AI Killed FP&A Busy Work. What’s Next for Early Career Professionals?

Finance teams are at a weird crossroads right now.

On one hand, artificial intelligence (AI) is finally taking soul‑crushing work off our plates. The endless reconciliations, the copy-paste gymnastics, and the "why is this variance here?" scavenger hunts."

On the other hand, a quiet question keeps popping up in the background: If the grind goes away, how do people actually "learn" finance?

This tension isn’t hypothetical. Over the last few years, expectations on finance teams have exploded. Finance leaders are being asked to move faster, think more strategically, and guide the business through constant uncertainty.

According to OneStream’s Finance 2035: Return to Investment research, investors now rank chief financial officer (CFO) competence as one of the most important factors in investment decisions. Investors also expect finance leaders to act as true strategic growth drivers, not just number crunchers. That kind of confidence doesn’t come from dashboards alone. It comes from deep financial understanding.

And that’s where things get interesting. Traditionally, that understanding was earned by living in the details, touching the data, and repeating the same processes until patterns started to click. But today’s analysts are coming into a world where AI can do a lot of that work instantly. So what happens to the learning curve when that busy work disappears? Is something valuable lost?

What I’ve found is that AI isn’t eliminating the need for finance fundamentals. But it is forcing leaders to rethink how those fundamentals are built.

The Expectation Gap

There’s a growing generational gap between those entering the financial planning and analysis (FP&A) workforce and those who have been here for a while.

Newer entrants to the profession are showing up expecting AI to be part of the job, and not as a “nice to have,” but as table stakes. In fact, a OneStream study found that 86% of finance students and professionals expect to use AI tools at least “somewhat often” in their careers. A full one-third expect to rely on AI tools “significantly.” For younger talent, AI dominates expectations.

But that expectation doesn’t always line up with reality inside organizations. While two‑thirds (66%) of current corporate finance professionals say they use AI at work today, confidence in applying AI drops off as careers progress. Nearly 90% of finance students say they have enough AI experience to use it at work. Comparatively, just 54% of professionals with 10+ years of experience say the same. There’s a canyon between how finance could work and how it often still does.

The result of these mismatched expectations is frustration. Frustration with the pace of change, frustration with AI’s unknowns, and frustration with inexperienced colleagues making strategic decisions without enough of a financial background.

The problem is really trust — trust that the next generation can build deep expertise in a very different way than the last one did.

What the Grind Used to Teach Us

Let me be clear: No one actually misses the grind.

Nobody is nostalgic for doing late‑night reconciliations, using broken spreadsheets, or manually hunting down variances that turned out to be nothing.

The grind wasn’t fun. It wasn’t efficient.

The problem is that it worked. Not because the work itself was valuable, but because repetition was. Running the same processes repeatedly forced analysts to sit close to the data. You started to recognize patterns. You could sense when something was off before you could explain it. That instinct didn’t come from dashboards or summaries. It came from exposure.

The grind built judgment.

The grind taught people how data flows through a business, where it tends to break, and which assumptions quietly matter the most. The grind trained analysts to question outputs instead of blindly trusting them. And the grind gave analysts the confidence to say “This doesn’t look right,” even when the numbers were technically tied.

That’s what leaders are worried about losing — not the work itself, but the intuition that comes with practice.

When removing this friction, AI also removes forced learning. You can get answers instantly without ever understanding how they were produced. In finance, speed without context is dangerous. Clean outputs are easy. Defensible insights are not.

The grind was never sacred. But the understanding it created remains sacred.

Which Parts of the Grind Finance AI Should Kill

If a task exists purely because systems don’t talk to each other, that’s not “learning.” That’s a tax.

Manual data pulls. Reconciliations that don’t change the answer. First‑pass variance explanations that everyone knows will be rewritten anyway. Finance AI should wipe those out without hesitation.

AI is good at the mechanics of finance: aggregating data, flagging anomalies, surfacing trends faster than any human ever could. Letting machines handle that work doesn’t cheapen the role; it clears the runway for strategic work. Analysts spend less time proving numbers are accurate and more time asking what the numbers actually mean.

The mistake is assuming this work was valuable simply because it took time.

