Workers are ready, organisations aren't
Your most capable AI users are running Claude on their phones at lunch.
Your most capable AI users are running Claude on their phones at lunch. The work laptop is locked. They know what’s possible. Their organisation won’t let them near it.
Microsoft just put a number on how many of these people sit inside your business.
It’s 10%.
Microsoft’s 2026 Work Trend Index dropped recently. 20,000 workers, 10 countries, telemetry from Copilot across every segment worldwide. The most rigorous look at what’s actually happening inside organisations on AI right now.
It tells you three things every leadership team needs to sit with.
1. Your people are already using AI for the hardest work in their day.
The biggest single activity in the Copilot telemetry wasn’t drafting emails or summarising notes. It was making decisions and solving problems. 27.5% of all use. More than every other activity in its category combined. Half of all use sits in the broader analysing, reasoning and deciding bucket. The grunt work AI was sold to do? Just 17%.
For two years the pitch has been: free up time on the admin so your people can focus on higher-value work. Look at the data. Your people are already past that. They’re using AI for the higher-value work right now. Without your strategy, without your governance, without your training.
The question stops being whether to free up time. It becomes whether your operating model has caught up to what they’re already doing.
2. 10% of your workforce has the AI capability of your top performers. Your organisation is the reason they’re not.
Microsoft sorted the workforce on two axes. How capable people are with AI as individuals. And how ready the organisation around them is to support that capability. Five zones came out of it.
19% sit in the Frontier. High individual capability, high org readiness. The people doing AI well in environments built for it.
10% sit in Blocked Agency. Same individual capability. Same skill. Same drive. The org around them is the bottleneck.
That’s the part that should stop every leadership team for a minute. The 10% in Blocked Agency have the same AI capability as your top performers. The only variable is the environment they sit inside.
If your AI strategy is built around training and tool rollouts, you’re solving the wrong problem. The capability is already there. The real work is upstream.
3. Org culture is the #1 predictor of AI outcomes. By a long way.
Microsoft ran a predictor analysis on 28 variables against actual AI outcomes inside organisations. Demographics. Tenure. Industry. Company size. Job function. Generation. Manager support. Talent practices. Individual mindset. The whole stack.
The number one predictor came back at 100% relative importance. Org AI culture. How open and curious people are. How safe they feel suggesting new ways of working. How confident they are using it.
Number two, talent practices. 43%. Less than half the impact of culture.
Number three, manager support. 43%. Similar.
Number four was the first individual variable. AI mindset of the person themselves. 42%. Less than half the impact of the culture they sit inside.
Then the demographics. Generation, industry, tenure, company size, job function, decision-maker status. All under 2%. Most under 1%. The “Gen Z is leading on AI” headlines? Less than 0.5% of the variance in AI outcomes.
You don’t have a capability problem. You don’t have a generation problem. You don’t have an industry problem. You have a culture problem. And culture is downstream of leadership. Always has been.
Karim Lakhani at Harvard, who wrote the foreword to this report, put it cleanly.
Productivity gains at the edge do not automatically become enterprise transformation at the core.
The culture is what carries the productivity from one to the other. Without it, you get 10% of your people running ahead on their own time, locked out of the laptop you give them, and the other 90% waiting for permission.
The work for 2026 isn’t bolting AI onto your existing operating model. It’s rearchitecting the operating model so the capability your people already have can actually leave the lunch break and enter the business.
Two questions for your exec team this week.
How many of your people are sitting in Blocked Agency right now?
And have you ever measured your AI culture honestly?
If you did, what would it tell you?
One more thing...
56% of CEOs report zero financial return from AI. Not because the technology failed, but because they skipped the most important step: understanding where their people actually stand.
If you want to know where your organisation sits on what this article just covered, we have a globally proven approach. Your employees take a 15-minute assessment and we deliver a rich visual report that shows you how many of your people sit in Blocked Agency, where shadow AI is actually happening inside your business, what your culture score looks like, and what your 30/60/90-day plan should be. It's benchmarked against 296,000+ professionals across 70 countries and over a billion data points, showing where your organisation's AI proficiency stands globally.
Everyone who takes part gets a personalised 90-day AI roadmap. Real value for 15 minutes, plus a signal that leadership is thinking about their development. Leadership gets the diagnosis a top-down strategy was never going to surface: organisational AI maturity, shadow AI intelligence, the culture variables that predict outcomes.
Book a 30-minute call to see what this would look like in your organisation, or learn more about AI Compass, which powers this solution (we are proudly the only Australian partner that can offer this).