Anshuman Yadav is a strategic finance and AI leader with global SaaS, M&A, and ops experience. Kellogg MBA. Ex-civil engineer.
Every major leap in finance has come from a new tool: bookkeeping that made commerce accountable, spreadsheets that gave analysts the ability to test assumptions quickly instead of waiting days, or ERP systems that centralized information and became the company’s control tower. Finance always moves with its tools.
Now, AI is being touted as the next leap. From my decade working across three countries and multiple industries, I have seen the same problems repeat themselves—fragile automation, siloed systems, forecasts that don’t learn. Until recently, the tools weren’t there to fix them. Today, AI is mature enough that we finally have a chance to tackle these gaps.
But so far, what I see in practice still feels small: automating invoices, slightly better forecasts built on last year’s data, or ERP AI features that look impressive in a sales deck, but never change the boardroom conversations. That’s tinkering, not transformation.
Where AI Falls Short Right Now
Working with companies big and small, and from my time as a fractional CFO to startups, I have seen the same issues pop up time and again.
For one, automation seems fragile. It works fine until something changes—a new data source, a new format—and suddenly the process breaks. Additionally, forecast models lack memory and context, spitting out projections, but not remembering how far off they were last quarter, or that sales leaders tend to overpromise. I have sat in planning sessions where the forecast looked perfect on paper, but everyone in the room knew it would be off by 20%. Finally, ERP systems still trap information in silos. Treasury numbers rarely make their way into FP&A, risk analysis doesn’t influence capital allocation, and investor relations often ends up stitching together a story on their own.
This all leads to CFOs walking into board meetings armed with plenty of numbers, but not enough foresight. The questions that really matter: Do we double down on R&D or pay down debt? What if FX volatility drags on for another year? These are the questions that don’t always get answered.
What Needs To Change
Finance doesn’t need another automation tool. What it does need is AI that can remember, learn and connect the dots.
This is where the idea of the intelligent financial enterprise (IFE) comes in. The name doesn’t really matter. The point is simple: Finance should work less like a set of disconnected reports and more like a continuously learning co-pilot.
What would that look like?
• A treasury model that bakes delays in cash receivables into treasury forecasts in real time.
• FP&A projections that update as soon as the sales conversion rates dip, so you don’t have to wait until the next monthly or quarterly close.
• Risk tools should map how suppliers connect with each other and flag any weak links before they turn into costly problems.
• Investor updates should tie results back to strategy and explain what’s really driving the numbers.
This is not automation. This is the finance behaving like a living model of the business.
What Leaders Gain From Intelligent Financial Enterprise
Three big benefits would show up when teams adopt this way of working. First, capital finally goes where it matters. Instead of budget decisions being won by the loudest voice, they are backed by clearer trade-offs. Second, risks don’t sneak up as often as liquidity gaps, and supplier issues get spotted before they spiral. Treasury teams catch problems months earlier this way. And third, boards get more confident. Not because the numbers are perfect, but because there’s context and explanation behind them.
The point here is to give finance leaders sharper tools so their judgment calls are better informed.
The Hard Part Nobody Should Downplay
Of course, getting to this point isn’t simple. A few challenges come up every time:
• Governance: Some things can run on autopilot—a small cash transfer, for example. But a $50M capital decision always belongs with leadership.
• Data: If ERP and CRM systems are inconsistent, AI will only make the noise louder. In practice, teams often burn weeks cleaning up data, only to run one useful model at the end.
• Culture: This is probably the hardest piece. Finance is used to living in spreadsheets—it’s what we were trained to do. But the real shift is learning how to challenge and apply insights coming from AI. In my experience, the teams that adapt fastest aren’t the ones that resist it; they’re the ones that experiment with it and fold it into their process.
These aren’t just technical hurdles, but leadership ones.
The Takeaway
Every tool has changed finance. Bookkeeping made it accountable. Spreadsheets made it analytical. ERP made it central. AI can make it adaptive, but only if we stop treating it as a gimmick and start building systems that learn. I don’t believe AI will replace CFOs. But I do believe CFOs who ignore AI may be replaced by peers who embrace it.
Dashboards won’t define the next chapter of finance. Systems that learn and give leaders foresight will.
Forbes Finance Council is an invitation-only organization for executives in successful accounting, financial planning and wealth management firms. Do I qualify?