Role 1 · Leadership
Leadership
Execs who set strategy and allocate budget. Without them, training hits the ceiling.
Training · Leadership, Lab, and Crowd · $10-25k
Your team will use AI on Monday, or your AI strategy is theater.
Three tracks for three audiences. Execs need a mental model. ICs need hands-on patterns. The wider org needs the cultural setup. Pick one or stack all three. Anchored on Mollick's Leadership-Lab-Crowd. Built to survive without me.
§01
Leadership-Lab-Crowd (Ethan Mollick, Wharton, May 2025).
Three internal roles every AI-adopting org needs. Most programs train only the Crowd, hit the 20% ceiling, and stall. The fix: train all three in sequence.
This is also where "secret cyborg" lives. People use AI privately but won't share, because nobody signaled it's safe to admit.
Role 2 · Lab
Lab
A small builder team running experiments. Without them, no examples to teach from.
Role 3 · Crowd
Crowd
The 80% using AI in daily work. Without them, only power users benefit.
§02
Three tracks in sequence so Leadership, Lab, and Crowd reinforce each other.
Track 1 · 2 sessions, 90 min each, plus async reading
Leadership (executive layer)
For execs setting AI strategy. Three-Horizon portfolio framing. Budget defense language. Eval oversight without becoming a bottleneck.
Track 2 · 4 sessions, 2 hours each, plus hands-on builds
Lab (internal builder team)
For your 5 to 10 builders. Prompt engineering for production. Eval setup (Braintrust or equivalent). Multi-model routing and cost control. Agent literacy: single-shot vs multi-step vs multi-agent.
Track 3 · 2 sessions, 60 min each, plus async references
Crowd (everyone else)
For ops, sales, CS, engineering. Daily-use patterns. Prompting for the use case, not the demo.
§03
Curriculum tied to your workflows, not vendor PDFs.
Generic AI training flatlines after the trainer leaves. This one lives in the knowledge base your team and your AI both work from.
Without · What most AI training looks like
Generic and disposable
Prompt workshops disconnected from real workflows. Trainer leaves, team forgets. Vendor tutorials that lock you in.
With · What you get here
Tied to your workflows, kept by your team
Built on your actual use cases. References in the knowledge base your team and your AI share. Adoption tracked, 30/60/90 follow-up.
§04
4 to 8 weeks depending on org size and number of tracks.
Me, an exec sponsor, one Lab lead. Live workshops, recordings, async. Reinforcement at 30/60/90.
Phase 1 · Week 1 · Diagnostic
Map current AI usage
Interview 6 to 10 across all three roles. Surface the secret cyborg. Pick highest-leverage Crowd workflows.
Phase 2 · Weeks 2 to 3 · Leadership track
Lock the strategy
Two exec sessions. Lock strategy, allocation, metrics. Signal it's safe to admit you use AI.
Phase 3 · Weeks 3 to 6 · Lab track
Hands-on builds
Four sessions with builders. Real internal use cases. Eval framework stood up.
Phase 4 · Weeks 6 to 8 · Crowd track
Use-case workshops
Two sessions per team. Use-case prompting. Contribution path back to the knowledge base.
Phase 5 · 30/60/90 post-engagement
Reinforcement
Adoption review. Course corrections on what stuck.
§05
End of engagement, all three tracks delivered. Six months later, the ceiling moves.
End of engagement · What you get
Three tracks delivered
Adoption baseline measured. Lab can run future training. Crowd has references in the knowledge base. Leadership has the diagnostic to flag regressions.
+6 months · Clients report
The ceiling moves
Adoption past 20%, typically 60% to 80%. Secret cyborg shrinks. Lab ships internal tools without outside help. New hires train through the OS.
§06
Frequently asked questions
A 4 to 8 week program that trains three internal roles every AI-adopting org needs: Leadership (execs who signal safety and set strategy), Lab (a small builder team running experiments), and Crowd (the 80% using AI in daily work). Anchored on Ethan Mollick's Leadership-Lab-Crowd framework (Wharton, May 2025). Most programs train only the Crowd, hit the 20% adoption ceiling, and stall. The fix is training all three in sequence.
$10-25k, 4 to 8 weeks, across one to three tracks. Leadership track: 2 sessions, 90 min each, plus async reading. Lab track: 4 sessions, 2 hours each, plus hands-on builds. Crowd track: 2 sessions, 60 min each, plus async references. Scoped against team size and track mix. Quoted in writing within 48 hours of the discovery call.
Most programs train only the Crowd. The Leadership layer never signals that AI use is safe to admit, so the "secret cyborg" pattern dominates (people use AI privately but won't share). The Lab layer never produces internal examples, so the Crowd has nothing to copy. Training all three in sequence, starting with Leadership, breaks the ceiling.
Companies whose AI strategy is set but whose adoption has stalled. Execs who need a mental model. ICs who need hands-on patterns. Wider orgs who need the cultural setup. Mid-market service firms, roughly $5M to $200M+ ARR. Pairs with Fractional Head of AI Transformation engagements where adoption is the bottleneck.
You can, but it underperforms. Leadership track first sets the safety signal. Without it, Crowd training teaches skills people are still nervous to use openly.
Crowd track scales. I've run 50-person workshops. Lab track stays small (5 to 10 builders). Leadership track is the exec layer, usually under 12.
Yes. Everything goes into the shared knowledge base your team and your AI both work from, or your shared drive. You own the curriculum. New hires train on it without me.
Optional. Most clients add it to the Lab track so one or two internal leads can run future Crowd sessions.
§07
What's included beyond the workshops.
The three tracks are the deliverable. These come with the engagement.
Bonus 01 · Standalone value $2k
Recordings of all sessions, yours to keep
Every Leadership, Lab, and Crowd session recorded. Yours to keep, for new-hire onboarding, refreshers, and the team members who couldn't attend live. Training scales without re-engaging me.
Bonus 02 · Standalone value $1.5k
The Leadership-Lab-Crowd playbook
The framework documented for internal reuse. Diagnostics, exercises, talking tracks for the exec sponsor. So Lab can run the next training cycle without me in the room.
Bonus 03 · Standalone value $2k
4 weeks of office hours after the last session
Weekly drop-in office hours for the first four weeks post-engagement. Live questions, prompt review, workflow troubleshooting. Catches the adoption regressions before they become structural.
Move past the ceiling
Move your team past the AI adoption ceiling.
Book the call. 30 minutes to diagnose where your team is on the Leadership-Lab-Crowd grid. $10-25k, 4 to 8 weeks.