The AI Deadlock

When Leadership delegates and IT can't deliver

Schedule consultation

4 min read · Oct 5, 2025

The AI Deadlock

Everyone wants AI. Every boardroom talks about it. Yet, for most mid-sized companies, AI adoption stalls before it even begins. Not because of cost, but because of a paradox in ownership and understanding. Leaders think in goals and outcomes, IT thinks in systems and dependencies. Somewhere between them, "AI strategy" becomes a loop of slides and slogans, driven by the illusion that real impact must be as spectacular as the hype itself.

The smartest place to begin with AI? Right where it hurts.

The AI deadlock isn't really about the technology; it's about focus. Leaders want transformation but often lack both a clear grasp of what AI can realistically deliver, and an honest view of their own process reality. When clarity is missing, leadership's directives to IT tend to be all about vision but vague on what actually needs to be built.

In a client workshop we held recently, the turning point came when a CFO realized that a report she would ask for regularly was taking a team member hours per week to put together because of data fragmentation and poor system integration. Through a straightforward AI-powered automation, the work disappeared and the CFO could pull up the report at her discretion.

The seemingly trivial inefficiencies that rarely surface at the leadership table aren't obstacles to adoption; they're the invitation. The manual handoffs, outdated tools, double data entry, too many systems, and quiet workarounds that slowly drain productivity (1).

When you apply AI where the friction lives, you don't just automate, you reveal how the business truly works. That's where momentum starts and strategy becomes real. One process, and one workflow at a time.

The irony of the AI deadlock: The transformation everyone is chasing often hides in the very inefficiencies no one stops to see.

The New Automation Frontier

Before the rise of Generative AI, and it's worth stressing that this is the scope of this article, workflow automation was a creature of certainty. It followed strict rules, clicked the same buttons, read the same forms, but froze the moment something looked different, breaking the flow. The more fluid workflows understandably remained beyond reach of automation.

Gen AI has quietly redrawn those boundaries. It can read what isn't uniform. It can understand messy language and handwritten documents, adapt to variation, and make sense where traditional scripts would stumble. The line between what we automate and what we leave to people has moved (2). Further than most realize, turning negative business cases for automation into positive ones.

Cracking the Deadlock

Breaking out of the AI deadlock starts by making things smaller, with a focused moment where leadership and operations sit together to answer three simple questions:

Where's the friction?
What value is being lost?
And what can we safely test within 90 days?

In a recent workshop session with one of our clients — a Swiss payroll business — a team leader captured these 3 questions beautifully in one statement: "I'm tired of my team manually checking hundreds of scanned timesheets. Our clients all use different HR systems, so we can't standardize. If this process slips, clients don't pay us, but it's costing us about 30 hours every month. Can we quickly test if AI can handle this?".

And that's how progress starts: Not by drafting ambitious moonshot plans, not by leaders getting lost in the weeds either, but by creating space for teams to surface the real obstacles that slow the business down.

The Lived Experience

Real change starts small. Solving a tangible problem with AI turns vague ambition into lived experience. Teams feel heard when their daily pain points are addressed. IT gains hands-on AI expertise on work that matters, and leadership sees strategy take shape on the ground.

These early wins do more than remove friction, they turn AI from a hot potato into shared ground, a fresh starting point. AI stops floating above the business and starts working within it. One meaningful step at a time.

For mid-sized businesses, AI strategy doesn't start with grand visions; it starts with precision. Identifying the right friction points, aligning leadership with operations, and proving value fast. That's where real transformation takes root. This is the work I help organizations do every day.