Among the takeaways from the Public Accounts Committee’s 30th January 2025 session looking at the government’s AI strategy is that achieving tangible results will require strong leadership, skills development and recognition that legacy IT needs updating.
During the session, Sarah Munby, permanent secretary for the Department for Science, Innovation and Technology (DSIT), said: “It’s always hard as an organisation to be truly self-critical, and I think we do have to be self-critical about digital leadership across government.”
Citing the State of the state report, she added: “Many non-digital public sector leaders with sizable delivery responsibilities have insufficient technical expertise and lack the digital orientation to implement tech-enabled programmes, or they don’t fully understand mission critical technology dependencies or high priority opportunities such as AI.”
Another challenge in the public sector, according to Munby, is that digital leaders are not well represented at executive level: “The report into the state of digital government says we have very serious challenges and an enormous amount of work to do to close the gap between reality and ambition.”
Speaking on the skills and hiring crisis, Catherine Little, chief operating officer for the civil service and permanent secretary for the Cabinet Office, said: “Just at macro level, we have got a huge challenge and we’re not alone. You know the private sector and other public sector organisations have exactly the same challenge in deploying these technologies at scale and making sure [they’ve] got professional experts at scale. We’re all competing for the same [people], and that’s a massive challenge for central government and for the civil service.”
During the session, David Knott, chief technology officer for the UK government’s Central Digital and Data Office, was asked about the technical challenges that government departments face. “I think it varies by department,” he said. “There are many challenges and they are not evenly distributed.”
Knott said that a series of workshops was run last year with departments to understand the challenges and obstacles they face. “Skills still remain a challenge, and the data the departments need to use to train models are often locked inside legacy systems,” he said.
Knott acknowledged that there has been “uneven progress” in remediating these legacy systems. The third area identified by the workshops is market maturity, since many AI products that could be deployed in government are developed commercially.
“Part of the reason we’re trying to find the right balance between guiding procurement centrally, without stifling the ability to buy locally from SMEs, is that lots of people want help to understand how they should think about the stability of the companies they’re buying from. We see lots of volatility in the market,” he added.
The fourth challenge Knott discussed is change management and how it relates to AI: “What people realise now is that these tools start to show up in people’s daily work and it can be a significant shift to their working patterns.” He compared the change as the shift to a graphic user interface from command line programs.