The use of artificial intelligence (AI) in finance functions has surged, with 58% of them now employing the technology in 2024, according to a new survey by Gartner. This marks a significant jump of 21 percentage points from 2023, underscoring the expanding role of AI in financial operations.
The survey, conducted in June 2024, gathered responses from 121 finance leaders and aimed to assess the level of AI implementation within finance departments.
“AI adoption in the finance function is advancing quickly,” said Gartner finance practice research senior director Marco Steecker. “It’s also encouraging to note that two-thirds of finance leaders feel more optimistic about AI’s impact than they did a year ago, particularly among those who have already made progress leveraging AI solutions.”
Uptake up, along with plans for AI expansion
Despite this increase, the survey shows that 19% of finance leaders reported no plans for AI implementation in 2024, down from 31% in 2023. Additionally, 21% of finance functions indicated that AI implementation is planned, a decrease from 30% the previous year.
Meanwhile, 32% of finance functions are developing AI pilots, up from 28% in 2023. Furthermore, the percentage of organisations using AI in production has decreased to 8% in 2024, compared to 20% in 2023.
Only 1% of respondents are focused on scaling AI to a larger group of users in 2024, down from 6% in 2023. The percentage of those who are uncertain about their AI plans remains constant at 2% for both years.
Among the finance functions already using AI, several key applications are prominent. Intelligent process automation is currently employed by 44% of finance functions. This involves utilising AI capabilities in existing automation tools, such as robotic process automation (RPA), to enhance data processing and automate repetitive tasks.
Another significant application is anomaly and error detection, where 39% of finance functions are using AI to identify and report inconsistencies in large datasets, such as internal claims, expenses, and invoices.
AI is also being used for analytics, with 28% of finance functions applying AI to improve financial forecasts and conduct deeper results analysis, enabling better decision-making. Operational assistance and augmentation, including the use of generative AI to replicate human judgment in operations, is utilised by 27% of finance functions.
However, the survey also highlights several challenges that finance leaders face in adopting AI. The primary obstacles include inadequate data quality and availability, along with low levels of data literacy and technical skills among employees. To address these challenges, chief financial officers (CFOs) must tackle three main issues, which are understanding the roles and skills necessary for AI implementation, attracting and retaining AI talent, and accelerating the development of AI capabilities within existing teams.
In terms of data management, Gartner experts suggest that the traditional “single version of the truth” philosophy—which aims for a perfect and singular source of data—may no longer be practical due to the sheer volume and variability of data in today’s environment. Instead, they recommend adopting a “sufficient versions of the truth” approach, which balances the need for data quality with its utility for decision-making purposes.
Market uncertainty about generative AI
In a separate survey focused on generative AI (GenAI), Gartner found that at least 30% of GenAI projects will be abandoned after proof of concept by the end of 2025. The main reasons cited for these failures include poor data quality, insufficient risk controls, rising costs, and unclear business value.
According to Gartner, one of the major challenges organisations face is justifying the substantial investment required for GenAI to enhance productivity, as the benefits can be difficult to directly translate into financial gains. While many organisations are using GenAI to transform their business models and explore new opportunities, these initiatives come with significant expenses. The technological research and consulting firm estimates the cost of deploying GenAI projects can range from $5m to $20m.
Read more: Just how high can Nvidia fly?
