AI tools are driving efficiency and strategic gains—and fundamentally reshaping finance professionals’ roles.
Artificial intelligence promises to ignite a revolution in the corporate finance function, enhancing efficiency, forecasting, and decision-making. As adoption accelerates, CFOs envision expanding their role from stewards of cost to drivers of longterm value as they integrate AI into both strategy and operations.
While the benefits are attractive—from improved cash flow to fraud prevention—accompanying them are cultural shifts, data challenges, and regulatory pressures. With governance practices and team capabilities evolving rapidly, finance professionals will have to adapt fast in an environment where strategic use of AI-driven technology is becoming a core differentiator.
The 2025 PrimeRevenue Annual CFO Trend Report notes that companies thriving in today’s volatile markets are those that are rethinking how capital flows through their operations. AI is leading this shift, transforming business-to-business payments through fraud detection, predictive analytics, and dynamic discounting.
“It’s not about replacing human judgment but extending it. You need to be human-centric.”
Raphael Savalle, former CFO at Weleda
Firms using AI in accounts payable, for example, are enjoying at least a $3 million return on investment (ROI) over five years from improved forecasting and stronger fraud prevention, according to the report. In retail, 60% prioritize digital transformation while logistics—key to global trade—is embracing AI and automation to ease cash-flow pressures, enhance efficiency, and boost supply chain visibility through early payment and predictive tools.
Armand Angeli, AI and Automation Specialist & Vice President of Finance Transformation and International Groups at DFCG
The reality of implementing AI across the finance function reveals intricate operational and cultural challenges, however. Unlike the earlier robotic process automation (RPA) wave, says Armand Angeli, an AI and automation specialist and vice president of Finance Transformation and International Groups at DFCG, the French network of CFOs and controllers, today’s AI is highly complex, which can be intimidating for finance professionals.
“As a CFO, the priority is to understand the substance behind the hype,” he says. “In practice, most CFOs are skeptical—particularly of GenAI and agentic AI. Predictive AI is more widely trusted, mainly as it doesn’t hallucinate. This form of AI is already proving useful in fraud detection, bank reconciliations, and smart cash posting.”
Pockets Of Innovation
Tangible value, increasingly, is outflanking the skeptics, particularly when it comes to cash flow forecasting, says Alexandros Koliavras, president of the Hellenic Association of Treasurers (HAT) and deputy chair of the European Association of Corporate Treasurers. “In the Americas and Europe, AI-driven models adjust in real-time based on internal and macroeconomic inputs, proving useful for liquidity planning,” he notes.
AI in transaction categorization and anomaly detection is another innovation helping finance teams identify risks early. And in some US banks, AI copilots are being tested to assist treasury teams during liquidity stress tests, adjusting scenarios as needed and effectively embedding innovation in decision-making.
“These tools don’t just replace manual work, they’re changing how professionals interact with data,” Koliavras says.
In the US and Europe, treasury teams are integrating AI copilots into their treasury management systems, where they advise on optimal funding or investment decisions based on real-time data. In Greece, HAT’s members are eagerly adopting best practices, with some firms piloting machine learning to optimize payment runs and detect fraud patterns.
Whether driven by strategic foresight or the need to stay ahead of the technological curve, efforts to embed AI into finance are accelerating globally.
Raphael Savalle, former CFO of beauty products maker Weleda, made AI a strategic priority at the Swiss-based global company. “It’s about staying competitive,” he says. “Everyone must adapt; individuals, firms, the entire profession. People also want to work in companies that embrace the latest tech. You have to get on the technology train.”
Having introduced an in-house GPT for firm-wide use, Savalle brainstormed deeper uses for the tool with the company’s head of data and digitalization. He also launched an inventory management initiative.
“For those kinds of applications, you need two to three years of data,” he says, “and the more granular, the better. More is more when it comes to AI.”
GenAI has transformed the company’s monthly financial reporting over 20 countries, covering profit and loss, receivables, and budgets. What took days now takes seconds, says Savalle, delivering 80% to 90% usable output in 10 seconds. Next up: embedding AI deeper into enterprise resource planning (ERP) for predictive analytics, inventory and lead time management behind the scenes.
But while innovation grabs the headlines, he cautions, it’s important not to neglect existing systems. “You need to change the wheels of the car while driving the car!” he says. “Weleda has 23 ERP [platforms] today; you can’t realistically aim to apply AI across everything at once. First, you need to get the foundation right. Then you can build new, innovative solutions on top—but at the same time. Only then can you harvest the benefits from this new foundation.”
Integration Challenges
Infrastructure is widely considered one of the biggest hurdles for AI, especially at large organizations where processes span multiple departments—procurement, logistics, finance—and multiple legacy systems. Some ERP and procurement platforms will have been in place for decades, often siloed and poorly integrated, but they are business-critical.
For Angeli, the challenge lies in building bridges between these systems and the new technologies.
“This is a serious concern not just for corporates but also for banks, many of which still operate on systems like SAP,” he warns. “Integration is far easier for start-ups unburdened by legacy infrastructure.”
The finance function is a critical inflection point in this respect, Koliavras argues.
“Technologically, CFOs and finance transformation leaders must think modularly,” he says. “Instead of large, monolithic system overhauls, the future lies in composable architectures: plugging AI into existing systems in ways that deliver fast ROI and are flexible to scale.”
CFOs are also coming to realize that AI doesn’t yield a straightforward ROI; rather, the real value lies in long-term gains in productivity, engagement, and smarter resource use. Forward-looking CFOs are budgeting for AI with a broader lens, focusing less on quick returns and more on sustainable efficiency and enablement.
Shifting The Finance Culture
The role of the finance function, meanwhile, is being transforming as automation and AI reshape roles, especially for accountants and mid-level managers. Leadership increasingly demands a proactive approach to innovation and a deep understanding of how technology is reshaping finance.
“We’re rethinking how we interact with banks, customers, suppliers—everything is evolving,” says Angeli, “and that kind of transformation requires adaptability: not just in process but in mindset and leadership.”
Hybrid roles like finance data translators, digital controllers, and treasury analytics leads are emerging, blending finance expertise with AI and data science knowledge. More uncomfortably, the emergence of these new roles is creating pay disparities in some firms, with finance managers earning less than some data engineers, which poses leadership challenges for CFOs.
CFOs must build data science literacy, uphold data integrity, and deepen their understanding of AI algorithms. Close collaboration with data and IT teams is essential, as is a solid grasp of compliance requirements. They also need to ensure transparency in how AI handles sensitive customer, employee, and supplier data—keeping people firmly at the center of decision-making.
Despite the strides many companies are making, AI adoption still faces resistance, particularly on boards wary of its reputation and risks. AI errors can spread quickly in finance, so strong controls, traceability, and human oversight are vital.
“It’s not about replacing human judgment but extending it,” Savalle says. “You need to be human-centric to explain how AI is going to help. In any large company, there will be advocates of change and others will be less enthusiastic. It’s challenging to convince a very conservative board member that we may not be artificial, but we need more intelligence.”
