UK Government AI Trial Finds Copilot Reduces Civil Servants Workload

A recent trial conducted by the UK Government has revealed significant benefits from integrating AI tools into the daily workflows of civil servants. The study, which focused on Microsoft 365 Copilot, demonstrated a notable reduction in the time civil servants spend on administrative tasks, freeing them up for more strategic and impactful work.

This initiative underscores a broader governmental push towards leveraging artificial intelligence to enhance efficiency and modernize public services. The findings suggest that AI has the potential to not only streamline operations but also to improve the overall quality of service delivery to citizens.

AI’s Impact on Civil Servant Workload

The UK Government Digital Service (GDS) led a trial involving 20,000 civil servants across 12 departments and agencies. This extensive experiment, running from September to December 2024, aimed to assess the practical impact of Microsoft 365 Copilot on daily tasks. The results indicated that participants saved an average of 26 minutes per day using the AI assistant.

If these time savings were to be replicated across a full working year, it could equate to approximately 13 additional working days per employee annually. This substantial gain in time allows civil servants to reallocate their efforts towards more complex, value-added activities.

The benefits were perceived across various roles and organizational levels, with over 70 percent of users agreeing that Copilot reduced the time spent on mundane tasks and information retrieval. This suggests a broad applicability of AI tools in alleviating the administrative burden common in public sector roles.

Microsoft 365 Copilot integrates generative AI assistance directly into familiar applications such as Word, Excel, PowerPoint, Outlook, and Teams. By allowing users to perform tasks through a natural language chat interface, it bypasses the need for extensive mouse movements and menu navigation, thereby simplifying and accelerating many common office functions.

The trial also highlighted that AI tools could significantly reduce the time spent on specific routine tasks. For instance, drafting documents saw an average time saving of 24 minutes, while creating presentations saved users around 19 minutes. Even smaller tasks, like scheduling meetings, demonstrated consistent time savings of approximately 9 minutes.

These findings align with broader research suggesting that AI can support a significant portion of public sector work. A separate report by the Alan Turing Institute indicated that approximately 40 percent of public sector employee time could be aided by generative AI, primarily through task automation. This points to a considerable opportunity for AI to free up public sector staff from laborious duties.

Enhancing Productivity and Strategic Focus

The average daily time saving of 26 minutes translates into nearly two working weeks per civil servant per year. This reclaimed time is invaluable, allowing individuals to focus on higher-value tasks, innovation, or direct public service impact, rather than being consumed by administrative work. The potential for this to rise significantly if used across the entire civil service could transform productivity and public service delivery at scale.

A significant majority of trial participants, 72%, agreed that Copilot saved them time on mundane tasks. Furthermore, 71% agreed that using the system provided them with more time to allocate to strategic tasks. This dual benefit—accelerating administrative processes while creating more capacity for important, non-routine work—is a key advantage of AI integration.

The trial also noted that routine tasks, including consultations, were streamlined, and response times for job seekers were notably sped up. Extrapolating from the government’s figures, an estimated 30,000 civil servants annually could potentially avoid routine administration tasks altogether.

Microsoft UK CEO, Darren Hardman, commented on the experiment, stating that empowering employees with the right tools allows them to spend “less time on admin, more time delivering what matters.” This sentiment aligns with the government’s broader objective of modernizing the state and achieving substantial savings through technological advancements.

The integration of AI into the civil service is seen as a direct support for the government’s ‘Plan for Change.’ It aims to drive innovation, foster economic growth, and modernize public service operations. At Companies House, for instance, staff utilized Copilot to manage routine customer queries and expedite tasks such as drafting responses and updating records. Similarly, at the Department for Work and Pensions, work coaches employed AI to personalize advice for job seekers, ensuring faster and more tailored support.

Key Features and Applications of Microsoft Copilot

Microsoft 365 Copilot is designed to be an integrated assistant across the Microsoft 365 suite, enhancing productivity within applications like Word, Excel, PowerPoint, Outlook, and Teams. It operates by providing generative AI assistance, allowing users to accomplish tasks through natural language prompts rather than manual input.

