We recently met with Chris Johnson, the Chief Scientific Adviser to the UK government’s Department for Science, Innovation and Technology (DSIT), to discuss the importance of data management and accessibility for UK growth.
As Chief Scientific Adviser (or CSA to use the acronym), Chris Johnson is responsible for bringing science into policy through evidence-based input, for bringing policy into science, as well as developing the next generation of scientists and engineers, and bringing them into government.
DSIT is now the digital centre of government, responsible for driving and accelerating science and innovation across the UK; so DSIT has a vast range of priorities and faces a number of key challenges.
Image: Queen’s University, Belfast
Chris explains, “The UK infrastructure is continually changing and yet parts of it haven’t changed for centuries, so it’s an incredible spectrum. The decisions we make today could affect the future of many generations. DSIT is central to the digital infrastructure of the country, which in turn is an absolutely critical asset for the country going forwards. We are increasingly involved in supporting the intelligence agencies and the national security agencies against external threats.We need to plan ahead for future capability like 6G, to accurately map our infrastructure, plan for contingencies and future national resilience, and more.”
So what does Chris see as the role of DAFNI in supporting DSIT’s priorities?
Chris explains, “Data is incredibly important in terms of the whole culture of the Chief Scientific Adviser and the GO-Science organisation. Data is often available but not known about or distributed throughout the system.
He adds, “Research platforms like DAFNI enable us to think through and work through what the implications of policy are, so that we can measure the services that we provide to the public in terms of things that have value, like whether you can go back to work or whether you can continue to study. There’s an increasing role for the exchange of data between companies, and there has been for at least a decade.
There are particular technologies, like homomorphic encryption, that enable you to compute functions without having direct access to the source data – and they are beginning to be used and allow organisations to participate in entities like DAFNI without compromising intellectual property.”
And where does the future National Data Library sit?
Chris says, “The Government holds, in public trust, a vast amount of data in order for us to deliver services to the public, and that data can improve the services we provide, and can reduce the cost of them. It can also create opportunities for charities or companies to have access to that data to make informed decisions that can improve services.
In our resource-constrained world it seems clear to me that, over the next 30-40 years, it’s data that will provide us with greatest lever to improve economic growth and to have an impact on healthcare and education, for example. Data allows us to see regional differences, differences in educational outcomes, differences in prosperity and health. And when you look at the data, you can go inside that and look at, say, whether gender plays a role in outcomes.
“We need data to see where the problems lie and to make improvements. The National Data library provides us with both the inspiration and lever to gain a more modern view of society and to really challenge the reasons why inequality exists in our society. The National Data Library provides a way of informing what we do in government but also in industry and charities.”
Do you have any final thoughts on how DAFNI might develop to support DSIT’s agenda?
“Going forwards there are at least two different types of research that DAFNI can support.
“The first is the mission-oriented focus, where the end effect for the public is what drives and inspires us and motivates us. The government is working on finding ways to link the research components with delivery organisations, and looking at how we overcome previous barriers to have a big impact for the public.
There is also a series of longer-term research objectives for the country where we’re blue sky thinking and considering long-term research agendas. These are just as important as linking to organisations that have a direct impact. Clarity over the purpose of research and not sacrificing one for the other is really important.
People and data and computing skills are also very important! Areas such as cyber security, machine learning, data stewardship… At the moment government data is relatively clean, without a lot of consequential inference within the data, but as machine learning develops more widely across other information sources, the provenance and quality of the data, whether it was derived through inference or derived through direct data collection, becomes increasingly important.
We can already see that large language models are suffering from learning from data that they themselves developed previously, the data now contains hallucinations, etc., so data stewardship and those type of skills, will be really important going forwards.”
Thank you Chris, we look forward to ongoing work with you and DSIT to support the UK’s research and innovation agenda