For years, work in the finance department has been synonymous with month-end overtime. Bank reconciliations, matching invoices, preparing journals, and tidying up reports before closing the books. Now, some of that burden is starting to shift to machines. The question is no longer whether AI will enter the finance function, but which parts should be handed over and which parts should remain in human hands.
Recent data shows that this shift is not just a discourse. Citing the Stanford AI Index 2025 report, AI adoption in the global business world has risen to 78 percent of organizations, up from just 55 percent previously. In the finance sector, the impact is quite measurable. A study of 500 companies using AI in the accounts receivable process noted productivity increases of up to 82 percent and operational efficiency improvements of up to 60 percent.
The Fastest to Automate
Looking at the adoption patterns, the first areas typically affected by AI are repetitive and rule-based tasks. Accounts payable, bank reconciliations, invoice recording, and expense management. Generative language models are already being used to read hundreds of lines of data, draft narrative reports, and flag transactions that appear suspicious before further examination.
Interestingly, automation here is not about replacing accountants, but about cutting down the time spent on administrative tasks. When data input and number matching are completed more quickly, there is more room for strategic analysis. It is no longer about who is the most meticulous at retyping numbers, but who is the sharpest at interpreting them.
Where Humans Still Hold Control
At this point, many people misunderstand. The ability of AI to read data quickly is not the same as the ability to assess it. Financial reports are subject to accounting standards, business context, and legal responsibilities that cannot be delegated to algorithms. A number that appears correct according to the machine may not necessarily be substantively accurate.
KPMG in its Global AI in Finance 2026 report notes that 95 percent of finance leaders plan to use generative AI in financial reporting within the next three years. However, at the same time, about one-third of organizations are actually increasing human oversight or human in the loop as a direct response to the risks of AI output. The greater the role of AI, the more important professional skepticism becomes to verify the results.
Academic literature in Indonesia has reached a similar conclusion. Research on AI in accounting emphasizes that data quality and governance are determining factors in whether AI truly enhances report quality, rather than just speeding it up. Accounting regulations that are not yet fully prepared to accommodate AI-based outputs also remain a point that cannot be overlooked.
The Challenge is Not Just Technology
Interestingly, the biggest obstacles are often not the tools themselves. At the Deloitte Indonesia CFO Forum 2026, Deputy Minister of Communication and Digital Nezar Patria reminded, "The biggest barrier is not technology, but leadership and organizational culture. Many AI initiatives stall at the pilot stage without making a real impact."
He mentioned three recurring challenges: the trap of pilot projects that never scale, the foundation of data that is not yet clean and integrated, and human readiness. Regarding the last point, Nezar emphasized the importance of a human in the loop approach. According to him, the true value of AI emerges when human capacity also grows, rather than being sidelined from the process.
For finance professionals, the message is quite clear. AI is most useful when used as an assistant that speeds up routine work, while the final judgment, context interpretation, and accountability remain in human hands. The skills that need to be honed are no longer just about inputting numbers, but about critically reading machine outputs and knowing when not to trust them blindly.
Sources
- Ministry of Communication and Digital (Komdigi), AI Adoption Reduces Costs by Up to 40 Percent, Corporate Finance Functions Change Rapidly (April 15, 2026). portal.komdigi.go.id
- KPMG, Global AI in Finance Report 2026. kpmg.com
- Jesya, STIE Al-Washliyah Sibolga, Artificial Intelligence (AI) in Accounting: Opportunities and Challenges for the Accounting Profession (2025). stiealwashliyahsibolga.ac.id
- Fiskusnews, Transformation, Not Extinction: The Future of the Accounting Profession in the Era of Artificial Intelligence. fiskusnews.com