UK Workers Spend Their Own Money on Workplace AI
Published by AINave Editorial • Reviewed by Ramit
UK workers are paying for AI tools they use on the job, even as many rely on free services or employer-provided access. A Deloitte workplace survey reports that one in six UK workplace AI users pays for at least one tool personally, with that group spending more than $1 billion a year collectively. The survey also found that 63% of UK working adults actively use generative AI for work. The survey findings and spending estimate point to adoption that is already happening beyond tools formally supplied by employers.
Personal, free and employer-provided tools coexist
Nearly half of workers surveyed said they use free AI tools, while 34% said they use generative AI provided by their employers. Those figures describe reported tool use; they do not establish that the categories are mutually exclusive. The Deloitte survey is described as drawing on 25,000 workers across the continent, while also being called the largest study of workplace GenAI use conducted in a single country. That description does not support treating all 25,000 respondents as UK workers. The survey’s reported usage figures and sample description are best read with that distinction in mind.
The spending total is collective, not an average subscription bill. Deloitte’s reported one-in-six figure applies to workplace AI users who said they paid for at least one tool themselves, and the excerpt does not give an individual spending amount. The headline figure therefore signals a meaningful pool of self-funded use, but not what a typical worker pays. The reported out-of-pocket share and annual total also do not explain why people chose personal tools.
Most reported use is still everyday work
The survey describes common uses such as drafting emails and using AI in place of search engines. Deloitte UK partner Paul Lee said workers’ use remained focused on searching for information and drafting content rather than more complex work. That distinction matters: broad workplace adoption does not, by itself, mean organizations have handed over complicated workflows to AI. The examples and Lee’s characterization of use describe practical, familiar tasks, not evidence that every worker uses AI in the same way.
Lee also pointed to limited training, patchy guidance and cases where employers did not know workers were using the tools. Personal, free and employer-provided access can therefore sit side by side, while employees may not share a common understanding of how to use AI at work. The survey excerpt does not quantify resulting security problems, so the supported concern is the guidance gap itself. When workers are already adopting tools on their own, workplace AI policy has to address use that may not begin with an employer-issued account. Lee’s comments on training, guidance and employer awareness make that gap part of the adoption story.
