps where we believe there are things that are harmful, we have already shown, at least in terms of online safety for children, that we will act.
Baroness Jones of Whitchurch: So the plan is to do it case by case, rather than have an overall statutory responsibility?
Liz Kendall: Yes, we are definitely looking at individual cases where we can see that this is causing problems. That is our approach.
Q21 Lord Duncan of Springbank: Minister, I have a very quick question. Most of the AI innovators and companies are not British; they are American or from elsewhere. The leverage that you have to impose your conditions upon them depends upon your willingness to be quite strong. Do you anticipate banning some of these companies if they simply did not adhere to what you felt were British values?
Liz Kendall: We said, on Grok, that if it did not obey British law, Ofcom had the power to fine it up to 10% of its relevant worldwide revenue or introduce business disruption measures to stop it being available in the UK. We were quite clear that, where that happens, it is the job of Government to make sure that the law of the land is followed.
Lord Duncan of Springbank: I accept that. Grok was the one that you were focusing on, but most of the AI platforms were offering the same facilities. My issue is that if the US as a whole simply said no, pushed back on that and was willing to be diverted and move away from the UK, what would you do in that circumstance? There is no British equivalent to that platform.
Liz Kendall: I do not believe in a blanket approach. I believe in being really specific, where there are harms, especially if they are against the law in this country, that we should stand up for that. My position is that I just want to make sure that the law is followed. I can only give you that example, because that is my approach. It did change, but we will monitor it closely to make sure that it continues to do what it said it was going to.
Q22 Lord Ranger of Northwood: I appreciate you being so generous with your time. I have some quickfire questions. I have some declarations, because of these questions. I have spent most of my career in strategic supplies to Government, so I understand the Government’s procurement rules. Right now, I advise some firms that are selling to Government—hopefully in the right way.
Liz Kendall: Which ones?
Lord Ranger of Northwood: VE3, and Fuel Finder has recently come in. It worked very hard to get that in place. It has been hugely successful. There is also Scrumconnect. I have not been advising the big firms but the SMEs and the ones that are getting in there with new innovation.
Right now, you are pushing AI and Government, quite rightly, and I welcome that. Quickly, how do you think it is going? Which tools and applications have proved most useful so far? Where do you feel that you have got more to do around AI in Government? Fundamentally, what productivity measures and gains are you seeking to achieve?
Liz Kendall: There are lots of different views and forecasts about the impact AI will have on productivity. At the Budget, the OBR downgraded productivity from the last Government by 0.3%. It believes that AI will increase productivity by the end of the forecast period by 0.2%. The IMF says much more—it could be up to 1.5%. You know what economists are like—there are lots of different forecasts; some of my best friends are economists. There are many different views about that.
On what has been the most successful, I can give you three very quick examples. When I was at the DWP, our amazing staff in our Welsh jobcentres had massive queues of people seeking help to write their CVs. You cannot be in a situation where people who need work are not getting help to write their CV. So, off their own back, they used AI to help people write their CVs. It was brilliant. Who would not want AI to help them write a CV?
I will give an example of a primary school in Leicester West. In the poorest part of my constituency, there is a fantastic school: Braunstone Community Primary School. It is using AI for personalised tutors for kids. Many of these kids come in and they start school 18 months behind. They use dummies, are not able to talk and have no fine motor skills. They cannot hold a pen. They have speech and language problems. The school has now got 80% of those kids up to the national average through personalised tutors and support. The teachers love the planning, because they do not have to do all that; they are present in the classroom. It is brilliant.
The third example I have seen of how good it is in the NHS, where we are using it for earlier diagnosis. We are using it so that GPs can spend more time face to face with their patient and not just on writing letters, freeing their time up. I am not starry-eyed about it, but it can help us provide more personalised support to individuals. We have got to get the best out of our public services, and we have got to do it quickly because people are impatient for change. Where it can work, we should use it. There are real benefits to real people.
Lord Ranger of Northwood: I agree with you in spreading it around the country with your AI growth zones. I should declare, I am supporting South Yorkshire in getting its growth zone. I must declare that, in Doncaster, I am advising Automated Analytics.
Q23 The Chair: Secretary of State, Lord Vallance and Mr Mian, you have been excellent in answering our many questions. I have one last question for you.
Liz Kendall: Chair’s privilege, of course.
The Chair: You mentioned your own experience of training for AI, and your predecessor, Peter Kyle, was very open about his use of ChatGPT in particular. Have you used the most up-to-date models? Are you using them?
Liz Kendall: I do. I do not use it for work; I use it for planning holidays.
The Chair: Excellent. Thank you very much for this session. It has been incredibly helpful to us. We will follow up in writing with some questions, probably particularly on the UKRI funding. We will probably follow up with some more detailed questions. I thank all three of you very much. That concludes the session.
