The numbers come first.
A project that once consumed three weeks now closes in one. A full-day report wraps before lunch. And the moment that time is recovered, a new assignment lands in the queue. The efficiency gain is real. So is the transfer of its benefits — straight up the org chart.
The Pulse of Work in 2026 study, conducted by GoTo and Workplace Intelligence, surveyed 2,500 employees and IT decision-makers across ten countries. Employees using AI tools save more than two hours per day. That sounds like relief. It is not. Sixty percent of those same employees say they feel pressured to use AI to lift productivity. Fifty percent say they rely on it too much. Thirty-nine percent say that reliance is making them less intelligent. The productivity gain and the cognitive cost are arriving at the same address.
ActivTrak analyzed 443 million hours of work activity across more than 1,100 organizations. AI doubled time spent on email and messaging. Focused deep work fell 9%. A Harvard Business Review study published earlier this year found that after AI adoption, workers moved faster, took on broader task loads, and extended their working hours — often without being asked. The tools generate more activity. They do not generate more capacity for the thinking that makes activity worth anything.
Gallup's data shows 65% of employees say AI has improved their productivity. Frequent AI use among managers has doubled, from 15% to 30% since 2023. But the gains are not distributed evenly. Leaders report the strongest improvements. Individual contributors remain the least likely to receive guidance on using AI effectively. And 54% of managers say workplace expectations have directly increased because of AI. The bar is rising. The measurement framework is not.
Speed-based evaluation misses what AI cannot do. Catching hallucinations before they become decisions. Applying contextual judgment a model lacks. Turning an 80-percent-done draft into something actually good. That is executive-level cognition running in the background all day. When employees are evaluated on pace alone, that invisible labor goes unrecognized. Eventually it goes exhausted.
Researchers have a name for what follows: workslop. Fast output, low value, flooding organizations when speed is the only metric that counts. The GoTo and Workplace Intelligence study found that 43% of employees have submitted AI-generated content despite suspecting it contained errors. Seventy-seven percent say AI-generated work takes more time to review than work done by a human. Faster output, in practice, often means slower net progress.
The infrastructure gap makes it worse. Sixty-five percent of employees say employers are failing to equip them with the skills they need as AI absorbs more tasks. Eighty percent say most workers are not being trained properly on the tools. Yet nearly one in four IT leaders say AI mistakes have already affected customers or the company's bottom line. Adoption is accelerating. Investment in the human side of the equation is not keeping pace.
The metrics that look clean in a board presentation get elevated. The metrics that would reveal cognitive overload, declining focus time, and eroding judgment quality have not been built yet.
CEO Times reads it this way. Free enterprise runs on productivity, and AI is a genuine force multiplier. But capital rewards clear rules — including honest accounting. When companies book the efficiency gain and bury the human cost, they are not running leaner. They are borrowing against workforce capacity without disclosing the liability. The first intervention is the simplest and the rarest: give the time back. Organizations that treat recaptured hours as an asset to reinvest in people — training, judgment, depth — will compound the advantage. Those that treat every freed hour as a slot for one more task are running a short trade on their own talent base. The market will eventually price that correctly.
