Leaders Are the Heaviest Users. And the Weakest Signal.
Oct 31, 2026
|
Key Takeaways â–¶ Leaders are now the heaviest AI users at work: 51% of US leaders use it frequently, against 26% of individual contributors (Gallup, 2026) â–¶ 64% of UK managers say their senior leaders encourage AI experiments. 13% strongly agree those leaders experiment with the tools themselves (CMI, 2026) â–¶ A leader’s own use becomes their proof that adoption is happening. So nobody ever tells the team what AI is for here. â–¶ A clear statement of what AI is for comes with 25 more points of measurable business impact; better tools alone, 5 (BCG, 2026) |
|
51% of US leaders use AI a few times a week or more, against 26% of individual contributors (Gallup, Q2 2026) |
13% of UK managers strongly agree their senior leaders experiment with AI themselves (CMI, 2026) |
+25 points of measurable business impact with strategic clarity, against +5 with better tools (BCG, 2026) |
Leaders did what every conference told them. They got hands-on with AI. It made the gap below them wider.
Leaders are now the heaviest AI users in the building. Gallup has 51% of US leaders using it a few times a week or more. Individual contributors: 26%.
So the top of the organisation is using it most. Has that pulled everyone else along?
Ask the managers. The Chartered Management Institute asked 1,019 UK managers in February and March. Of the 717 whose organisation uses AI, 64% agreed their senior leaders encourage people to experiment. 13% strongly agreed those leaders actively experiment with it themselves.
Two thirds have been told to have a go. One in eight is sure they’ve seen the boss do it.
Part 1 of this series argued that AI’s gain lands with the individual and gets spent there, because nothing the organisation asks for has changed. This post is about what happens when that individual is the chief executive.
A leader’s AI use is a private gain like everyone else’s. Leaders read it as the organisation’s progress.
The Heaviest Users in the Building
Three years ago fewer than one in five at any level used AI frequently. Then the top pulled away.
|
Interactive: frequent AI use by level The higher you sit, the more AI you use Share of US employees using AI a few times a week or more. Tap a year. Q2 20262023
Both groups roughly tripled their frequent use in three years. The gap between leaders and individual contributors grew from 8 points to 25. Source: Gallup, AI and Workplace Productivity: What Leaders Need to Know, Q2 2026 data (US employees), published 27 July 2026, updated 30 September 2026 |
The UK looks the same from a different angle. PwC asked 2,023 UK workers whether they had used AI at work in the past year: 85% of senior executives, 73% of managers, 35% of non-managers. And just 28% of those non-managers feel confident using it.
The View from the Top
A chief executive drafts the board paper with AI and summarises the week’s reports with it. Hours saved. She knows it works. And she has a feeling about the organisation no survey gave her: we’re on this.
|
The chief executive Uses AI every day and knows it works. Has said at two town halls that she wants people experimenting. Believes, honestly, that the organisation is well on the way. |
|
The head of department Has heard the encouragement. Has never seen the chief executive use it, doesn’t know what she uses it for, and can’t tell whether “experiment” means in the open or on the quiet. Says nothing and carries on. |
|
The team Most of them already use it, at their own desks, on their own work. Nobody has told them what it’s for here, so they keep it to themselves. Same deliverables, same deadlines. |
That’s the normal case.
Wharton and GBK Collective asked 1,007 managers and executives at large US companies the same questions as 409 entry- and mid-level employees, in July. 78% of the managers said staff get adequate AI training. 56% of the staff agreed. 79% of managers said leadership has set out a clear long-term AI vision. 64% of staff could see one.
|
Interactive: same organisation, two accounts Leaders describe a more trained, more confident, more changed workplace than their staff do Share agreeing with each statement. Tap a statement. Staff get adequate AI trainingStaff are confident using AILeadership has a clear long-term AI visionOur workflows have changed significantlyAI has improved efficiencies
Training, confidence, vision and workflows: managers and above (1,007) against entry- and mid-level employees (409) at US companies with 1,000+ staff. Efficiencies: CEOs against individual contributors, 16,000+ professionals in 11 markets. Source: Wharton Human-AI Research and GBK Collective, The Adaptive Advantage, 13 October 2026; Korn Ferry, Workforce 2026, 8 September 2026 |
|
There is no group in this study where managers and the employees they manage describe the same working environment. Wharton Human-AI Research and GBK Collective, The Adaptive Advantage: Who’s Winning With AI and How, October 2026 |
Every gap points the same way. The further up you sit, the more adoption you see, because the evidence in front of you is your own week, and your own week really has changed.
Once your own week is the evidence, you stop looking for the one thing the team still hasn’t had: a plain statement of what AI is for here.
Hang on, don’t leaders talk about AI constantly? They do. But “experiment” with no task attached, no deadline moved and no example to copy is a permission slip for private use. Exactly what Part 1 said the organisation already has too much of.
