A lot of businesses are measuring AI success by the wrong thing. They count how many tools they’ve rolled out, how many seats are provisioned, how many workflows have AI baked in. What they’re not measuring is what happens to the time AI saves. A major new report from Boston Consulting Group shows this is where most of the value is leaking out.

BCG released its fourth annual AI at Work survey on June 3, 2026, drawing on responses from 11,749 workers across 14 markets. The headline finding is one that should give pause to every business leader who thinks they’ve solved their AI problem: nearly half of workers now spend more time managing and directing AI than doing the actual work.

That is a fundamentally different kind of job than what most people were hired to do.

What the Numbers Actually Show

The survey found that 47% of workers report spending more time managing and directing AI tools than performing the underlying tasks themselves. Frontline employee adoption is now at 74%, up more than 20 percentage points over the past two years, so this is not a story about reluctant adoption. People are using AI. They’re just doing it in a company environment that hasn’t figured out what comes next.

The productivity gains are real. Among frontline workers who use AI regularly, 42% save at least one full workday per week from their schedule. A workday every week, returned to them. That is substantial.

But here is the problem: 66% of workers say they receive little or no guidance on how to reinvest that recovered time. The hours are saved, and then they evaporate into busywork, informal tasks, or expanded workloads instead of going into strategic thinking, higher-skill work, or anything that creates actual business value.

BCG calls this the defining organizational challenge of AI in 2026: AI is moving faster than companies are.

The Strategy Gap Is Bigger Than the Tool Gap

One finding from the report stands out as particularly relevant to how businesses should be thinking about their AI investments. Companies with a clear AI strategy see 25 percentage points more impact from AI than companies without one. Companies that simply buy better tools, without the strategic framework, see only 5 percentage points more impact.

That is a five-to-one multiplier. Strategy beats tooling by a factor of five.

This is not a new idea in theory, but the data makes it strikingly concrete. Most of the conversation in enterprise AI has been about which model is best, which platform to choose, which workflow to automate first. The research says all of that matters far less than whether leadership has a coherent view of what AI is actually for, and how the people in the business should change how they work as a result.

The Agent Shift Is Already Happening

The survey also captured a notable acceleration in AI agent adoption. Thirty percent of workers now report that AI agents are integrated into their workflows, which is more than double the percentage from the previous year’s survey.

And the forecast from inside those same organizations is striking: 65% of managers and leaders believe AI agents will take over at least half of their job responsibilities within the next three years.

Whether that number turns out to be accurate or overstated, the direction is clear. AI agents are moving from pilot to production faster than most analysts predicted even 12 months ago.

The “Joy Paradox” in Practice

The report also documented a dynamic BCG describes as the “joy paradox”: 67% of workers say AI has improved their job satisfaction, but 41% simultaneously report that AI has increased their cognitive load. Work is both more rewarding and more demanding at the same time.

This happens because AI handles the repetitive parts of a role, but what remains is often the harder, more ambiguous, more judgment-intensive work. The easy decisions get automated. The difficult ones land more directly on the human. For workers without clear frameworks for how to approach that work, the result is more pressure, not less.

The Skills Conversation Is Urgent

Seventy-two percent of survey respondents say AI has significantly changed the skills required in their role. That is nearly three in four workers describing a meaningful shift in what their job actually demands.

This number matters beyond individual workers. It has direct implications for hiring, training, team structure, and how businesses think about retaining and developing people. If the skills profile of most roles has materially shifted, and two-thirds of employees are getting no guidance on how to adapt, the gap between AI investment and AI outcome will keep widening.

What This Means for Business

The BCG findings paint a clear picture of where most businesses are stuck. They have invested in AI tools. Adoption is genuinely high. Workers are saving real time. But the organizational layer of strategy, guidance, skill development, and cultural change has not kept up.

Three things are worth taking from this data:

1. Time savings without redirection are not productivity gains. If workers save a day per week and then fill it with low-priority tasks or expanded administrative work, that saved time goes nowhere useful. Companies need an explicit answer to the question: what should people do with the hours AI returns to them?

2. AI strategy is not a technology question. It is a business design question. The five-to-one multiplier on strategy versus tools is a strong argument for treating AI governance as a leadership priority rather than an IT priority. The companies getting the most from AI have leaders who made deliberate choices about how AI changes how the organization operates, not just which software to buy.

3. The agent transition needs a plan. With 30% of workflows already involving AI agents and that number accelerating, businesses without a structured approach to agent deployment are making decisions by default. Every agent integration is implicitly a decision about which tasks stay human and which do not. The companies that make those decisions deliberately will have more control over what their workforce looks like in three years.

The data from BCG’s survey suggests that the gap between AI investment and AI return is not primarily a technology problem. It is an organizational one. The businesses that close that gap fastest are the ones with clear strategies, strong data fundamentals, and people who understand what AI can and cannot do.

Building those capabilities is exactly the work that turns AI spend into AI value.

---

Enterprise DNA helps businesses build the data skills, AI strategies, and operational systems needed to capture real value from AI adoption. Learn more at enterprisedna.co.