Finance teams embrace AI amid growing accuracy fears

Research from ACCA and Chartered Accountants Australia and New Zealand shows a sharp rise in the use of artificial intelligence in accountancy, while also highlighting widespread concern about the accuracy and integrity of AI-generated insights. The study found that 93 per cent of finance professionals have concerns about those outputs.

The survey of 1,600 finance professionals globally, including more than 360 in the UK, points to a finance function relying more heavily on real-time operational data and broader business information to support decision-making. More than 60 per cent of finance teams said they had increased their use of real-time data over the past two years.

That shift reflects a move away from finance teams acting mainly as reporting units focused on completed periods. Finance professionals are increasingly expected to provide current and forward-looking analysis, drawing on operational metrics, internal text data and AI tools to interpret larger volumes of information.

Changing role

The findings also suggest that traditional barriers between finance, data and IT teams are weakening. Almost 60 per cent of respondents said finance teams now work closely with data and IT colleagues, pointing to a broader organisational role as companies seek to extract more value from business data.

At the same time, concern about AI reliability remains widespread. Respondents cited AI hallucinations, inaccuracies, incomplete data sets, lack of transparency and bias among the main risks undermining trust in machine-generated analysis.

These concerns come as AI becomes more common in day-to-day finance work. Use of AI in accountancy has risen by 60 per cent in two years, according to the report, a pace of adoption that is outstripping the skills base in many finance departments.

The study found that 72 per cent of respondents reported only basic or no generative AI skills. A substantial minority, 41 per cent, said they were seeking training and upskilling in their own time, suggesting demand for practical education is emerging faster than formal employer-led development.

The report identified several pressures behind the wider use of data analysis in finance. Globally, 45 per cent of respondents pointed to strategic priorities as the main driver, while 43 per cent cited regulatory requirements.

Skills gap

Despite the increase in data use, finance teams still face structural obstacles. The most commonly cited barriers to improving business insight were data quality issues, mentioned by 42 per cent of respondents; a lack of appropriate skills, also cited by 42 per cent; and difficulty integrating multiple data sources, cited by 40 per cent.

Those findings reflect a broader challenge for the profession as the volume of available information expands. Finance teams may have greater access to real-time and unstructured data, but turning that material into reliable analysis requires stronger controls, clearer governance and staff able to test the quality of outputs rather than simply accept them.

The research also examined the talent profile of modern finance functions and found a mismatch between ambition and execution. Areas identified as weak points included generative AI literacy, predictive analytics, collaboration, storytelling, and data governance and ethics.

For professional bodies representing accountants, the issue is not simply whether finance teams are adopting new tools, but whether they can do so while preserving trust in the numbers and judgements they present to management. The high level of concern about AI integrity suggests many see that balance as unresolved.

"This research shows how finance teams are evolving from retrospective reporting engines into strategic enablers of enterprise-wide insight. This is a great opportunity, but upskilling is critical. CFOs and finance teams need to lead in the responsible adoption of AI across organisations, ensuring robust training and governance is in place. Critical thinking, sceptical validation and an ethical approach is vital," Helen Brand, Chief Executive of ACCA, said.

Ainslie van Onselen, Chief Executive Officer of Chartered Accountants Australia and New Zealand, linked the issue directly to judgement and risk management inside finance teams.

"AI is now a core part of the finance toolkit, but it's not a shortcut. CFOs and finance teams need to use it to sharpen judgement and generate real value, not just speed up old processes. That means investing in structured learning and working more closely with IT and data teams. Upskilling isn't optional. It's how you manage the risk," she said.

Overall, the report suggests finance teams are taking on a broader strategic role as data becomes more central to business decisions, but that transition is exposing weaknesses in skills, governance and data quality. With most respondents expressing concern about AI-generated outputs, the findings indicate that adoption has moved ahead of confidence in the systems being used.