2026 Top Struggles in AI Initiatives
On March 16, I ran a poll to figure out where the real, unvarnished friction lies in enterprise AI initiatives right now. I wanted to see if we are still lost in the fog of defining the destination, or if organizations have matured enough that messy execution, politics, and ownership have become the bottleneck.
To map this out, I forced the chaos of AI into a classic framework: WHAT, HOW, WHY, WHERE, and WHO. I desperately wanted to add WHEN (as in, "When is this actually going to drive ROI?"), but LinkedIn’s five-option limit killed that dream.
211 of you chimed in, and the breakdown is telling:
WHAT (Vision) | 30.9%
WHO (Ownership & Authority) | 19.9%
WHERE (Prioritization) | 17.5%
WHY (Justification & Buy-in) | 17.0%
HOW (Strategy) | 14.7%
I intentionally structured the poll options to mirror the natural progression of an AI journey from left to right:
WHAT ▶ HOW ▶ WHY ▶ WHERE ▶ WHO.
In a perfect world, you figure out what you want to change before mapping out how to build the strategy. Then you secure the why (the business justification), decide where to narrow your focus, and finally, assign who actually owns the risk and the budget. The reality is that most companies are feeling all five of these pains simultaneously, just at different turning points:
Vision hurts early when you are staring at a blank canvas.
Strategy hurts when those vague, lofty ideas desperately need structure.
Justification hurts when the CFO steps in and demands hard math before cutting a check.
Prioritization hurts when your use-case spreadsheet bloats to 200 unmanageable rows.
Ownership hurts the exact moment AI touches live workflows, real user data, and daily operating decisions.
The fact that WHAT(Vision) completely dominated the vote tells me everything I need to know.It means that despite the non-stop press releases, we are largely still in the absolute infancy of this cycle. I look around and see companies aggressively hiring, setting up AI steering committees, and taking endless vendor meetings. But there is a hilarious disconnect between cash and clarity: organizations are spending money as if they are deeply mature and have it all figured out, while their internal teams are secretly raising their hands saying, "We are still trying to define what we are actually doing here." 😂
I am convinced that true AI maturity has absolutely nothing to do with the size of your budget, the number of pilots you run, or how many software licenses you buy. It is measured entirely by how clearly your leadership can answer WHAT, HOW, WHY, WHERE, and WHO, in that exact order, and how honestly you are willing to rewrite those answers when the work gets real.
