Select Committee · Public Accounts Committee

Use of AI in Government

Status: Closed Opened: 31 Oct 2024 Closed: 16 Jun 2025 8 recommendations 19 conclusions 1 report

Government’s ambition is for the public sector to set an example in the safe and ethical deployment of artificial intelligence (AI), which both has the potential to transform services for the taxpayer and presents risks and concerns. Cross-government work began in 2023 on developing a draft strategy for AI adoption in the public sector. The …

Reports

1 report
Title HC No. Published Items Response
18th Report - Use of AI in Government HC 356 26 Mar 2025 27 Responded

Recommendations & Conclusions

27 items
2 Conclusion 18th Report - Use of AI in Government Accepted

Update committee on Algorithmic Transparency Standard compliance and high-risk AI spend controls.

Public trust is being jeopardised by slow progress on embedding transparency and establishing robust standards for AI adoption in the public sector. Public confidence that the AI technology used by government is fair, accurate, secure and safe is key to successful adoption. Transparency is fundamental to building that trust but …

Government response. The government has updated internal guidance for AI spend assurance, with spend controls now checking Algorithmic Transparency Recording Standard (ATRS) compliance, and DSIT will provide an update in Autumn 2025 on further changes to spend assurance and service assessments.
HM Treasury
3 Conclusion 18th Report - Use of AI in Government Accepted

Explain how Digital and AI Roadmap reforms will resolve public sector skills gaps.

There are persistent digital skills shortages in the public sector and DSIT’s plans to address the skills gap may not be enough. 70% of government bodies responding to the NAO’s survey identified difficulties in recruiting and retaining staff with AI skills as a barrier to AI adoption. In 2024, around …

Government response. The government is implementing reforms including requiring digital leaders on executive committees and boards by 2026, refining pay and capability frameworks, and developing talent pipelines, with monitoring through bi-annual reporting and public updates in the forthcoming Digital and AI Roadmap.
HM Treasury
4 Conclusion 18th Report - Use of AI in Government Acknowledged

Establish mechanism for AI pilot learning and scale successful products across government.

DSIT has no systematic mechanism for bringing together learning from pilots and there are few examples of successful at–scale adoption across government. At the time of the NAO report (March 2024), levels of AI use in government were low, but 70% of surveyed government bodies were piloting or planning AI …

Government response. The government agrees but states a proposal is currently being developed within the Public Sector AI Adoption programme, which will define scope, secure funding, and build a minimal viable product for a mechanism to scale AI initiatives.
HM Treasury
5 Recommendation 18th Report - Use of AI in Government Accepted

Detail how AI sourcing framework will ensure value and competition for government procurement.

DSIT and Cabinet Office have a long way to go to strengthen government’s approach to digital procurement to ensure value for money and a thriving AI supplier market. The technology market is rapidly evolving with a shift from upfront capital purchases to subscription–based models, and greater consolidation of the supply …

Government response. The government agrees and commits to defining a digital procurement strategy focused on AI sourcing, publicly articulating an AI procurement framework, and providing training and support resources like an AI Playbook for government professionals.
HM Treasury
6 Conclusion 18th Report - Use of AI in Government Acknowledged

Ensure Government Digital and AI Roadmap has clear implementation plan with milestones and metrics.

Realising the benefits of AI across the public sector will require strong leadership from DSIT. Addressing the skills shortage, tackling the persistent issues with poor data quality and out–of–date technology, and providing the support and guardrails the public sector needs to build public trust and adopt AI safely and responsibly …

Government response. The government agrees, stating DSIT is exploring mechanisms to enhance digital expertise and integrate senior digital professionals, aligning with existing intent to require digital leaders on boards and a dotted reporting line to the Government Chief Digital Officer, but does …
HM Treasury
1 Conclusion 18th Report - Use of AI in Government Accepted

Committee hears evidence on the government's use of artificial intelligence.

On the basis of a report by the Comptroller and Auditor General, we took evidence from the Department for Science, Innovation and Technology (DSIT) and the Cabinet Office on the use of artificial intelligence (AI) in government.1

Government response. The government agrees with the implied recommendations, committing to GDS working with HMT to develop a methodology for tracking legacy system remediation funding and to improve cross-government data management practices, including for the National Data Library.
HM Treasury
7 Conclusion 18th Report - Use of AI in Government Accepted

Legacy IT systems pose significant risks to government AI adoption and cybersecurity.

