For the past two years, business leaders have talked about artificial intelligence as if the main question were access: who has the best model, the best copilots, the best stack, the best budget. But that framing is already starting to feel dated. In much of corporate life, AI is no longer the scarcest thing in the room. Judgment is.

That is the awkward twist now confronting boardrooms from Toronto to Singapore. The software is improving faster than the institutions using it. AI can summarize the meeting, draft the memo, generate the forecast, translate the deck and propose five strategies before lunch. What it still cannot do is decide what matters most, which risks are acceptable, or who will take responsibility when a confident answer turns out to be wrong.

That gap is showing up in the data. McKinsey says more than three-quarters of respondents now report that their organizations use AI in at least one business function, while companies are beginning to redesign workflows and elevate senior leadership oversight to capture value from generative AI. IBM, meanwhile, found that 42% of enterprise-scale companies had already actively deployed AI, with another 40% still exploring or experimenting, even as limited skills and expertise remained one of the top barriers to deployment.  

In other words, the tools are arriving faster than the operating model.

From knowledge advantage to judgment advantage

For decades, office power often belonged to the people who could synthesize information fastest. The executive who could turn chaos into a clean slide deck. The manager who always had the spreadsheet. The analyst who could digest a 70-page report overnight and emerge with the three takeaways that mattered.

AI has begun to commoditize much of that performance. The ability to produce polished first drafts, instant summaries and passable analysis is becoming widely available, and quickly. What used to signal rare capability increasingly looks like table stakes.

That does not make humans irrelevant. It makes the human contribution more exposed.

The World Economic Forum’s Future of Jobs Report 2025 found that employers increasingly value analytical thinking, resilience, flexibility, agility and leadership-related capabilities as technology reshapes work. The report draws on more than 1,000 leading global employers representing over 14 million workers across 55 economies. That scale matters, because it suggests this is not a niche consulting obsession. It is a broad management shift.  

If AI lowers the cost of producing information, then the premium moves to interpretation, prioritization and action. The winner is not the company that generates the most material. It is the one that can decide faster what to trust, what to ignore and what to do next.

Why so many companies still feel stuck

This helps explain one of the strangest features of the AI boom: adoption is widespread, but satisfaction is uneven.

BCG says only 5% of companies worldwide are truly “future-built” around AI, while 35% are scaling AI and beginning to generate value. That top 5% is not just marginally ahead. According to BCG, those companies are seeing five times the revenue increases and three times the cost reductions of others from AI. Deloitte describes the same problem in more sober corporate language: many organizations still need to bridge the gap between AI capability and operational reality if they want durable value at scale.  

That may be the most revealing statistic in the current AI cycle. The problem is no longer whether companies can buy access to powerful systems. Most can. The problem is whether leaders can redesign work around those systems instead of simply layering them onto old habits.

A badly run company with AI does not become a smart company. It becomes a faster version of its own confusion.

You can see this in small ways every day. Multiple teams generate different AI summaries of the same issue. Strategy meetings become longer because there are now more options, not fewer. Junior staff produce more output, but senior staff spend more time checking it. Everyone appears busier. Not everyone is moving.

Microsoft’s 2025 Work Trend Index captures the mood neatly: 53% of leaders say productivity must increase, but 80% of the global workforce says they lack the time or energy to do their work. AI may help close that capacity gap, but only if companies learn how to distribute decisions more intelligently rather than flooding workers with even more information.  

The new executive skill is orchestration

The business cliché for the past decade was “digital transformation.” The more revealing phrase now might be managerial transformation.

McKinsey’s latest survey notes that companies creating more value from generative AI are not just deploying tools. They are redesigning workflows, assigning senior leaders to oversee governance, retraining workers and embedding AI into business processes. That is a management story before it is a technology story.  

The modern executive, then, is becoming less of a pure decision-maker and more of an orchestrator. Not because judgment matters less, but because it now has to be applied at the right points in the system. The real question is not “Where can we use AI?” It is “Where must a human still own the call?”

That distinction matters. AI is excellent at scanning, sorting, drafting and pattern recognition. Humans remain essential when incentives conflict, when reputations are at stake, when trade-offs are ambiguous and when accountability cannot be outsourced. The companies that understand this early will build cleaner handoffs between machine speed and human responsibility. The ones that do not will create elegant-looking bottlenecks.

This is not just a tech-sector story

One reason this shift matters is that it is no longer confined to Silicon Valley or China’s platform giants. PwC’s 2025 Global AI Jobs Barometer, based on close to a billion job ads across six continents, found that industries more exposed to AI are seeing stronger growth in revenue per employee and faster wage growth. It also found a 56% wage premium for workers with AI skills in the same job compared with those without them.  

That suggests the upside is spreading beyond the obvious winners. AI is becoming a general business issue, not a specialist one. Retail, banking, logistics, consulting, healthcare, manufacturing and media are all being forced into the same uncomfortable realization: once the cost of routine knowledge work falls, culture and management quality matter more, not less.

The old fantasy was that AI would remove friction from work. In reality, it is exposing where the friction always lived.

Not in the machine.

In the meeting after the machine speaks.

The next divide in business

The next divide will not simply be between companies that have AI and companies that do not. It will be between companies that use AI to multiply clarity and those that use it to multiply noise.

That is a harder challenge, because it requires something software cannot purchase on a subscription plan: leaders who can choose a direction, absorb uncertainty, make trade-offs and create trust around difficult decisions.

The irony of this era is that as intelligence becomes cheaper, good management becomes more valuable.

AI may be transforming business. But the real test is whether business can transform itself.