Google Cloud research highlights potential for GenAI to transform public sector service delivery

Encouraging the public sector to ramp up its adoption of automation and generative artificial intelligence (AI) tools could generate annualised savings of £38bn by 2030, according to Google Cloud research.

The public cloud giant commissioned policy research consultancy Public First to look into the potential benefits that expanding the use of AI and automation technologies could bring to UK-based public sector organisations.

As part of this process, 415 UK-based public sector workers were asked how they think AI could change the way the public sector works.

Their feedback was combined with modelling by Public First to conclude that increasing the adoption of AI in the public sector could save it around £38bn a year by 2030.

“By facilitating more effective working practices and automating repetitive, bureaucratic tasks, generative AI could save enough time to allow for an extra 3.7 million GP appointments, a 16% increase in the teacher-to-student ratio and free up the equivalent of over 160,000 police officers,” said Public First partner Jonathan Dupont in a blog post.

The feedback received from survey respondents further suggests AI could have a transformative impact on how “administrative tasks” are tackled in the near term, before going on to create efficiencies in more complex public sector workflows and operations.

For example, 81% of respondents said they expect AI to cut the time it takes to fill out paperwork, while 79% predicted the technology would aid with transcriptions and the taking of meeting notes.

“In the longer term, AI will enable more significant innovations: increasing efficiency, reducing waste, and ultimately allowing for [a] redesign of how public sector services and jobs roles work,” said the accompanying 49-page report by Public First.

“This will take a few years … as technology such as AI agents matures, and the wider public sector works to put in place supporting data and skills infrastructure.”

By facilitating more effective working practices and automating repetitive, bureaucratic tasks, generative AI could save enough time to allow for an extra 3.7 million GP appointments, a 16% increase in the teacher-to-student ratio and free up the equivalent of over 160,000 police officers

Furthermore, more than half (56%) of the respondents anticipate that AI will complement the work of public sector workers, rather than replace it, and assist them in meeting the sector’s ongoing challenge of having to do more with fewer resources.

For the time being, while there is a lot of optimism about the potential for AI to change the way the public sector works, adoption of the technology is still at relatively low levels. For example, just 12% of respondents said they had “significantly” deployed AI tools.

“Around half of those we spoke to (47%) thought AI adoption in their institution would take place slower than other technologies such as the cloud or social media,” said the report.

“When we asked them about the barriers to their institute making more use of AI, it became clear that this was not because they could not think of good use cases. Instead, the larger concerns seemed to be around security, legal requirements, lack of skills and reliability.”  

However, when it comes to unlocking the full potential of AI in the public sector, the report urges the government, public sector organisations and tech companies to collaborate and focus on addressing these concerns.

There were also concerns raised by respondents about how skilled public sector workers are in using AI, and also about the quality of the datasets they have at their disposal to train their models on.

For instance, the research highlighted that AI skills shortages exist in the public sector workforce, given that only 34% of public sector respondents said they were confident their employees had the right skills to use AI tools.

Meanwhile, 55% of public sector respondents said they would need access to different or better-structured datasets to fully take advantage of AI.

“AI will work best when existing public sector datasets are linked together in a secure and private way,” the report added. “The lesson from previous examples of digital government reform is that this often takes a strong lead and enforcement from the centre.”

Responding to the results, technology secretary Peter Kyle said the survey’s findings are a “powerful reminder” of the “revolutionary” impact GenAI stands to have on government services.

“Today we have an opportunity to drive forward public service reform by empowering citizens with the information and tools they need to make better choices and hold services to account,” he said.

“Now the digital centre of government, my department is testing how we can put AI to work in the public sector, whether that’s speeding up finding information on Gov.uk or empowering teachers by reducing administrative burdens, allowing them to dedicate more time to what they do best.”