DSIT told us that it was a matter of urgency that the issue of legacy systems in government is addressed, not only to take advantage of the opportunities offered by AI, but also to address other risks including cyber security vulnerabilities. It emphasised prioritising the “systems that have the most …

Government response. DSIT, working with HMT, has conducted an assessment of proposals to address legacy IT and will use this to inform HMT's prioritisation of funding. GDS will also work with departments to ensure the highest risk systems receive the funding they …
HM Treasury
8 Conclusion 18th Report - Use of AI in Government Accepted

Government introduces new review and tracking for legacy technology remediation funding.

A blueprint for modern digital government sets out government’s intention to identify remediation funding for legacy technology. DSIT told us the new Government Digital Service will work with HM Treasury to review all spending review bids for legacy remediation to assess their credibility and ambition, while also tracking the allocated …

Government response. DSIT, working with HMT, has conducted an assessment of proposals to address legacy IT and will use this to inform HMT's prioritisation of funding. GDS will also work with departments to ensure the highest risk systems receive the funding they …
HM Treasury
9 Conclusion 18th Report - Use of AI in Government Acknowledged

Inadequate access to good-quality data impedes government's AI implementation efforts.

Access to good–quality data was identified as a barrier to implementing AI by 62% of the 87 government bodies responding to the NAO’s survey.16 We received written evidence expressing a range of views on this topic, illustrating the complexity of the issue and the challenge government has in getting it …

Government response. DSIT will set out details of its programme of work to address barriers to data sharing in the upcoming Government Digital and Artificial Intelligence (AI) Roadmap, including work to improve cross-government data management practices with a focus on data quality, …
HM Treasury
10 Recommendation 18th Report - Use of AI in Government Accepted

DSIT highlights need for cultural change and new initiatives to overcome data sharing barriers.

We challenged DSIT on what it was doing to address barriers to data sharing in government and it told us a change in culture is needed and government bodies need to think about the collective interest of the public when assessing the risks of data–sharing.20 A blueprint for modern digital …

Government response. DSIT will set out details of its programme of work to address barriers to data sharing in the upcoming Government Digital and Artificial Intelligence (AI) Roadmap, including work to improve cross-government data management practices with a focus on data quality, …
HM Treasury
11 Conclusion 18th Report - Use of AI in Government Acknowledged

Public trust is vital for AI adoption but remains low, requiring greater transparency.

67% of government bodies responding to the NAO’s survey said that support from the centre of government in fostering public trust in AI was very important.23 Written evidence we received from a wide range of stakeholders emphasised the vital importance of public trust. For example, the Ada Lovelace Institute told …

Government response. The government acknowledges the importance of public trust and data privacy in AI adoption.
HM Treasury
12 Recommendation 18th Report - Use of AI in Government Accepted

The Algorithmic Transparency Recording Standard remains underused, hindering public sector AI transparency.

The Algorithmic Transparency Recording Standard (ATRS), is intended to support public sector bodies to improve transparency and provide information about the algorithmic tools they are using, but the NAO found it was not widely used.27 We challenged DSIT on this lack of transparency as, at January 2025, only 33 records …

Government response. DSIT will update the Committee in six months (January 2026) on departmental compliance with the Algorithmic Transparency Recording Standard (ATRS) and further action it is taking to tackle gaps in transparency to strengthen public trust, including to address public concerns …
HM Treasury
13 Recommendation 18th Report - Use of AI in Government Accepted

DSIT acknowledges more work needed for AI transparency and redress mechanisms.

We asked DSIT for reassurances that there would be sufficient transparency and mechanisms for citizens to challenge AI assisted decisions. It told us that there were provisions in the Data (Use and Access) Bill to allow for redress and challenge in cases of automated decision–making. It also acknowledged that it …

Government response. DSIT will update the Committee in six months (January 2026) on departmental compliance with the Algorithmic Transparency Recording Standard (ATRS) and further action it is taking to tackle gaps in transparency to strengthen public trust, including to address public concerns …
HM Treasury
14 Conclusion 18th Report - Use of AI in Government Acknowledged

Significant recruitment and retention challenges hinder government's adoption of AI and digital skills.

