The vast majority of artificial intelligence investments are failing to generate a financial return for firms, according to new research.

In a study of 500 UK-based business leaders commissioned by digital product studio Studio Graphene, 78 per cent of respondents said their firm had adopted AI tools.

Businesses employing between 100 and 249 staff were the most likely to invest in AI technology, with an adoption rate of 85 per cent.

Among the small minority of firms yet to embrace AI, many are considering changing this. Fourteen per cent of respondents said they are either open to researching AI use cases or planning to adopt the technology over the coming year.

Not all firms are convinced by AI’s transformative potential, however. Eight per cent of UK business leaders have yet to use the technology and do not plan on changing this any time soon.

Caution around AI adoption may be justified, as the research shows that just 31 per cent of UK firms are currently experiencing returns from their AI investments.

A further 18 per cent of respondents said their AI investments have not been as beneficial as originally expected. Some UK leaders are more optimistic, with 16 per cent believing they need more time to determine whether their AI projects have been successful.

Determining the success of an AI project may be difficult, however, with 41 per cent of respondents struggling to define what this means in practice. Mid-size businesses, which are using AI the most, appear to have a better grasp of AI success, with 46 per cent of their leaders confident in defining a successful AI project.

Ritam Gandhi, director and founder of Studio Graphene, said: “There has been a rush to adopt AI amidst huge hype and a proliferation of new tools – this is certainly true of private equity-backed mid-sized companies looking to AI for automation, scalability and competitive edge.

“The problem, however, comes when AI is deployed without first defining where it sits within workflow, the decisions it will inform, the processes it will support, and the criteria for measuring success – often teams have not agreed whether AI is meant to save time, improve decision quality, reduce risk, support growth or all of the above.”

---