An Exeter digital economist last week appeared before the government's Public Accounts Committee to highlight some of the issues with deploying AI across public sector organisations.
The government has repeatedly stressed its desire to increase economic growth through the mass take-up of AI systems, including in the public sector, and an official plan published in January called for the government to "rapidly pilot" AI-powered services.
However, a new report from the Public Accounts Committee on the use of AI in government expressed a number of concerns about the speed of progress in preparing for AI, with too few departments reporting transparently on their use of AI and a shortage of digital and data skills in government.
These issues were highlighted by the University of Exeter Business School's Professor Alan Brown, who appeared before the Committee and submitted written evidence.
Professor Brown outlined three challenges the government faces in adopting and scaling AI technology.
The first issue Professor Brown described as a "misalignment" between how governments and AI firms operate, with government procurement cycles lasting months or even years, while AI capabilities evolve more rapidly.
"This misalignment can result in governments procuring solutions that may be outdated by the time they're implemented, or missing opportunities to leverage cutting-edge capabilities," Professor Brown said, adding that AI firms with agile business models who are used to private sector partnerships will struggle to adapt to the public sector's stringent security protocols, data sovereignty requirements and need for extensive accountability measures.
He also highlighted the difficulty of attempting to integrate modern AI solutions into the 'outdated architecture' of many government IT systems, describing a 'chicken and egg' situation where "AI adoption requires modernized systems, but the promise of AI could help justify and drive that very modernization".
Finally, Professor Brown also set out the "unique challenges" faced in the public sector to scale up the numerous successful AI pilots and proof-of-concepts that have run across government agencies.
He said: "The challenge of scaling these successes across broader government operations introduces much greater complexity in terms of change management, resource allocation, and risk mitigation.
"The transformative nature of AI technologies means that scaling efforts must address not just technical implementation, but also organizational culture, workforce adaptation, and public trust."