70% of government bodies responding to the NAO’s survey identified difficulties in recruiting and retaining staff with AI skills as a barrier to AI adoption.31 In January 2025, the government’s State of digital government review set out the digital and data skills challenge faced by government, reporting difficulties in recruitment …

Government response. The government acknowledges the digital skills gaps and is taking steps to implement reforms, including refining frameworks, developing talent pipelines, and encouraging skills exchange. More detail is promised in the forthcoming Digital and AI Roadmap.
HM Treasury
15 Recommendation 18th Report - Use of AI in Government Acknowledged

Addressing the AI skills gap is crucial for responsible and ethical public sector AI adoption.

Written evidence we received emphasised the importance of addressing this skills gap, raising concerns about the public sector’s ability to take advantage of the opportunities of AI responsibly and ethically without 28 Qq 10–11 29 Q 57 30 Q 20 31 C&AG’s Report, figure 9 32 DSIT, State of digital …

Government response. The government agrees with the recommendation to address the digital skills gap in the public sector but provides only general plans such as refining frameworks, developing talent pipelines, and encouraging skills exchange. More detail is promised in the forthcoming Digital …
HM Treasury
16 Conclusion 18th Report - Use of AI in Government Accepted

Government's long-standing digital and data skills gap requires renewed efforts and diverse initiatives.

The digital and data skills gap in government is long–standing and this committee has raised concerns about this issue in the past.35 We challenged the Cabinet Office and DSIT on how they were tackling the problem and they told us they need to do more to communicate the benefits of …

Government response. DSIT is implementing reforms, including refining the Government Digital and Data Pay Framework and Capability Framework, to address digital skills gaps. Talent pipelines are being developed and cross-sector skills exchange is encouraged through initiatives like TechTrack and the Digital Secondments …
HM Treasury
17 Conclusion 18th Report - Use of AI in Government Accepted

DSIT's blueprint prioritises reforms for digital leadership and employment models to address skills gaps.

DSIT’s A blueprint for modern digital government also sets out a series of priority reforms. These include commitments to assess the optimum employment models and the competitiveness of the overall package for digital and data staff. To strengthen digital leadership, the blueprint also sets out plans to require all public …

Government response. DSIT is implementing reforms, including refining the Government Digital and Data Pay Framework and Capability Framework, to address digital skills gaps. Talent pipelines are being developed and cross-sector skills exchange is encouraged through initiatives like TechTrack and the Digital Secondments …
HM Treasury
18 Conclusion 18th Report - Use of AI in Government Acknowledged

AI deployment not widespread across government, with most bodies still piloting or planning use.

At the time of the NAO report in March 2024 AI was not widely used across government, with just over a third (37%) of 87 government bodies surveyed reporting they had deployed AI. In contrast, over two–thirds (70%) were piloting or planning AI use.41 Examples of pilot activity included use …

Government response. The government acknowledges the committee's observation about the new Government Digital Service and its aim to unite efforts to grasp the opportunities of technology and AI.
HM Treasury
19 Conclusion 18th Report - Use of AI in Government Accepted

Systematic knowledge sharing on government AI adoption remains absent.

The NAO report set out the importance of knowledge sharing to learn from pilot activity, encourage innovation, promote consistent standards, and avoiding duplication of effort in AI adoption across government. However, it also reported that there was no systematic dissemination of knowledge on AI opportunities and adoption across government.46

Government response. DSIT is establishing workstreams to gather and share insights from AI pilots, including the Prime Minister's AI Exemplars, to identify conditions for successful scaling. This includes assessing maturity, identifying core components for scaling, resolving bureaucratic blockers, and creating guidance with …
HM Treasury
20 Conclusion 18th Report - Use of AI in Government Accepted

Government still faces challenges in robustly evaluating and sharing AI pilot learning.

We questioned DSIT on how it is evaluating and sharing learning from AI pilot activity across government to avoid reinventing the wheel and to support AI adoption at scale. It told us that it was taking a range of approaches including developing guidance and identifying good practice case studies, establishing …

Government response. DSIT is establishing workstreams to gather and share insights from AI pilots, including the Prime Minister's AI Exemplars, to identify conditions for successful scaling. This includes assessing maturity, identifying core components for scaling, resolving bureaucratic blockers, and creating guidance with …
HM Treasury
21 Conclusion 18th Report - Use of AI in Government Accepted

Government AI procurement processes remain ill-suited to dynamic technology markets.

