The shift is already happening
In the next twenty-four months, finance is going to see a shift unlike anything the profession has been through before. Roles will move from transactional to commercial. The work will look substantially different from the work that trained the current generation of finance leaders.
That is exciting. It is also a real responsibility. As senior finance professionals, we have to protect the staged transition from junior to senior, and we have to pass on the knowledge that makes finance a profession rather than a set of tasks. How we handle this change is going to determine the quality of the next generation of finance leaders.
Stop selling AI on efficiency. Start measuring it on impact.
SaaS software was sold on efficiency and time saved, and positioned as resource and cost savings. A lot of AI-native products are heading down the same route. That framing misses the biggest opportunity.
A significant uplift in output from the same team is far more valuable than a slice of that team's time saved. I have seen too many marketing org charts where the bottom layers have been cut and the senior roles halved. Let's not do that for finance.
The goal of AI should not be doing the same work with fewer people. It should be doing exponentially more with the same people. That reframing changes everything about how you plan the transition.
What that reframing means for junior roles
If we automate the entry-level work, the honest question becomes: how do we train the next generation of finance professionals?
The answer is not to remove the junior role. It is to uplevel it.
Instead of cutting a junior because AI can now do the data entry, use that junior to manage the accuracy of the data, run variance analysis, and spot trends. This is the proactive value-add work that finance teams have always wanted to do but rarely had time for. It is also the work that develops commercial judgment, which is exactly what the profession needs more of.
The value split needs to invert:
- Previously: 80% data gathering, 20% insight
- Future: 10% data oversight, 90% strategic partnering
The old finance model vs the new one
The rows on the right are the roles finance has been arguing it should be doing for the last decade. AI is finally the tool that makes the shift possible. The question is whether the profession is willing to take it.
Reimagining the career path
We still need structure for career progression and purposeful learning. If anything, the training element of the profession is now more important than at any point in my career, because on-the-job learning alone is no longer enough. Senior professionals, and the businesses they work for, have to actively make space and time for structured learning. This is not something we can leave to the professional institutes on their own.
With AI shrinking the early transactional phase, the traditional three-year junior period might reasonably compress to twelve months. That is faster progression for the individual and faster impact for the business. It is only a good thing if the accelerated route is genuinely well-structured. If we don't get that right, we will have a serious skills gap in five years and no easy way to close it.
To support entry-level roles through this transition, four things need to be in place.
What senior finance leaders should be doing right now
1. Adopt an apprenticeship mindset
Businesses should treat all entry-level finance roles the way apprenticeships are treated, with training as important as output, and structured into the job rather than tacked onto it. Finance is already set up culturally to do this. Most other professions aren't. Use the advantage.
2. Keep teaching foundational theory
The fact that we have calculators didn't mean we stopped teaching children how to count. The fact that we had accounting systems didn't mean we stopped teaching the theory of T-accounts. AI does not change the case for foundational theory. If anything, it strengthens it, because without foundational theory a junior cannot spot when the AI is wrong.
3. Insist on visibility in the tools you buy
The AI tools finance teams use must provide visibility into what is happening underneath, so juniors can still learn the mechanics. This is a purchasing decision as much as a training decision.
4. Build purposeful knowledge sharing into the working week
Dedicated internal training sessions or external mentoring should become a core part of career progression rather than a nice-to-have. I have run dedicated training sessions for junior team members throughout my time as a fractional CFO. Separate from one-to-ones, they act as safe spaces to ask questions, deepen understanding for exams, and bridge the gap between theory and practice. They work. They should be standard.
Let junior team members teach the seniors too
We should also actively involve juniors in figuring this transition out. We shouldn't assume we know where the gaps are. Juniors typically understand the interface and the prompting of new AI tools faster than seniors do. That is an asset. Create a culture where they teach the seniors the tech, and the seniors teach them the context. Both directions matter.
The "white box" principle: why transparent AI matters more than clever AI
To avoid the skills gap, the profession has to reject "black box" AI tools where data goes in, an answer comes out, and nobody knows why. For finance to survive the AI shift with its expertise intact, the tools we adopt have to be transparent and to integrate learning into the way they work.
