AI assistants could save government coders almost an hour per day, according to new trial results.
More than 1,000 tech experts across 50 government departments and agencies trialled Microsoft's GitHub Copilot and Google's Gemini Code Assist.
The trial aimed to understand how the coding assistant tools could help public sector engineering and development teams work better and be more productive.
An independent report on the trial, published by the Department for Science, Innovation and Technology and the Government Digital Service, found that trial participants saved an average of 56 minutes over the course of a working day when using the AI assistants. DSIT said this was equivalent to 28 working days a year on average.
More than half of the participants in the trial reported spending less time searching for information or examples, completing tasks faster, solving problems more efficiently and enjoying their job more.
The biggest impact reported was on the creation of code and analysis, where an average of 24 minutes a day were saved. Only 15% of code generated by the AI assistants was used without further edits being made by coders. DSIT said this showed that engineers were taking care to check and correct AI-generated code where needed.
The tools also received positive feedback from the users, with 72% agreeing they offered good value for their organisation, and 58% saying they would prefer not to return to working without AI assistance, whilst 65% reported completing tasks faster and 56% said they could solve problems more efficiently.
A lower proportion, 39%, believed their productivity would decrease if they lost access to the tool.
The report also found some limitations which made it difficult to fully measure the impact of the tools.
It said participants were not tracked across the surveys conducted at the start, middle, and end of the trial, and so “comparing changes in satisfaction or productivity over time proved challenging”.
The trial was also disrupted by the festive period, leading to reduced availability of both users and departmental resources.
The report also noted that it is possible that survey responses about time saved on different tasks "might overlap, potentially leading to some overestimation in the overall savings". And it said it is also possible that survey respondents overestimated time saved due to optimism bias.
However, the report said that, despite these limitations, “the findings strongly suggest that AICAs are valuable tools which can increase productivity and efficiency for government developers and engineers”.
“Further investigation could explore their impact across varying experience levels and refine methods for measuring time savings,” it added.
The report also said the trial would not influence ongoing or future procurement activities.
Commenting on the findings, technology minister Kanishka Narayan, a former civil servant in the Cabinet Office, said: “For too long, essential public services have been slow to use new technology – we have a lot of catching up to do. These results show that our engineers are hungry to use AI to get that work done more quickly, and know how to use it safely.
“This is exactly how I want us to use AI and other technology to make sure we are delivering the standard of public services people expect – both in terms of accuracy and efficiency. With a £45bn jackpot at stake, it’s not an opportunity we can pass up, as it can help cut backlogs and save money.”
The results follow those from a trial of Copilot run by the Department for Business and Trade which found reported time savings but no evidence of productivity benefits.