In its January 2025 State of digital government review, DSIT reported that government procurement processes had not adapted to the changing nature of the technology market, which has seen a shift from upfront capital purchases to subscription–based models, and greater consolidation of the supply chain.51 Written evidence we received from …

Government response. DSIT will publicly articulate an AI procurement framework, aligned with the overarching digital technology sourcing strategy, with a focus on Digital Partnerships and collaboration with Crown Commercial Service. DSIT will also help technology and procurement professionals work in a unified …
HM Treasury
22 Conclusion 18th Report - Use of AI in Government Accepted

Departments are strengthening digital procurement to balance centralisation and SME opportunities.

We questioned DSIT and the Cabinet Office on how they were strengthening digital procurement in government to support competition and innovation. They told us these issues were an ongoing challenge and that they wanted to find a balance between guiding procurement centrally without stifling opportunities to buy locally from SMEs.54 …

Government response. DSIT will publicly articulate an AI procurement framework, aligned with the overarching digital technology sourcing strategy, with a focus on Digital Partnerships and collaboration with Crown Commercial Service. DSIT will also help technology and procurement professionals work in a unified …
HM Treasury
23 Conclusion 18th Report - Use of AI in Government Accepted

Government is improving efforts to leverage public sector buying power for digital services.

The State of digital government review reported that government is not doing enough to ensure the public sector benefits from the scale of its buying power.56 The Cabinet Office and DSIT told us they are working to tackle this, giving as an example a recent agreement reached with 48 Q …

Government response. DSIT will publicly articulate an AI procurement framework, aligned with the overarching digital technology sourcing strategy, with a focus on Digital Partnerships and collaboration with Crown Commercial Service. DSIT will also help technology and procurement professionals work in a unified …
HM Treasury
24 Conclusion 18th Report - Use of AI in Government Acknowledged

AI governance fragmented across government, but recent departmental transfers consolidate responsibility.

At the time of the NAO report, responsibility for AI in government was split across the Cabinet Office—which was primarily responsible for AI adoption in the public sector, through CDDO, i.AI and the Government Digital Service (GDS)—and DSIT, which held responsibility for wider AI policy. The NAO concluded that limited …

Government response. The government acknowledges the committee's observation about the new Government Digital Service and its aim to unite efforts to grasp the opportunities of technology and AI.
HM Treasury
25 Recommendation 18th Report - Use of AI in Government Accepted

Departmental moves aim to resolve AI accountability issues, but full impact unproven.

We questioned the departments on what benefits we would see from these departmental moves, and the Cabinet Office told us that the aim was, in part, to address the issues of accountability and complexity highlighted by the NAO report which it felt the changes had pretty comprehensively addressed.64 DSIT also …

Government response. The government is introducing a requirement for all executive committees to include digital leaders and for boards to appoint digital non-executive directors by 2026, refining the Government Digital and Data Pay Framework and Capability Framework, and developing talent pipelines through …
HM Treasury
26 Recommendation 18th Report - Use of AI in Government Accepted

New inter-ministerial group and political leadership are key for AI coordination.

The Departments told us that the departmental moves had been supported by a new digital inter–ministerial group (IMG) which provides greater focus and links the work with wider cross–government work on public service reform and the government’s missions.66 DSIT emphasised that this was about “about having multiple members of the …

Government response. The government is introducing a requirement for all executive committees to include digital leaders and for boards to appoint digital non-executive directors by 2026, refining the Government Digital and Data Pay Framework and Capability Framework, and developing talent pipelines through …
HM Treasury
27 Recommendation 18th Report - Use of AI in Government Accepted

Public sector digital leadership found inadequate, DSIT to mandate roles and publish roadmap.

We challenged DSIT and the Cabinet Office on the scale of the challenge and whether the public sector would be able to grasp the opportunities offered by AI. DSIT told us that it had to be self–critical about digital leadership across government, observing that digital leaders are not well represented …

Government response. The government is introducing a requirement for all executive committees to include digital leaders and for boards to appoint digital non-executive directors by 2026, refining the Government Digital and Data Pay Framework and Capability Framework, and developing talent pipelines through …
HM Treasury

Oral evidence sessions

1 session
Date Witnesses
30 Jan 2025 Cat Little · Cabinet Office, David Knott · National Lottery Community Fund, Sarah Munby · Department for Business, Energy and Industrial Strategy View ↗

Correspondence

1 letter
DateDirectionTitle
1 Dec 2025 To cttee Letter from the Permanent Secretary at the Department for Science, Innovation a…