BCG asked frontline staff about this directly in its 2026 AI at Work survey. 33% say leadership communicates clearly about AI. 28% see strong alignment between what leaders say and what the organisation actually does.
We see the same gap before any training starts.
|
G3NR8 data: leaders and their teams, same questions In one organisation we assessed, a Gulf financial regulator, we surveyed 95 people while designing their AI programme: 10 leaders and 85 team members. 70% of the leaders rated their teams “absolutely ready” for AI. 38% of the people assessed felt confident using it. 94% were already using it. |
Ten leaders convinced. Eighty-five staff, most already using it and most of them unsure. Wharton’s gap in one room.
What Actually Moves a Team
BCG split 9,923 regular AI users by two things: whether their organisation had strategic clarity (AI named as a priority, a clear plan, guidelines for what to do with saved time) and whether they had good access to tools. Then it asked who could point to measurable business impact.
With neither, 55%. Better tools on their own took it to 60%. Clarity on its own took it to 80%. Both together, 83%.
|
+25 points of measurable business impact where there’s strategic clarity about what AI is for. Better access to tools on its own: 5. Source: BCG, AI at Work 2026 survey, June 2026 (9,923 regular AI users of 11,749 workers surveyed, 14 markets) |
And it costs a leader one decision and one plain paragraph.
The other thing that moves a team sits one level down. Gallup’s Q2 2026 data, in organisations that have adopted AI: where employees strongly agree their manager supports the team’s use of AI, 78% use it frequently. Where they don’t, 44%. The signal that works comes from the person who hands out the work.
What to Do This Month
|
1. Stop reading your own use as the organisation’s progress Wharton calls the seniority gap “the most consistent finding” in its 2026 data. Your week has changed. Ask the team whether theirs has. Don’t ask the leadership meeting. |
|
2. Decide what AI is for here and say it once, plainly Which tasks, what happens to the time it saves, and what you’ll now ask for differently. Only 33% of frontline staff say they’ve heard anything that clear (BCG, 2026). |
|
3. Be seen using it, on real work, in the room Run one of your own recurring tasks with AI in front of the team, show the prompt and where it fell short, then change the deadline to match. Encouragement from a distance is what 64% already get. |
|
4. Equip the manager one level down Adoption is decided where the work is handed out. Just 12% of UK managers feel very confident supporting their team’s AI adoption (CMI, 2026). Give each manager one named task to change and the authority to change the ask. |
|
5. Make the reward and the safety real below your own level Microsoft’s 2026 Work Trend Index found leaders are twice as likely as employees to say reinvention is rewarded whatever the outcome (21% against 10%), and more likely to feel safe suggesting new ways of working (81% against 67%). If the only people who feel safe experimenting are at the top, the experiments stay at the top. |
|
6. Re-measure from the staff side Wharton: 39% of managers and above describe an organisation that clears four of five adoption conditions. 26% of their staff do. Ask the team the same questions you asked the leadership team and put the two sets of answers side by side. |
Run the first and the sixth this month and you’ll find out which organisation you’re running.
Which of your senior team has run real work through AI in front of their people, then changed what they asked for? If you can’t name one, the team is still guessing.
Next in the series: the squeezed middle. Managers are the one layer that can change the ask for a team, and the one organisations have equipped least. Next week, what the 2026 data says about them.
|
FREE REPORT: The AI Adoption Gap 2026 Everyone’s using AI. So why isn’t anything changing? We’ve pulled the whole series into one report you can take into your next leadership meeting. âś… The 2026 numbers from Gallup, McKinsey, Microsoft and BCG, all sourced âś… What the 6% making money from AI did differently âś… Leaders, managers, HR and operations: where each one is stuck âś… The collection test, ready to run with your teams this week G3NR8 has trained 150+ large organisations. This is what we see on the ground. |
|
Get UPSKILL in your inbox every week Our free weekly newsletter on what’s actually working with AI at work, with prompts and skills you can use straight away. Five minutes, once a week. |
Sources & References
| Gallup: AI and Workplace Productivity: What Leaders Need to Know, 27 July 2026, updated 30 September 2026 (Q2 2026 data, US employees) - frequent AI use (a few times a week or more): leaders 51%, managers 36%, individual contributors 26%, up from 17%, 15% and 9% in 2023; in organisations that have adopted AI, 78% of employees who strongly agree their manager supports the team’s AI use are frequent users, against 44% of those who don’t. gallup.com |
| PwC UK: One in five UK employees are now using AI everyday at work but 44% report a heavier workload, 1 October 2026 (2,023 UK workers, UK cut of the Global Workforce Hopes and Fears Survey 2026, fielded May to June 2026) - 85% of senior executives, 73% of managers and 35% of non-managers have used AI at work; 28% of non-managers feel confident in their AI use. pwc.co.uk |
