Finance professionals used to learn their jobs by doing the work themselves. They processed invoices, reconciled accounts, investigated exceptions, chased approvals, and gradually developed the judgment needed to handle more complex situations.
The work was often repetitive, but it trained the next generation of finance leaders. Artificial intelligence is changing that process.
Across finance departments, AI systems are taking on tasks that were once considered entry-level responsibilities. They can extract data or validate information, route approvals, and even identify and solve issues without human intervention.
In invoice lifecycle management, that means AI is increasingly touching work that affects compliance, payment accuracy, supplier relationships, and financial integrity. For organizations under pressure to improve efficiency, the appeal is obvious.
However, if AI is taking the entry-level work, what happens next?
Where will future expertise come from?
Most conversations about AI and employment focus on whether jobs will disappear. But a different challenge might be emerging inside finance organizations. If software handles a large chunk of the work that traditionally taught people how finance operations function, where will future expertise come from?
“The current senior cohort built its judgment by doing the manual grind,” said Kevin Kamau, Director of Product Management at Basware. “The incoming cohort is AI-native and will never do that work. That raises an uncomfortable question. You cannot supervise what you never learned to do, so where does expert judgment come from once the apprenticeship that produced it has been automated away?”
The concern reflects a broader shift taking place throughout enterprise finance.
Organizations are no longer limiting AI to generating insights. Now, they are deploying systems capable of participating directly in operational processes.
Tasks that once consumed hours of employee time now take minutes or even seconds. That creates opportunities for greater productivity, but it also changes the role people play within those processes.
The wrong question about AI and jobs
Jason Kurtz, Chief Executive Officer of Basware, believes much of the public discussion around AI has focused on the wrong question.
“The ‘AI is coming for your jobs’ narrative makes for good headlines,” Kurtz said. “Fear, uncertainty and doubt sell stories. But we’ve seen this before. Every major technological shift sparks anxiety about widespread job loss. Technology evolves, roles adapt, and work changes, yet it rarely results in the wholesale replacement of people.”
Historically, a significant portion of accounts payable and finance operations involved processing transactions and resolving routine exceptions. As automation expands, organizations are discovering that the remaining work often requires a different set of skills.
“Agentic AI is taking over manual tasks like chasing invoices, checking data, and routing approvals,” Kamau said. “That allows AP professionals to focus on activities that directly impact the bottom line: cash flow optimization, supplier relationships, working capital, and risk. It also moves their role closer to protecting financial integrity, making sure transactions are accurate, compliant, and trusted before decisions are made or payments are released.”
The change sounds straightforward until organizations begin thinking about how expertise develops.
Supervising what you never learned to do
A finance professional responsible for overseeing AI systems still needs to understand the process those systems are managing. They need to recognize when something looks unusual, when a recommendation should be challenged, and when an exception requires human intervention.
That becomes especially important when AI is operating across complex workflows, ERP environments, approvals, compliance checks, and payment decisions. Those skills traditionally emerged through experience.
AI is removing more of the repetitive work that once served as an entry point into the profession. As it does, companies may need to rethink how they train employees and develop future leaders.
That challenge is one reason Basware recently launched an AI certification program designed for finance professionals. The initiative focuses on helping users understand how AI agents operate inside financial workflows and where human judgment remains essential. It also covers how employees can effectively supervise increasingly autonomous systems.
The scale of its invoice network shapes Basware’s perspective. More than 2.5 billion invoices inform its platform, and it supports more than one billion AI actions annually across its customer base. That gives the company a practical view of how AI changes finance work in production, not just in controlled pilots.
Treating AI agents as digital employees
For Kurtz, the objective of this course is not simply helping people use AI. It is helping them learn how to work alongside it.
“To get the most out of AI agents, organizations need to treat them as digital employees,” he said. “That means training people not just to use AI, but to manage it.”
After all, no finance leader would simply hand a new hire authority over millions in spend on day one. We build trust in AI much the same way we build it in people. Authority is earned by setting guardrails and monitoring performance. This governed approach allows finance teams to increase autonomy as trust and confidence increase.
The idea reflects a future in which AI agents become a routine part of finance operations rather than a specialized technology project. Employees may find themselves overseeing systems that handle large portions of transaction processing. They would remain responsible for governance, exceptions, and strategic decisions.
Donna Wilczek, Chief Product and Technology Officer at Basware, said that shift makes enterprise control essential.
“As AI becomes part of everyday finance operations, it’s not enough to ask whether it can complete a task,” Wilczek said. “You need to know how that decision got made and who’s on the hook for it. And you need people who’ll actually step in when something looks off. That is what separates useful automation from trusted autonomy.”
From advisor to operator
That transition is already influencing the skills many organizations value. Analytical thinking, business judgment, data literacy, and the ability to evaluate AI-generated recommendations are becoming increasingly important.
Organizations are asking finance professionals to understand systems, interpret results, and make decisions when situations fall outside established patterns. They are also asking them to understand controls, auditability, risk signals, and the impact those decisions have across the wider invoice lifecycle.
Wilczek believes the profession is moving toward a model where people spend less time executing processes and more time supervising them.
“The direction for AI is moving from advisor to operator,” she said. “What’s left for people is the hard, ambiguous, interesting work. That’s the better job, if you ask me.”
The long-term impact of AI on employment remains uncertain. What appears likely is that the path into many professions will look different than it did for previous generations.
Finance departments have spent decades teaching people by immersing them in the work itself. As AI assumes more of those responsibilities, organizations will need new ways to build judgment, experience, and expertise.
The companies that solve that challenge may discover that preparing people for an AI-driven future is not simply about adopting new technology. It is about ensuring the next generation learns enough about the work to know when the technology is right, when it is wrong, and when human judgment still matters.