That means three specific things.
Auditability of AI logic. Junior staff should be tasked with auditing the AI's reasoning. If an AI flags an issue or takes an action, the junior must be able to explain the reasoning behind it, why the issue was flagged, and how to prioritise the response.
System visibility. AI tools should let users drill down into the raw data and produce an audit log of actions taken and why. This maintains the link between modern automation and foundational theory. Juniors still learn the "why" while the AI handles the manual element.
Trust. In finance, "the AI said so" is never going to be an acceptable answer to an auditor or a CFO. A white-box system provides a defensible audit trail for every step. Nothing else will.
Without insisting on transparency, we risk producing a generation of finance "button-pushers" who can operate the software but don't understand accounting. If the software glitches or the tax code changes, they won't have the foundational theory to spot the error or make the adjustment. Core accounting judgment will slowly be lost.
Why Finzu is being built as a white-box system
We have deliberately committed Finzu to the white-box approach. Every action is auditable. Every AI decision is traceable back to the data it was made from. The system is built so that a junior using it is learning finance, not just operating a tool. Learning is designed into the way the platform is used, not layered on afterwards as a training module.
This is a commercial choice as much as a philosophical one. Businesses that adopt white-box finance tools will produce stronger finance professionals over the next decade. Businesses that adopt black-box tools will find themselves in five years unable to explain their own numbers to auditors, investors, or acquirers. The market will sort out which was the right call.
The bottom line
This is the most exciting moment finance has had in a generation. The profession can move away from processing and towards genuine commercial partnership. Juniors can reach interesting, high-impact work in twelve months rather than three years. Senior leaders can shift from reviewing transactions to mentoring the next generation and shaping business decisions.
Or, we can do the opposite. We can cut entry-level roles because AI can do the data entry. We can adopt black-box tools because they demo well. We can measure AI on hours saved and end up in five years with smaller teams, weaker training pipelines, and a shortage of the senior talent every business will still need.
The choice is being made now, by the finance leaders reading this. Let's build a profession that does more, knows more, adds more value, and grows faster.
Frequently asked questions
Will AI replace junior finance jobs?
Some of the transactional work junior finance staff have historically done, yes. The role itself, no, not if it is redesigned properly. The teams getting this right are using AI to remove the data-entry element and giving juniors work that used to sit further up the ladder: variance analysis, exception review, first-pass commercial commentary. The junior finance role is changing, not disappearing.
How should finance leaders train the next generation in an AI-driven environment?
Treat entry-level roles as apprenticeships with structured training built in. Keep teaching foundational accounting theory. Insist on transparent, "white-box" AI tools that let juniors see the mechanics underneath. Build regular knowledge-sharing sessions into the working week. And let juniors teach seniors about the tools while seniors teach juniors about the context. It is a two-way exchange, not a one-way transfer.
What is "white box" AI in finance and why does it matter?
A white-box AI tool is one where the reasoning behind every action is transparent, auditable, and traceable. The opposite is a black-box tool that produces an output without letting users see how it was arrived at. For finance, only white-box tools are acceptable, because "the AI said so" is not a defensible answer to an auditor, a CFO, or a regulator. It is also the only kind of tool that lets junior staff genuinely learn accounting rather than just operate software.
Will the finance career path get shorter as AI takes on more work?
Probably. The traditional three-year junior period is likely to compress to around twelve months as AI absorbs the transactional work that used to fill those years. That is a good thing for the individual, provided the compressed period is genuinely well-structured. If it becomes a badly-supported acceleration, it will produce senior professionals with real gaps in their foundational understanding, and the profession will pay for that later.
What is the biggest risk of AI in finance careers?
Producing a generation of finance professionals who can operate AI tools but don't understand accounting. That happens when businesses adopt black-box AI, cut entry-level roles rather than uplevel them, and stop investing in structured training. The technical failure mode is easy to describe: the tools glitch, the tax code changes, an audit surfaces an unusual transaction, and there is nobody in the room with the foundational knowledge to spot what's wrong. Avoiding this is a training and purchasing decision, not a technology decision.
