---
title: Getting Executive Teams Hands-On with AI
source: https://steadman.ai/newsletters/david/executive-sessions.html
published: 2026-07-23
summary: A planning guide for AI sessions at executive retreats and leadership conferences, six goals, what works and what doesn't, and six formats, from three and a half years of running them.
---

# Getting executive teams hands-on with AI

What works, what doesn't, and six ways to run it: a planning guide for retreats and leadership conferences, from three and a half years of running them.

By David Boyle, Steadman (steadman.ai). 23rd July 2026. From the thinking behind David's Saturday AI Thoughts: https://steadman.ai/newsletters/david/

Our first AI session for a leadership audience ran on 2nd February 2023, nine weeks after ChatGPT launched: a hands-on workshop at a global consumer goods brand's international leadership conference, with executives generating campaign images for their own products using tools most of the room had never opened. This month the same company came back to start planning the next one.

We have run these sessions continuously since: for an international strategy consulting firm, for boards, for private equity firms, for broadcasters, for founders. The requests increasingly sound the same: "our leadership team gathers once a year; how do we use that time on AI well?"

This page is our answer. Use it to plan your own. And if we can help, we will run it with you.

## I. Start with the goal, not the agenda

The most useful question in planning one of these days is about purpose: what do we actually want to achieve? Most sessions are designed backwards, agenda first. The agenda is the last thing to decide.

A leadership session can do six jobs. Pick two. A day that tries to do all six does none.

* **Believe.** "This is real and it applies to us." For rooms that have not yet felt it. Needs live demonstration on your business, using your data and your problems.
* **Upskill.** Individual capability. Needs hands-on practice, small groups, real tasks, and someone working the floor.
* **Accelerate.** Fuel on the fire for the pioneers you already have. Needs advanced content, competition, and peers who stretch each other.
* **Align.** A shared language and shared priorities at the top. Needs the executive team in a room on its own, working the tools themselves.
* **Unblock.** Decisions only this room can make: licensing, permissions, budget, governance. Name them before the day and leave with them made.
* **Signal.** Visible sponsorship that makes the rest of the organisation want in. Selective invitations and leaders seen using the tools do more than any memo.

Then write the agenda as questions the group must answer rather than topics to cover. Steven Rogelberg (University of North Carolina at Charlotte) finds a topic agenda invites drift while a question agenda forces a decision. It also gives you a scoreboard: at the end of the day, either the questions got answered or they didn't.

The test for every design choice: the Monday after. Judge the day by what people do differently at their desks the following week, and by what their colleagues notice. The feeling in the room at 5pm is a poor proxy for either.

## II. Know which room you are in

Every organisation we have worked with shows the same split. A small group does most of the innovating; the rest use AI like a slightly better search box.

That split gives you three different rooms, and they need different days. The pioneers need fuel: advanced material, competition, each other. The majority need nudges, foundations and proof from peers they recognise. The executive team needs personal fluency and decisions. Mixing them is the most common design mistake there is: content pitched at everyone lands with no one.

Leaders reliably misjudge the audience. McKinsey's "Superagency" research (published January 2025, fielded late 2024) found C-suite leaders estimated 4% of employees used AI for at least 30% of their daily work; the employees themselves reported 13%, three times higher. The gap the day needs to close is often the leadership's own.

## III. What works really well

* **Hands-on beats watching, and the evidence agrees.** Harvard Business Publishing and Degreed (2025) found people fluent with AI were twice as likely as everyone else to have learned by experimenting. Design for at least half the time spent doing.
* **Their business, never a canned demo.** At one firm's annual partner conference, tables of six built genuine client pitches from scratch in a morning: research, simulated client interviews, a finished proposal. It worked because the work was real.
* **Peers on stage beat outside experts.** A short panel of colleagues showing real work they have already shipped grounds the day in proof rather than promise. Moderna built the same principle into its rollout: a prompt contest surfaced its top 100 power users, who became a network of internal AI champions, with office hours in every business line. Its own chatbot, mChat, launched at the start of 2023 and passed 80% adoption, and Moderna set itself a six-month target of getting everyone proficient (OpenAI's case study on Moderna). The outside expert's best work happens backstage: finding and coaching the presenters.
* **Nobody starts cold.** Before the day, everyone gets set up and completes one small piece of machine-graded homework with personal feedback. Everyone walks in having done one real thing. Pre-work needs teeth: at one large conference we ran, where it was optional, barely one person in seven did it.
* **A power user on every table.** Identified in advance, ideally from usage data, and seated deliberately. The single biggest driver of a good room.
* **Mix the room on purpose.** Offsites' durable payoff is rewired working relationships, and only attendees get it (Kneeland, Gardner and Kleinbaum, Strategic Management Journal, 2025). Seat across functions, protect informal time, choose the invite list deliberately.
* **Structure who speaks.** Hierarchy suppresses candour even among the senior: in Amy Edmondson's survey of 805 senior leaders, just 7% believed their own silence had had no impact. Small groups, round-robins, rotating facilitation.
* **Competition, with locked rules.** Tables submit real work; it is judged; there are prizes. AI can judge at scale, against criteria agreed with leadership in advance.
* **Leadership opens, and stays.** McKinsey's State of AI (November 2025): high performers were three times more likely to have senior leaders who visibly own and commit to AI, not just approve the budget.
* **End with something written, and someone waiting.** A report: what we saw, what it says about where you are, recommendations, next steps. At conference scale, commitment pairs with a 15-minute check-in a month out (David Burkus).