In Word, Copilot assists in drafting and revising content, significantly reducing time spent on writing tasks. For Excel users, it simplifies complex data analysis by automating formula creation and generating actionable insights. PowerPoint users benefit from Copilot’s ability to help create visually compelling presentations from existing files or simple prompts, offering AI-driven design suggestions.

Collaboration in Teams is enhanced through Copilot’s capacity to summarize discussions and streamline project management. Furthermore, Copilot can summarize lengthy email conversations, draft emails with a specified tone and length, and help adjust the tone and structure for clearer communication in Outlook.

The tool’s voice-to-text capability is particularly powerful, enabling users to dictate notes during meetings or summarize training sessions, converting spoken words into clear text. This eliminates the need for manual note-taking, allowing for greater engagement during live interactions.

Copilot also excels in proofreading and editing, improving grammar, clarity, and flow across various written materials. It can reformat existing content, transforming a detailed document into a presentation outline or adjusting content length for different needs.

In Excel, Copilot can identify trends, create visualizations, and extract insights from complex datasets. It can also be used for brainstorming ideas by generating content outlines or research angles, and for tailoring content for different audiences by adjusting tone, style, or detail levels.

Challenges and Considerations for AI Adoption

Despite the positive outcomes of the trial, the adoption of AI in the public sector is not without its challenges. A significant hurdle identified is the prevalence of outdated legacy IT systems within government operations. These systems can hinder the quality of data essential for effective AI implementation.

Securing public trust is another crucial element for successful AI adoption. Progress in transparency regarding algorithm-assisted decision-making has been slow, with concerns remaining about the handling of sensitive data. Addressing public concerns over data privacy and security is paramount, especially given the sensitive nature of the information handled by government agencies.

The dominance of a few large technology suppliers in the AI market also poses a risk, potentially stifling competition and innovation. The government’s current procurement approach may lead to an over-reliance on specific companies, compromising adaptability.

Furthermore, persistent digital skills shortages present a major obstacle. Around half of the roles advertised in civil service digital and data campaigns went unfilled in 2024, with many departments facing challenges in recruiting and retaining AI-skilled staff. This skills gap is a key factor contributing to uneven AI adoption across the public sector.

Concerns about the reliability and safety of AI outputs also impact confidence in adoption. The UK’s Data and AI Ethics Framework provides principles and activities to guide the responsible development and use of AI in the public sector, aiming to help public servants understand ethical considerations and address them in their work.

Limitations were observed when dealing with complex, nuanced, or data-heavy aspects of work, indicating that AI is most effective when supporting rather than fully replacing human judgment in intricate scenarios. A proactive approach to AI adoption, focusing on both technological integration and workforce enablement through training, is essential for maximizing benefits while mitigating risks.

The Future of AI in the UK Public Sector

The UK Government’s ongoing exploration of AI, exemplified by the Copilot trial, signals a commitment to harnessing technology for public service transformation. AI offers the potential to improve government operations by providing policymakers with enhanced information and the ability to query AI for potential strategies.

The integration of AI can lead to better predictions, aiding in tasks such as external threat detection, health crisis management, and economic forecasting. By understanding potential future events more quickly, governments can make more informed decisions that minimize the impact of challenges.

AI’s role extends to streamlining decision-making processes, enabling quicker budget allocation and more efficient distribution of funds to dependent agencies and non-profits. This improved efficiency and automation of repetitive tasks can lead to significant time and cost savings for government entities.

Looking ahead, the UK Government’s AI strategy aims to foster responsible and transparent AI use to enhance public services and outcomes. This includes ensuring that public and civil servants have the necessary tools, information, and skills to leverage AI effectively. The potential for AI to contribute to economic growth and modernize public services is a key driver for its continued adoption.

The successful integration of AI hinges on addressing systemic barriers such as skills gaps, outdated technology, and procurement practices. Continued investment in training and development, alongside a focus on ethical AI principles, will be critical for realizing the full transformative potential of AI in the UK public sector.