AI for Finance: What CFOs Must Understand Before It Is Too Late

The conversation around artificial intelligence often begins with excitement and ends in uncertainty. There is broad agreement that it is reshaping finance, yet far fewer professionals can clearly explain what that transformation demands from those responsible for decision-making.

That gap shaped our most recent webinar session titled “AI for Finance: What CFOs Must Know and Act On,” and delivered through a collaboration between the ACCA Women Network Nigeria and dbrownconsulting, with David Brown facilitating the session. From the outset, the tone was deliberate. This was not positioned as a discussion on trends or emerging tools, but as a practical conversation about responsibility, leadership, and execution within finance functions.

In the opening remarks, Victoria Ajayi the host, made this intention clear by emphasising that the session was not about “buzzwords” or surface-level understanding, but about how finance professionals can actively shape the way artificial intelligence is applied within their organisations. That framing immediately shifted the discussion away from curiosity and towards capability.

AI Feels New, But It Is Not

Before exploring applications, David Brown chose to step back and address a common misconception. Artificial intelligence may feel recent because of its current visibility, but it is far from new.

As he explained during the session, AI as a field has existed for decades. Its current form is the result of multiple breakthroughs gradually coming together, including rule-based systems, machine learning, computational power, and the vast availability of data.

The difference today is not its existence, but how well it works and how widely accessible it has become. What used to sit within laboratories and specialised research environments is now being deployed directly into finance workflows.

The Shift Most Professionals Have Not Made

A critical point in the session was the way AI is still being misunderstood in practice. Most professionals continue to treat it as a tool, something to be used occasionally, often disconnected from core processes.

David reframed that view in clear terms:

“Artificial intelligence is not necessarily a tool. It is like you now have this extremely intelligent employee… but that employee needs guidance from you as the domain expert.”

This distinction determines how AI is approached.

When treated as a tool, usage remains inconsistent and shallow. When treated as an employee, expectations change. It must be trained, guided, reviewed, and held within a defined structure. That is where its value begins to compound.

The New Leverage in Finance Has Changed

One of the most practical insights from the session was how the barrier to entry has shifted. AI has effectively removed the need for technical programming skills in many finance-related applications. Instead, it has introduced a new requirement: clarity.

As David put it: “The new programming language now is English… AI has cracked language, so the way you communicate with it determines the quality of what it produces.”

This changes the nature of ability. The professionals who will benefit are not necessarily those with technical backgrounds, but those who can think clearly and communicate precisely. Instructions now drive outcomes.

Why Expertise Still Sits at the Centre

Despite its capabilities, AI does not eliminate the need for expertise. In fact, the session made it clear that expertise is now more important than ever.

AI can generate structured outputs quickly, but it cannot reliably validate them without oversight. This introduces a new category of risk.

David addressed this directly by saying: “AI is an amazing pattern predictor, but it does not judge. It can be confidently wrong and if you are not a domain expert, you may not even realise it.”

Outputs may look correct. Reports may appear complete. Calculations may align superficially. Without the ability to interrogate and interpret those outputs, errors can move faster and remain undetected for longer.

This is where the role of the finance professional becomes more, not less, critical.

The Foundation Most Organisations Overlook

At one point, the discussion returned to something fundamental but often ignored: data quality. Before organisations talk about adopting AI at scale, they must deal with the condition of their data.

“If your data is dirty, your output will be dirty. Before you start talking about AI transformation, fix your data.”

This principle is not new, but its consequence is amplified in an AI-driven environment. Poor data does not simply produce poor output. It produces poor output faster, at scale, and with greater confidence.

To bring structure to the conversation, David outlined how both AI and finance professionals evolve together.

AI progresses through clear stages. It starts as an assistant, moves into drafting work, develops into a watchdog, and eventually becomes an operator capable of executing tasks independently.

At the same time, the role of the professional shifts alongside it. The individual moves from being a preparer to a reviewer, then to an advisor, a judge, and ultimately an orchestrator of systems and decisions.

The implication is straightforward. If professionals remain at the level of execution, they risk being replaced at that level. Progression requires moving into interpretation, control, and oversight.

The Question That Changes Perspective

During the session, one scenario brought this idea into sharp focus. Participants were asked to consider whether they would sign off on financial statements that had been entirely prepared by an AI system.

The response was immediate and most of them declined. That hesitation was not based on whether AI could produce output. It was based on accountability.

This led to one of the most important statements in the session: “You can delegate to AI, but you cannot abdicate responsibility. The output may come from AI, but the accountability remains with you.”

Beyond efficiency and automation, the session also addressed the broader implications of AI within finance, particularly around risk.

AI does not operate in isolation. It introduces cybersecurity concerns, regulatory uncertainty, and exposure to new forms of operational risk.

David drew comparisons between AI and other powerful technologies, noting that its impact depends entirely on how it is used and governed. In finance, where decisions carry direct consequences, this responsibility cannot be delegated.

The conclusion was clear. Finance leaders must not only understand how AI works, but also how it can fail, and where those failures can have material impact.

What Separates Those Who Will Benefit

As the session progressed, a pattern became clear.

The advantage does not come from simply using AI frequently. It comes from using it deliberately.

Those who benefit most are able to:

• understand the financial systems behind the output
• structure clear instructions
• apply professional judgment
• and validate results before acting on them

Others may use the same tools but remain dependent on them.

Moving From Awareness to Capability

Understanding the shift is only the beginning. The real advantage comes from building the capability to operate effectively within it.

At dbrownconsulting, this is being addressed through a growing set of structured learning programmes tailored to different levels of leadership and application.

These include:

• AI for Work, focused on practical, day-to-day application across professional roles
• AI for Strategic Leaders, designed for decision-makers responsible for integrating AI into organisational direction
• AI for Board Members, built around governance, risk, and oversight at the highest level

During the session, there was significant interest in a dedicated AI for Finance programme, reflecting the need for a more focused approach within finance functions. That programme is currently in development, and further details will be shared once registration opens.

In the meantime, professionals looking to strengthen their financial capability alongside these emerging tools can explore the Project Finance training calendar, which provides the structured foundation required to interpret, evaluate, and apply financial models in real-world contexts.

Artificial intelligence is no longer a future consideration for finance. It is already shaping how work is done, how decisions are made, and how outcomes are evaluated.

The difference now lies in how it is approached. Some will use it as a shortcut. Others will take the time to understand it, structure it, and control it. Over time, that difference determines who remains responsible for decisions, and who simply participates in them.

For more details or enquiries, please send an email to [email protected]