Dive Brief:
- Only 14% of CFOs completely trust artificial intelligence technology to deliver accurate accounting data on its own, according to a Wakefield Research study released Wednesday.
- The vast majority (97%) of survey respondents said human oversight is critical for ensuring data accuracy in accounting when AI is deployed.
- “While most finance teams are already using AI tools, confidence in accuracy is not absolute, and many have experienced unreliable or hallucinated outputs,” the report said.
Dive Insight:
The study, commissioned by technology vendor Maximor AI, points to a “massive trust gap” that could hinder CFOs’ ability to fully execute their AI ambitions, according to the company.
The research comes as experts predict that many business leaders will be forced to sharpen their AI strategies this year, with a keener focus on data accuracy risks and other challenges.
Despite the potential hurdles, more than half (54%) of finance chiefs say integrating AI agents in their departments will be a digital transformation priority in 2026, Big Four accounting and consulting firm Deloitte found in its latest CFO Signals Survey released earlier this month. “This likely reflects growing interest in advanced tools to augment workflows and decision-making,” the report said.
Of the CFOs surveyed by Wakefield, 88% said they were already using at least one agentic AI tool in their accounting and finance processes. But less than half (40%) said they had fully integrated AI into their company’s finance function. Most respondents (86%) said their finance team had encountered at least one instance of inaccurate or “hallucinated” data while using AI.
“CFOs overwhelmingly agree that AI must know when to act autonomously and when to escalate decisions to humans to maintain control and auditability,” the report said.
Wakefield polled 100 CFOs at mid-market U.S. companies in late September.
