A week after the U.K. government announced a sweeping plan to make big investments into AI, it’s laying out more details around how this will take shape in the public sector. On the agenda: AI assistants to speed up public services; data-sharing deals across siloed departments; and a new set of AI tools — dubbed “Humphrey” after a character on an old U.K. TV political sitcom — to speed up the work of civil servants.
The plans will be formally unveiled at a press conference Tuesday headed up by the Department of Science, Innovation and Technology (DSIT), along with two other departments, Work and Pensions and Health/Social Care.
Per the U.K. government’s AI site, the projects’ progress appears to be in the very early stages. For example, a plan to bring more AI services into the customer-facing side of the NHS are only at the stage of a “charter” committing to the concept. Other projects include links to Github repositories to check out some of the work to date.
It’s not clear how many people in total are working on these projects nor what third-party tools, such as LLMs, are being used. (We have asked these questions and will update readers when we learn more.)
At their heart, the projects are all about efficiency. The government, DSIT said, currently spends some £23 billion annually on technology, and the idea will be to redeploy that money in a more modern way.
“Sluggish technology has hampered our public services for too long, and it’s costing us all a fortune in time and money… Not to mention the headaches and stresses we’re left with after being put on hold or forced to take a trip to fill out a form,” said Peter Kyle, the Secretary of State for DSIT, in a statement. “My Department will put AI to work… We will use technology to bear down hard to the nonsensical approach the public sector takes to sharing information and working together to help the people it serves.”
The plans include a new team within DSIT to head the projects, a little like DOGE in the U.S. but conceived of and run by government people rather than tech moguls.
DSIT is honing in on three areas initially:
1. The work of government employees. Humphrey, named after the wry, clever assistant played by the late Nigel Hawthorne in “Yes, Minister” and then “Yes, Prime Minister,” is a set of apps aimed at reducing the typical daily workload of civil servants, much of it centered around vast amounts of data that they are required to read and process.
“Consult” is designed to read and summarise “thousands” of responses to consultations in hours. (Responses, which can be lengthy and numerous, are a central part of how the government takes feedback from stakeholders and the public into account.) “Parlex” will let government employees query and read conversations in Parliament that are relevant to bills or other policy documents they are working on. “Minute” is a secure transcription service to take notes from meetings. “Redbox” helps prepare briefings and policy documents. And “Lex” is focused on helping government workers find relevant legal data.
2. Another strand of the efficiency push will be around speeding up public-facing services. The idea here is to take aim at legacy bureaucracy, of which there is a lot in the U.K., such as the 100,000 calls that the tax authorities get daily, or the need for people to appear in person to register a death, or, bizarrely, posting ads in local papers as part of the process of getting a license to drive a truck.
DSIT’s thinking is that overhauling processes like these with more AI-fueled automation could save £45 billion annually. It’s not clear if that estimate is before or after deducting the cost of building and running the AI services.
3. A final area will be focused on more collaboration between departments in aid of sharing data to speed up how services are procured, and then how they work.
Taken together, the various projects are a signal that the government means business on its new AI push. But they also raise a number of questions.
For example, in the case of data sharing, DSIT says the operating idea here will be “a common-sense approach to sharing information.” Central government departments, like HMRC (revenue and customs) and the Department for Business and Trade, could, for example, share data with each other and local councils in fraud investigations. But what happens to data protection for individuals when data is shared in unintended ways?
Another possible question is around Humphrey: right now, DSIT said that some of the early apps are in testing phases only, but the big question will be how far will the government go in trusting some of the AI’s conclusions.There will also be more human challenges. As one former civil servant who now works for an AI company notes, past efforts to create programs that cut across departments have not always worked. Collaboration, money and authority are ultimately the levers that will make or break any of these plans.
