When the former chancellor Jeremy Hunt said last year he wouldn’t advise graduates to enter accountancy because of the threat posed by AI, it was always bound to hit a raw nerve across the profession.
That was last year; today the debate about how AI will impact accounting is very much alive, especially after KPMG claimed their AI agents performed a better audit than their human equivalents this week. But it’s not as simple as saying AI is going to replace jobs, not least because the uptake of AI is still relatively modest.
According to FloQast’s The State of Accounting AI 2026 report, while 85% of accounting teams have made AI a strategic priority, only one in 10 is using it extensively. Many finance teams, it seems, are still doing many tasks by hand and managing tasks with limited automation.
The study found that the barriers holding them back have less to do with technology and more to do with issues such as trust, training and governance, something I know the profession is looking to address as part of the recently launched CCAB AI Hub.
Whatever the reasons, it’s fair to say that we still have a way to go until AI has the kind of impact on our profession that some people are predicting.
And yet, the firms that have embraced AI to carry out repetitive tasks reported that they spend nearly half as much time on manual accounting and close their books two days faster.
If I had to sum up the state of accounting, I’d say it was still a ‘work in progress’. That said, based on what we do know, the direction of travel is clear, even if the pace of change is less certain. What’s more, it still continues to fuel debate about the future of our profession and how our roles may change in the coming years.
The nature of finance jobs will change
What’s clear is that finance teams will increasingly move away from repetitive tasks such as account reconciliations, transaction matching and variance analysis and, instead, move into roles that oversee AI agents carrying out these tasks.
Of course, the logical conclusion is that once AI becomes embedded across the profession – and it will – then we’ll need fewer people to do the basic work.
For some people, this is the very definition of progress. But it also harbours a problem, because as we know, junior accountants learn their trade by doing this relatively basic – yet fundamental – work.
They spend years dealing with transactions, reconciliations, accounts and processes. And it is this grounding that gives them the all-important foundation which, in turn, allows them to become the senior accountants, finance directors or CFOs of tomorrow.
If AI strips away that formative work, where will the next generation of accounting professionals come from?
A new skills mix
Thankfully, the accounting profession is right on the money, and one of the key areas it’s exploring is training.
Instead of spending their first few years mainly learning routine accounting processes, junior professionals may need much earlier exposure to the skills that AI cannot simply take over.
That means learning how to interpret AI-generated outputs, understand the data behind them and manage the finance systems producing them.
This shift may also mean junior accountants being encouraged to develop professional judgement sooner, which means knowing when an answer looks wrong, when assumptions need challenging and when human intervention is required.
Just as importantly, junior accountants will need to become better communicators so that they can explain financial information and work effectively with colleagues across the wider business.
It’s a point discussed at length in an article by the ICAEW. It quotes David Shogelola, talent director, accounting operations, at recruitment firm Robert Half, who said that junior roles are already adapting to the change.
‘As routine processing continues to be automated, early-career professionals are spending less time on repetitive data entry and more time on reviewing outputs, validating the quality of information produced by AI tools and spotting anomalies that require human insight,’ Shogelola said.
But this is just the start. As these AI-native juniors progress up the career ladder and begin to take on more senior roles, there’s every chance that the operating model itself will change.
Rethinking the finance function
Finance teams are likely to become less hierarchical and less dependent on large numbers of people carrying out routine transactional work. The boundaries between traditional accounting, technology and data roles are also likely to blur.
Rather than handing work from one specialist team to another, finance functions may bring those capabilities together in smaller, multidisciplinary teams.
In other words, AI won’t simply change how finance does its work. It could change what a finance department is built to do. But – and it’s a big ‘but’ – that is still some way off.
For all the talk of AI transforming finance, most organisations are still working through much more basic questions around data, systems, governance, skills and trust. The technology may be moving quickly, but changing how a finance function operates is going to take longer. But change it will.
To me, this doesn’t sound like a profession in decline. Far from it. Instead, I see a future in which accountants combine technical expertise with data literacy, AI fluency and professional judgement.
About the author
Hugh O’Neill FCA, principal, accountant in residence at FloQast
Useful links
FloQast report, The State of Accounting AI 2026 and CCAB AI Hub