Growing use of automation and digital technology could shoulder the burden of tackling fraud, while also offsetting the impact of staff turnover, according to minister for AI efficiency Alex Burghart
Artificial intelligence could reduce the need for civil servants in areas such as fraud detection, help drive up wages and tackle the loss of institutional memory, a Cabinet Office minister has said.
Speaking at an AI in Whitehall event at the Centre for Policy Studies, minister Alex Burghart, who has responsibility for public sector AI efficiency, set out some of the ways the government is exploring integrating the technology into its ways of working and why it is choosing to do this in-house.
Asked about the potential impact of AI on jobs in the civil service, Burghart said: “We’re always going to have a civil service. There are always going to be people working in it. But, at the moment, in some areas we employ thousands of people on fraud detection; we may or may not need to employ thousands of people to do fraud protection and in the future, I hope that that’s something that we can make infinitely easier and cheaper for the British public.”
The Department for Work and Pensions and HM Revenue and Customs have been asked to employ large numbers of extra officials to tackle fraud in the last few years, with DWP currently recruiting for thousands of jobs to investigate fraud in Universal Credit.
At the same time, the government is attempting to stop the rise of the civil service headcount through a numbers cap and a planned return to pre-pandemic levels which could mean around 66,000 job cuts.
Burghart said he could see AI eventually being used to reduce the size of the civil service – and increase wages.
“As we master this technology, you can certainly envisage a future in which you have a smaller civil service than the one you have today, that is better trained, capable of using the new technological capability that’s out there and, consequently, because you’ve got efficiencies, is probably better paid,” he said. “But that’s not tomorrow, that’s some way off.”
Burghart also set out how a trial of AI ministerial red boxes could help departments to retain institutional memory that is lost through high turnover of staff. In 2021-22, 46,000 officials left the civil service, the highest number in a decade.
Currently, a few ministers are using the digital ministerial briefcase, which uses AI to summarise documents that go into a minister’s red box. The technology will be fine-tuned and eventually offered to all ministers.
Burghart said he thinks the system could become “the institutional memory of the department”.
He said staff in the Cabinet Office, for example, “don’t always stay that long”, which can lead to a situation where the people who remember things that happened “three, four or five years ago” have left.
“But with an effective AI red box, that won’t be a problem. We will be able to retain the experiences of previous policies, previous successes, things that weren’t so successful, as well as consultations, policy, tech records, and so on.”
He said the technology will also free up time for officials working in ministerial private offices to work on other projects.
Burghart also said the government is attempting to develop new AI systems in-house rather than outsource after “a period where in-house has often been balanced or completely outstripped by outsourcing”.
“We’re having those conversations right now, building the systems right now, that we hope will enable us to go to Treasury for budget in spring and say ‘we are starting to prove the potential of these systems in in Whitehall’, and ‘help us go further’,” he added.
However, the Cabinet Office minister said the government also needs to continue to take advantage of technology being developed outside of government, picking out Microsoft’s Copilot chatbot as a tool which could “upskill huge numbers of officials” through its integration into Microsoft Office.