Two years ago the challenge was convincing sceptics. Today I see the opposite problem in most organisations: people are not sceptical, they are worn out. Another tool, another training, another pilot, another town hall about the future of work. The organisation nods along and quietly returns to the old way of working.
I call this AI fatigue, and I treat it as seriously as any strategic risk, because it burns something expensive: your people’s willingness to change.
The numbers behind the exhaustion
The fatigue is measurable. In Gallup’s 2026 State of the Global Workplace research, half of US employees used AI at work at least a few times in the first quarter of 2026, but only 13 percent use it daily, and just 9 percent feel very comfortable with it[1]. Pew Research finds a third of workers overwhelmed by AI at work, and more than half more worried than hopeful about it[2]. MIT’s State of AI in Business research adds the organisational side: 95 percent of corporate GenAI pilots deliver no measurable P&L impact[3]. People are not tired of AI. They are tired of AI programmes that go nowhere.
Why fatigue happens
Fatigue is rarely caused by too little communication. It is caused by too much initiative and too little decision. When everything is a priority and nothing is ever formally stopped, every new AI announcement lands on top of unfinished ones. People learn that the safest response is polite passivity: attend the workshop, praise the demo, change nothing.
The same MIT research contains the most telling number of all: while only about 40 percent of companies have an official LLM subscription, roughly 90 percent of employees already use personal AI tools for their work[3]. The energy is there. It is simply not flowing through the official programme, which is exactly why another official programme is not the answer.
Activation is different from adoption
Most AI programmes measure adoption: licences, logins, tools rolled out. I care about activation: does someone actually work differently on a normal Tuesday? Activation needs three things adoption metrics never show. A use case anchored in someone’s real workflow, not a generic one. Explicit permission about what to stop doing, because time for new ways of working has to come from somewhere. And a visible early result, small and real, that colleagues can see without a slide deck.
| Fatigue signal | Activation response |
|---|---|
| Every initiative still called a pilot | Two or three cases, shipped |
| Polite passivity in workshops | One real result, visibly used |
| New tools on top of unfinished ones | Something explicitly stopped first |
| Enthusiasm as a KPI | Ownership as the KPI |
Two findings show what actually moves the needle. Gallup finds employees whose manager actively supports AI use are more than twice as likely to use it frequently, yet only about three in ten say their manager does[1]. McKinsey’s operating-model research points the same way: the companies that create value redesign workflows and train people first, tools second. Emirates Global Aluminium put thousands of employees through a dedicated academy before scaling AI across its operations, and documented more than 120 million dollars in impact[4]. Activation is a leadership job, not a licence purchase.
What I see in the field
I see this pattern from both sides: in a DAX headquarters environment and in conversations with mid-sized owners. The honest version usually surfaces after the official meeting ends: people quietly admit they stopped opening the chatbot weeks ago. Not because it is bad, but because nobody connected it to a decision they actually own. Fatigue is not resistance. It is unanswered effort. Treated with respect, it converts back into energy remarkably fast.
What this means for you
If your teams are AI-tired, do not schedule another inspiration session. Pick one workflow that matters, one team that wants it, one owner with authority, and produce one real result within weeks. Then let that result do the internal marketing. Energy returns when people see decisions and outcomes, not announcements.
This is the work we track at Quintellix week by week: which AI moves create real advantage, what the EU AI Act actually requires, and where activation beats another rollout. If your organisation is somewhere between chatbot and impact, that is exactly what our AI Practice is built for.
Personalised answer within 48 hours.
Sources
- Gallup, State of the Global Workplace 2026: The Human Side of the AI Revolution. gallup.com
- Pew Research Center, U.S. Workers Are More Worried Than Hopeful About Future AI Use in the Workplace, February 2025. pewresearch.org
- MIT NANDA, The GenAI Divide: State of AI in Business 2025; reporting by Fortune, August 2025. fortune.com
- McKinsey & Company, The Operating Model Advantage: Why AI Winners Are Rewiring Their Organizations, July 2026. mckinsey.com
The AI Act Just Moved. Your Deadline Did Not.