This article is part of Global Outlook from IREX, a monthly insights series for our community, featuring perspectives from Aleksander “Aleks” Dardeli, President and CEO.

I keep returning to a line from our IREX AI Readiness report: universities know AI matters, but few have built the strategy or governance to act on that belief. That finding pushed me to look wider, past universities, at the organizations most likely to win the AI era.

The gates are open, and they will not close. Stanford's 2026 AI Index puts organizational adoption at 88 percent, with generative AI reaching 53 percent of the population in three years. Adoption is no longer a differentiator. Everyone has walked through the gate, and capability keeps accelerating behind them, across benchmarks, investment, and education.

What separates institutions now is what happens after. AI remains a tool, and tools solve problems only as well as we understand them. A tool cannot own change for us. BCG surveyed 640 CEOs and found 72 percent now call themselves the main decision maker on AI in their organization. Half of those CEOs believe their own job depends on getting AI right. A smaller group, BCG's “Trailblazer” CEOs, roughly 15 percent, push further still, investing faster and upskilling teams sooner. Morgan Stanley built an evaluation framework around its AI assistant and reached 98 percent adoption among advisor teams. Moderna paired training with internal AI champions and drew 2,000 weekly participants to its AI forum. Each case points to the same pattern: leaders redesigned work, not just deployed software.

Most institutions have not made that shift. Deloitte finds only 34 percent are deeply transforming with AI, while 37 percent use it at the surface, with little change to process. Worker access to AI rose 50 percent last year, yet skill remains the binding constraint. The gap sits between ambition and ability. McKinsey reports nearly 60 percent of organizations cite knowledge and training gaps as the leading barrier to responsible AI, and nearly two-thirds name security and risk as the top obstacle to scaling agentic AI. Our own research at IREX found similar patterns in higher education: one in three institutions had a clear AI strategy, fewer than one in five had governance structures in place, despite leadership ambition running well ahead of both. Interest is not the constraint. Readiness is.

This brings me to people in the room, and the other differentiators around them. Leadership matters, and BCG's CEO ownership numbers back that up. Leadership alone will not carry an institution though. McKinsey finds organizations with named ownership for responsible AI score 2.6 on maturity, against 1.8 for those without it. Skills, governance, workflow redesign, and data readiness compound leadership, or they limit it. Weakness in any one area drags down the rest.

Technology access is becoming common, and the price of entry keeps falling. What remains scarce: leaders who can organize people, process, and accountability around AI, and who move past pilots into sustained value. That is the divide I am watching now, and I expect it to define which institutions lead the next decade, and which fall behind.

More from IREX

Have you read IREX higher education AI readiness report? Access From Ambition to Adoption: Insights into University AI Readiness from Around the World.