By Farooq Shaikh | Senior Data Scientist, Kefron
Somewhere right now, an accounts payable (AP) clerk is keying invoice data into a system that could read it automatically. Perhaps someone else is chasing an approval by email that a workflow engine could have routed in seconds. No one chose a career in finance to do these tasks, and as technology evolves, they increasingly won’t have to.
Naturally this sparks fear: If artificial intelligence (AI) can do all of this, what is left for humans?
The truth is somewhat more nuanced. I build the AI systems that power these workflows and my perspective is that AI is not coming for AP jobs. It is coming for the parts of the AP job that nobody actually wants to do.
Accounting is deterministic but AI is probabilistic
Accounting, at its core, is a deterministic discipline. An invoice is either fraudulent or it is not. The ledger balances or it does not. There is no “probably correct” in audited accounts.
AI is built differently. Every prediction an AI model makes has a probability attached to it. Even the most capable models today can only infer and predict based on patterns. For example, a system cannot tell you with absolute certainty that two line items match, it can only tell you that it is 97% confident they should.
In accounting, this creates a critical gap. The output of an AP process lives in a ledger that must be right, with certainty. Even with impressive automation capabilities, a business processing thousands of invoices will have a meaningful proportion of items where a human must still step in, understand context, and take accountability for the system’s outputs. High automation and human oversight must work together.
Same story, different chapter
This isn’t the first disruption to finance. Years ago, automation software cleared out floors of office workers doing calculations manually. Yet, despite this, finance did not die as a profession. What looked like the end of a job at the time turned out to be the start of a more valuable one. Those who adapted moved into new roles involving analysis, forecasting, and decision-making that software could not handle.
Human judgment is becoming more valuable
As the cost of execution falls thanks to automation, the value of human judgment rises. When invoice routing becomes autonomous, AP professionals can stop spending days on process and start making important decisions. They can focus on quality control, stakeholder relationships and compliance. This is work that directly drives spend optimisation and better financial control.
Areas where human value will grow include:
- Cash flow strategy: Engaging meaningfully with treasury to leverage early payment discounts. In other words, AP moves from a downstream processor to an active contributor to liquidity strategy.
- Supplier risk: AI is good at flagging anomalies. It is not good at interpreting them (yet). Trust remains human-led.
- Complex exceptions: Navigating ambiguity in pricing disputes, partial deliverables or contract interpretation. Here AP professionals earn their place.
- Governance: Someone needs to define the policies AI agents operate under and audit their decisions.
How AP professionals can future-proof their careers in an AI era
To adapt, AP professionals must become curious rather than defensive:
- Build basic AI literacy: Understand how the software model extracts data and why it fails. This makes you difficult to replace.
- Own the exceptions: Become the expert who resolves the transactions that algorithms cannot. Regardless of automation, a good portion of invoices will need human judgment.
- Strengthen relationships: Leverage human connections because relationships are not automatable.
- Move up, not out: Understand cash flow, supplier strategy and compliance well enough to make decisions on information that AI can only surface.
Will some entry-level, manual data entry roles shrink? Yes, and it is important to confront that truth. But the core skills that make someone effective at AP, such as attention to detail, understanding controls and spotting anomalies, are not becoming obsolete. They are the foundation for overseeing a more powerful system.
The manual burden was never the point of the job. Removing it is long overdue. Done well, AI transforms AP from transaction processing into a powerful source of financial intelligence. The tools may handle scale, but people will handle consequence.
