HR & L&D PLANNING RESOURCE
How to Build an AI Training Plan for Your Employees
An employee AI training plan should define who needs training, which tasks they will practise, what tools they may use and how progress will be assessed. The example below shows how to organize a four-week programme and adapt it for different roles.
Prepared by School of Core AI · Review date: 12 November 2026
1. Training-Needs Assessment
Roles
List the roles that will use AI weekly (business users, managers, developers) and their daily workload.
Current skills
Record what each group already does with AI tools today, if anything. Self-reported plus one practical check.
Tool access
Which tools and licences participants may use, under whose account, with what data boundaries.
Task selection
Pick recurring tasks with clear inputs and outputs — documents, briefs, analyses, code reviews.
Data restrictions
What material may enter a tool, what must stay out, and who verifies compliance.
2. Role-to-Curriculum Map
Role
Curriculum focus
Starting depth
Evidence produced
Business users
Everyday AI workflows on their own documents
No coding; tool familiarity
Reviewed work sample + accuracy checklist
Managers
Use-case selection, pilots and review habits
Process literacy; no coding
Pilot brief with success criteria
Developers
Building and evaluating AI-assisted workflows
Working Python
Working exercise with evaluation notes
3. Illustrative Four-Week Schedule
One two-hour session per week per participant, with a practical exercise between sessions. Adjust pacing to the role and workload.
Week
Theme
Exercise
Review point
1
Foundations and task selection
Map one real weekly task; identify data boundaries
Task suitability check
2
Guided practice
Produce a draft with source material using approved tools
Source-supported draft reviewed
3
Applied workflow
Complete a role-based workflow exercise end to end
Work sample reviewed against checklist
4
Quality and independence
Self-review with checklist; plan follow-up tasks
Repeatable checklist demonstrated
4. Cohort Capacity: the Arithmetic Behind a Plan
Example: 50 employees, four weeks
- · Two groups of 25 participants.
- · Each group: one two-hour session per week for four weeks.
- · Scheduled training: eight hours per participant.
- · Live instructor delivery: 16 hours across the two groups.
- · Excludes preparation, marking, administration and follow-up.
Stress test: 200 employees, eight weeks
- · At 25 participants per cohort and one parallel session per week, 200 employees need 8 cohorts — 16 weekly instructor delivery hours.
- · Adding a second parallel instructor halves calendar duration; three cohorts can run concurrently.
- · Session frequency, cohort size and instructor count are the three levers — name your assumptions in the plan.
This is a planning calculation for the reader, not a statement of SCAI's minimum size or available instructor capacity.
5. Before/After Assessment Rubric
Criterion
Before training
After training
Task quality
Work meets only part of the task definition
Work meets the task definition with named limitations
Verification
Outputs accepted without source checks
Key claims traced to sources; errors flagged
Tool use
Tool use ad hoc; data rules unclear
Approved tools and data boundaries applied consistently
6. Follow-Up Scorecard
Usage check (2–4 weeks)
Are participants still completing the trained tasks with the checklist? Owner: L&D.
Task outcome (6–8 weeks)
Compare rework and turnaround on the trained tasks. Owner: department manager.
Escalation trend
Track questions and errors reaching experts. Owner: team lead; feed back for refresher scope.
7. When Internal Delivery Is Enough — and When It Is Not
Internal delivery usually fits when
- · The tasks are simple, repeated workflows with clear rules.
- · A subject-matter expert inside the company can run the sessions.
- · The tool stack is already approved and supported.
Commission external support when
- · Multiple roles need structured, different curriculum at the same time.
- · You need an independent before/after assessment design.
- · Engineering teams need reviewed retrieval, tool-use and evaluation work.
Need a trainer to deliver this programme for your teams? Review the corporate AI training programmes and request a proposal →