## IV. What doesn't work

* **Mixed rooms.** Pick a room (see section II).
* **Death by deck, and a cast of thousands.** Presentation marathons crowd out conflict and decisions (Doug Sundheim, Forbes, 2025); oversized invite lists turn working sessions into town meetings.
* **The offsite high.** Inspiration with no decisions, no owners and no dates evaporates within days.
* **Leaving the plumbing to chance.** The technology needs the same careful preparation as the content: hundreds of people downloading, logging in and submitting at once is a real load test. Pre-stage small files, test the venue connection, health-check tools minutes before doors, and agree a plan B for every dependency.
* **An ending that fizzles.** Kahneman's peak-end rule: people judge an experience by its peak and its close. Size the judging to the break; protect the ending.
* **Changing the rules mid-game.** Lock the rules in advance and let them play out.
* **Teaching speed without judgement.** AI's first drafts are mostly impressive, and that is exactly what makes the weak minority dangerous: it arrives with the same confidence as the rest. Reviewing and improving the output is a skill in its own right. Build a Check, Edit, Own beat into the run of show; add a red-team step where the model attacks its own draft.
* **The relapse.** Frequent AI use is 78% among employees who strongly agree their line manager actively supports it, 44% among everyone else (Gallup, February 2026). Book booster sessions before the day itself.

## V. Six ways the day can work

Formats we run, in rising order of ambition. Every one ends with the written report. Most rooms start from our signature keynote.

### Our signature keynote: The Past, the Present and Two Futures

A 20 to 25 minute keynote with live demonstrations throughout, built for mixed rooms of executives and builders. It answers the question every leadership audience actually has: where are we in this story, and what should we do about it? Four buckets:

1. **The past.** The tools and habits we have already left behind. Quick and affectionate: weren't they fun at the time. Then we move on, because you should have too.
2. **Now, but not everywhere.** What is already working brilliantly in some corners of your organisation and invisible in others, and how to get more from what you already have. This bucket doubles as the call to action for the rest of your day.
3. **Coming, certainly.** Technology we can demonstrate live on stage today, even though your organisation cannot use it yet: awaiting licences, permissions and policy. AI as a teammate you can email, message on Teams or WhatsApp, and hand work to. You watch it happen on screen.
4. **Speculative, but visible.** Where the signs point next. No product yet, though the ingredients are on the table. What it would mean for your business if they combine the way we think they will.

Three practical buckets and one lovely retrospective.

1. **The conference keynote.** 20 to 40 minutes, any size room. Job: believe. The keynote above, applied to your context.
2. **The peer showcase.** 60 to 90 minutes, any size room. Job: believe + signal. Your own people on stage showing real shipped work. We find, coach and compere.
3. **The structured debate.** 45 to 60 minutes, any size room. Job: believe + align. A motion with teeth ("AI will replace us"), two teams arguing assigned positions, timed rounds, live audience questions, the same poll before and after; the winner is whoever moves the room. We did not invent the format, but we have run it at conferences since 2018 and it is the best way we know to surface the issues a room actually holds. The first time, a room that began 77% against the motion ended roughly even.
4. **The executive working session.** Two to three hours, the executive team only. Job: align + unblock. Personal fluency first, then what this means for us, what we will refuse, what gets unblocked today. We arrive with a straw model to react to.
5. **The hackathon with an AI judge.** Half a day, 20 to 300+ people. Job: accelerate + upskill. Tracks by function, real challenges and data, AI judging against criteria agreed with leadership, locked rules, prizes, results reveal in the room.
6. **The full immersion day.** One day, your most active users, with executive sponsorship in the room. Job: accelerate + signal. Keynote, showcase, hackathon and an executive working session in parallel.

Every format ends with the written report.

Other shapes worth knowing: prompt-a-thons (Michael Schrage, MIT Sloan: 100+ run since 2023; iteration beats one-shot, teams beat individuals), reverse mentoring (KPMG, Fidelity, GE), and board-level AI war-games adapted from cyber-crisis exercises.

## VI. Where this thinking comes from

Most of these ideas appeared first in David's Saturday AI Thoughts, the weekly email: https://steadman.ai/newsletters/david/

Relevant editions: Average by default (20th June 2026); What boards accept (16th May 2026); The bill and the harness (2nd May 2026); A thousand small bargains (27th June 2026); How We Got Here (30th May 2026); Futures and fears (18th July 2026); The system and the surrender (28th March 2026).

## Get in touch

Use this freely: it is what we would do. AI transformation failures are organisational, not technical, and a leadership gathering is the one moment in the year when the organisational part is all in one room.

If you would like help: david@steadman.ai