CFOs are throwing money at AI like drunk sailors at a casino, and Gartner just called them out for it. The harsh truth? Most finance leaders are confusing AI deployment with actual value creation — a mistake that’s costing companies millions while delivering nothing but expensive digital theater.
This isn’t just another case of corporate waste. This is a systematic failure of financial leadership that echoes some of the most spectacular technology boondoggles in business history. And it’s happening right now, in boardrooms across the globe.
The AI Delusion: Deployment ≠ Value
Here’s what’s actually happening: CFOs are purchasing AI tools, implementing AI systems, and checking AI boxes on their strategic initiatives. Then they sit back and wait for the magic to happen. Spoiler alert: it doesn’t.
Gartner’s warning cuts straight to the bone — finance leaders are making the same fundamental error that killed countless technology investments over the past three decades. They’re measuring activity instead of outcomes, spending instead of returns, and adoption instead of transformation.
This pattern isn’t new. Remember when every company had to have a website in the late 1990s? Most of them built static, useless digital brochures that served no real business purpose. Or the ERP implementation disasters of the early 2000s, where companies spent millions on systems that made their operations worse, not better.
“How I Cut Claude Spend by ~40% Every CFO is looking into the ROI of AI spend right now Not because AI isn’t useful (obviously is), but where can the company be more efficient while not decreasing AI usage? And is the incremental $ spend worth it?” — @OnlyCFO
The difference this time? The stakes are higher, the investments are larger, and the opportunity cost of getting it wrong is massive.
The Real Problem: Finance Leaders Don’t Understand Their Own Data
CFOs are deploying AI solutions to problems they haven’t properly diagnosed. They’re automating processes that shouldn’t exist in the first place. They’re applying machine learning to data that’s fundamentally flawed or incomplete.
This is like trying to fix a broken engine by painting the car.
The fundamental issue isn’t technological — it’s strategic incompetence. Most finance organizations don’t have:
- Clear visibility into their actual value-creation processes
- Measurable baselines for the problems they’re trying to solve
- Defined success metrics beyond “we implemented AI”
- Change management capabilities to capture the value AI could create
Compare this to how companies approached automation in manufacturing. Toyota didn’t just buy robots and hope for the best. They spent decades perfecting their processes first, then applied automation strategically to amplify human capabilities and eliminate specific inefficiencies.
“We’re in an era where complex financial tasks don’t have to be managed manually, with automation layers such as @w3arew3 emphasising this at the enterprise level by enabling firms to deploy complete & verifiable workflows at compute speed & AI scale Agentic finance is the future” — @Erin_Trivett
The Historical Parallel: Why This Feels Like the Client-Server Disaster
This AI deployment chaos mirrors the client-server computing transition of the early 1990s. Back then, companies ripped out perfectly functional mainframe systems and replaced them with distributed networks that were theoretically superior but practically disasters.
The pattern was identical:
- Technology vendors promised revolutionary improvements
- IT departments focused on technical deployment rather than business outcomes
- Finance leaders approved massive budgets based on theoretical ROI
- Operations suffered through years of decreased productivity and increased complexity
Companies that succeeded during the client-server transition did three things differently: 1. They started small with pilot projects that proved value before scaling 2. They focused on specific business problems rather than broad technological transformation 3. They measured actual performance improvements, not just system utilization
The companies implementing AI successfully today are following the exact same playbook.
What CFOs Should Actually Be Doing
Stop buying AI tools and start auditing your finance processes. Here’s the roadmap that actually works:
- Identify your three most expensive manual processes — not your most complex ones, your most expensive ones
- Measure the actual cost of errors, delays, and rework in your current systems
- Test AI solutions on small, contained problems where success and failure are immediately obvious
- Scale only after proving measurable value — and only to similar, well-understood problems
- Build internal capabilities for AI implementation rather than outsourcing strategy to vendors
The companies getting this right aren’t deploying enterprise-wide AI platforms. They’re starting with accounts payable automation that cuts processing time from hours to minutes. They’re implementing fraud detection that reduces losses by specific, measurable amounts. They’re using predictive analytics to optimize cash flow in ways that directly impact working capital.
These aren’t glamorous AI deployments. They’re boring, practical applications that make money.
The Brutal Truth About AI in Finance
Most finance AI implementations fail because CFOs are solving the wrong problem. They think the problem is efficiency, when the real problem is effectiveness. They think they need automation, when what they actually need is insight.
The brutal reality? Many finance organizations could achieve better results by simply fixing their data quality, standardizing their processes, and training their teams properly. No AI required.
AI amplifies your existing capabilities — both good and bad. If your finance processes are messy, inconsistent, or poorly understood, AI will make them messier, more inconsistent, and even less understood. But at 10x the cost.
The companies that will dominate finance AI aren’t the ones spending the most money on tools. They’re the ones who understand their operations well enough to apply AI strategically, measure its impact precisely, and scale its deployment systematically.
Everyone else is just burning money on digital theater while their competitors build actual competitive advantages.
Published in Stream · Dispatch #399 · May 28, 2026 · 5 min read.
Reply to paolo@mont3.ch - every email gets a human answer within 24h.