It wasn’t. The value was never in manually stitching together reports. The value was in what people noticed while doing it. And that’s an important distinction. AI can handle the repetition, but it can’t replicate curiosity. It can surface a variance, but it can’t decide whether that variance matters in the context of a pricing shift, a supply issue, or a half‑baked sales forecast.

So yes, automate aggressively. Just don’t confuse automation with abdication.

The goal isn’t to protect busy work. The goal is to free finance teams from busy work while deliberately preserving the moments that build judgment. AI should handle the “what changed.” Humans should own the “why” and the “so what.”

Yes, Junior Analysts Still Need to Touch the Data

Learning doesn’t happen by osmosis. Junior analysts need to understand how numbers come to life and where they originate. Juniors need to learn how assumptions sneak in and why a perfectly logical model can still lead to the wrong conclusion. That kind of understanding only comes from getting your hands dirty.

I’m not advocating that juniors should be buried in manual work all day. But I do believe they need intentional exposure. They need to see how inputs change outputs. They need to trace a number back to its source at least a few times before trusting automation to do it for them. Otherwise, they’re learning how to click buttons, not how finance actually works.

To be clear: The goal is to sequence AI, not withhold it.

Smart teams design learning moments on purpose. They ask analysts to explain the story behind the numbers. They make space for “walk me through how this was built,” even if the tool did most of the heavy lifting.

I don’t see the risk as entry-level professionals using AI — that's inevitable. The real risk is that they’ll never build the confidence to challenge it.

What “Deep Understanding” Actually Looks Like Now

A lot of confusion stems from mismatched expectations of what “deep understanding” means in FP&A for juniors. Before AI, deep understanding was about who can build the most complex model or manually trace every formula. Those skills still matter, but they’re not the signal they once were.

Today, deep understanding reflects how one reacts to numbers, not how one generates them.

When someone really understands the business, it's obvious because they don’t just accept the output. They want to know why something moved, what assumptions are doing the heavy lifting, and whether the story makes sense given what’s happening on the ground.

Deep understanding sounds like these statements:

• “This result technically checks out, but it doesn’t align with what sales is seeing.”
• “That variance isn’t new. We’ve been masking it with timing differences.”
• “If we follow this logic through two quarters, here’s where it breaks.”

No mechanical skill can replace judgment.

Deep understanding hasn’t disappeared; it’s just moved upstream. It lives in interpretation, skepticism, and decision‑making. AI will never replace that skill. AI just exposes whether the skill is actually there.

The Real Risk Isn’t AI. It’s Getting AI Adoption Wrong.

Most finance teams are no longer debating whether to use finance AI. The questions are how they will use it and what will break if it’s implemented incorrectly.

When AI is ignored, the consequences are obvious. Teams stay buried in low‑value work. High‑potential talent gets frustrated. Competitors move faster, forecast better, and look smarter doing it.

But I’m here to warn you that swinging too far in the other direction is just as dangerous.

When everything is automated by default, teams can lose the ability to challenge what they’re seeing. Outputs look clean. Dashboards look confident. And suddenly no one remembers how the numbers were built or whether the assumptions still make sense. That’s how bad decisions sneak through with great formatting.

The best finance leaders avoid both extremes.

Good finance leaders don’t treat AI as a magic fix or a threat to be contained. They treat it like infrastructure. Something powerful, necessary, and dangerous if misunderstood. They invest in automation and in capability. They modernize workflows and redesign how people learn inside those workflows.

Finance teams don’t win by producing answers faster. They win by producing answers the business can trust. AI can absolutely help with that. But only if humans stay firmly in the loop, doing the part of the job that’s always mattered most: thinking.

Build Judgment. Then Scale It With AI.

AI isn’t the end of finance fundamentals. It’s the moment we must get serious about how they’re built.

The teams that will win aren’t the ones clinging to manual processes or blindly automating everything in sight. They’re the ones who understand that judgment still matters and that AI simply changes where and how that judgment is developed. Automation should remove friction, not thinking. Speed should amplify confidence, not replace it.

If you’re figuring out how to do that — how to modernize finance without losing the instincts that make it effective — the Finance AI Academy is a great place to start. It’s designed to help finance leaders and teams understand where AI creates real value, how to apply it thoughtfully, and how to build trust in AI‑driven insights without sacrificing rigor.