| Chartered Management Institute (CMI): Artificial Intelligence; Real Leadership: The Management Imperative in AI Adoption, 10 June 2026 (Managers’ Voice Poll of 1,019 UK working managers, 20 February to 9 March 2026; leadership questions asked of the 717 whose organisation uses AI) - 64% agree senior leadership encourages staff to experiment with AI tools; 13% strongly agree senior leaders actively experiment with the tools themselves; 12% very confident in their ability to support their team’s AI adoption; Lloyds Banking Group Leading with AI programme cited on p. 16. managers.org.uk |
| Wharton Human-AI Research and GBK Collective: The Adaptive Advantage: Who’s Winning With AI and How (2026 AI Adoption Report), 13 October 2026 (1,007 managers to C-suite and 409 entry- and mid-level employees at US enterprises with 1,000+ staff, fielded 9 to 29 July 2026) - adequate AI training 78% against 56%; staff confident using AI 79% against 62%; clear long-term AI vision 79% against 64%; workflows changed significantly 73% against 58%; 39% of managers and above against 26% of staff describe an environment clearing four of five adoption factors. knowledge.wharton.upenn.edu |
| Korn Ferry: Workforce 2026, 8 September 2026 (more than 16,000 professionals, 11 markets, all organisational levels) - 79% of CEOs report improved efficiencies from AI, against 51% of individual contributors. kornferry.com |
| BCG: AI at Work 2026 survey slideshow, June 2026 (11,749 workers, 14 markets) - measurable business impact among 9,923 regular users: limited clarity and limited tools 55%, limited clarity and strong tools 60%, strong clarity and limited tools 80%, both 83%; 33% of frontline employees say leadership communicates clearly about AI; 28% see strong alignment between what leaders say and what the organisation does. bcg.com |
| Microsoft: 2026 Work Trend Index Annual Report: Agents, human agency, and the opportunity for every organization, 5 May 2026 (20,000 AI-using knowledge workers, 10 markets including the UK) - reinvention rewarded regardless of outcome: leaders 21%, employees 10%; feel safe suggesting new ways of working with AI: leaders 81%, employees 67%. microsoft.com |
| G3NR8: Assessment survey, a Gulf financial regulator, 2026 (95 people assessed: 10 leaders, 85 team members, during programme design) - 70% of leaders rated their teams “absolutely ready” for AI; 38% of the people assessed felt confident using it; 94% already using AI tools. |
Frequently Asked Questions
|
Do leaders use AI more than their employees? Yes, by a wide margin. Gallup’s Q2 2026 data has 51% of US leaders using AI a few times a week or more, against 36% of managers and 26% of individual contributors. In 2023 those figures were 17%, 15% and 9%. In the UK, PwC UK’s 2026 survey, published in October, of 2,023 workers found 85% of senior executives had used AI at work, against 73% of managers and 35% of non-managers. |
|
Why doesn’t a leader’s own AI use spread to their team? Mostly because the team can’t see it. A leader’s personal use saves the leader time and convinces the leader that adoption is happening. It doesn’t tell anyone else what AI is for in this organisation, which is the message teams are waiting for. |
|
How big is the gap between what leaders and staff say about AI at work? Wharton and GBK Collective’s October 2026 report compared 1,007 managers and executives with 409 entry- and mid-level employees at large US companies. 78% of managers said staff get adequate AI training; 56% of staff agreed. 79% said leadership has a clear long-term AI vision; 64% of staff saw one. Korn Ferry’s Workforce 2026 survey of 16,000 people found 79% of CEOs reported improved efficiencies from AI, against 51% of individual contributors. Wharton calls the seniority gap the most consistent finding in its data. |
|
What should leaders actually do about AI adoption? Decide what AI is for in your organisation and say it plainly: which tasks, what to do with the time saved, and what changes in what you ask for. BCG’s 2026 AI at Work survey found that clarity comes with 25 more points of measurable business impact; better tools alone, 5. Then be seen using AI on real work in front of your team, equip the managers one level down, and make it safe and rewarded to change how work is done below your own level. |
|
How can a leader tell whether their organisation is really adopting AI? Measure from the staff side. Ask teams the same questions you ask the leadership team and compare the answers. Wharton found 39% of managers and above describe an organisation that clears four of five adoption conditions, against 26% of their staff. Then run the collection test from Part 1 of this series: which recurring task is done differently because of AI, where the saved hours went, and what has changed in what the organisation asks for. |
This is Part 2 of 6 in The Adoption Gap series, on what the 2026 research says about who’s using AI at work and what it’s actually changing. Published 31 October 2026.
Keep informed with the newsletter for PE operating partners and the portfolio companies they back.
Â
Get operational insights and trends, AI frameworks, resources and real deployment stories.
