AI Can Write the Board Pack. Can You Defend the Numbers?

AI is changing what finance teams can achieve. But the real opportunity is not simply faster reporting — it is better financial decision-making, supported by reliable data, stronger processes and human judgement.

A board member looks at a polished financial report and asks a simple question:

“Where did this number come from, and what happens if the assumptions behind it are wrong?”

The report may have been prepared with AI. The commentary reads well, the charts are compelling, and the explanation sounds exactly like something an experienced CFO would write.

But can the finance team substantiate it?

This is where the next chapter of AI in finance becomes particularly interesting.

We have spent considerable time discussing what AI can automate. I believe the more important question is what it can help finance professionals achieve — and whether the underlying financial information is reliable enough to support better decisions.

What KPMG's latest research tells us

KPMG's recently published Global AI in Finance 2026 report provides compelling evidence of how quickly the profession is evolving.

Based on a survey of 1,013 senior leaders across 20 countries, it reports that 75% of surveyed organisations actively use AI in finance, compared with 30% in its 2024 research.

Meanwhile, 70% report improvements in decision-making quality and 71% in decision-making speed over the preceding year.

These findings are encouraging. But there is another side to the story.

Only 42% strongly agree that their organisations can produce AI-related audit evidence efficiently and without disruption, while just 29% formally track where AI adoption fails.

An important qualification: these are self-reported findings from large organisations, with technology and financial services accounting for 58% of the sample.

Comparisons with 2024 are directional because the research scope changed. The findings indicate meaningful patterns, rather than proving that AI itself caused the reported improvements.

Nevertheless, they highlight a fundamental challenge.

The ability to generate financial analysis is advancing faster than many organisations' ability to demonstrate its reliability.

And that distinction matters.

AI cannot compensate for weak financial foundations

One of the most interesting implications of KPMG's findings is the relationship between AI performance and financial discipline.

Organisations with more established data structures, controls and evidence-based processes generally report stronger outcomes.

This makes intuitive sense.

Imagine a company reporting declining gross margins despite growing revenue.

An AI tool might produce a convincing explanation involving pricing pressure, product mix or rising supplier costs.

But what if the underlying issue was an incorrectly recorded cost-of-sales accrual?

The explanation could sound entirely reasonable while being fundamentally wrong.

A capable CFO would first question the numbers, reconcile the movement and establish whether it reflected a genuine commercial problem or an accounting issue.

AI can accelerate analysis, identify potential explanations and help communicate findings. But it cannot automatically compensate for unreliable underlying information.

Equally, this should not lead us to dismiss its potential.

Used within properly structured workflows, AI can help finance teams introduce greater consistency, identify anomalies, explore scenarios and strengthen reporting processes.

AI alone does not create financial discipline. But combined with sound financial processes and experienced judgement, it can help make that discipline more effective and accessible.

That is a much more interesting proposition than automation alone.

From processing information to creating decision-making capacity

In my corporate finance advisory work, I have seen how AI can support the preparation and analysis of information, allowing professionals to spend more time challenging assumptions, evaluating alternatives and supporting better decisions.

The opportunity is to spend less time on preparation and more time understanding what the information actually means.

For financial professionals, that distinction is fundamental.

Consider financial forecasting. Reducing the time required to prepare a forecast is useful. But the greater value lies in being able to challenge the assumptions, assess alternative scenarios and understand the implications for liquidity, investment requirements or future funding.

The same applies to corporate finance and M&A.

Preparing information more efficiently has value. Identifying a potential weakness in sustainable earnings, working capital or an investment case before it becomes a transaction issue can be considerably more important.

This is where experienced finance professionals bring something that technology alone cannot replicate: commercial context, professional scepticism and accountability.

As AI becomes embedded in everyday financial workflows, I expect the value of those capabilities to increase rather than diminish.

The opportunity for growing businesses

There is another dimension to this transformation that deserves greater attention.

KPMG's research focuses on large organisations. Its findings cannot simply be extrapolated to smaller businesses.

Yet the implications for founders, scale-ups, fractional CFOs and smaller finance teams may be particularly significant.

Historically, sophisticated financial planning, structured management reporting and robust board communication have often required significant investment in people, systems and processes.

Today, increasingly capable AI tools are becoming more accessible.

However, access to AI is not the same as access to reliable financial intelligence.

A growing business may have several sources of financial data, limited internal resources and increasing expectations from investors or lenders.

What it needs is not another beautifully written report disconnected from the underlying numbers.

It needs a practical way of moving from financial information to trustworthy analysis and, ultimately, informed decisions.

The opportunity is to make disciplined financial management more accessible — not to remove the finance professional, but to give them better tools and greater analytical capacity.

Four questions every CFO should ask

As finance teams adopt AI, I believe the quality of their outputs should be assessed against four straightforward questions.

Traceability: Can every material financial figure be traced back to its source?

Reproducibility: Will the same financial inputs and calculation rules produce the same numerical results?

Transparency: Are the key assumptions visible, understandable and capable of being challenged?

Accountability: Has an appropriately qualified person reviewed the analysis and accepted responsibility for its use?

These are not new principles. They are fundamental disciplines of good financial management.

What is changing is the speed and scale at which technology allows us to apply them.

The wider financial industry is also paying closer attention to AI-related risks. In the Bank of England's September 2026 Systemic Risk Survey, 63% of participating financial firms identified AI-related risks among their five greatest potential risks to UK financial stability, an increase of 32 percentage points from the previous survey.

The Financial Stability Board has also consulted on 12 proposed sound practices for responsible AI adoption by financial institutions.

Neither exercise directly addresses the management reporting needs of growing businesses. But both illustrate the increasing importance placed on understanding and governing AI-related risks.

Making sophisticated finance more accessible

This broader evolution has also influenced my thinking behind ValuEdgeCFO, the AI-enabled CFO Co-Pilot being developed through ValuEdge.AI, a business I founded.

The approach is to combine structured financial calculation logic with AI-assisted analysis, commentary and reporting, while preserving human review and professional judgement. The aim is to make disciplined financial workflows and clearer communication more accessible to founders, finance teams and professional advisers.

It is one example of how I believe AI can support the evolution of financial management without attempting to replace the expertise on which sound financial decisions depend.

The real measure of progress

Over the coming years, I believe the most successful applications of AI in finance will be judged by more than hours saved or reports produced.

The more meaningful questions will be whether businesses can identify problems earlier, understand financial risks more clearly, evaluate alternatives more effectively and make better-informed decisions.

For CFOs, that means a gradual shift in emphasis: less time spent preparing information and more time devoted to interpretation, challenge, strategic planning and communication.

For founders and growing businesses, it creates the possibility of accessing more sophisticated financial capabilities without necessarily replicating the infrastructure of a large corporate finance function.

And for the wider profession, it presents an opportunity to redefine how financial expertise is delivered.

The future of AI in finance should not be about producing more answers. It should be about reaching better decisions, faster with the confidence to explain and defend them.

That, to me, is where the real value lies.