Generative AI · Capability building
AI training should change the work, not just the vocabulary
Five design choices that help generative AI learning transfer into everyday team practice.
Do Marketing Right · 18 August 2026
Knowing the names of AI tools is not the same as knowing how to use them well. Useful corporate training must connect new capability to the decisions, documents and workflows people already own.
Start with real work
Use relevant tasks rather than generic demonstrations. Familiar work makes it easier to compare quality, spot risk and understand where AI helps.
Teach a shared method
Teams need a common way to frame a task, supply context, set constraints and check outputs. A method travels better than a list of prompt tricks.
Build review into the workflow
Human review is not an optional final slide. Participants should practise checking evidence, judgement, tone, privacy and downstream consequences.
Design for different confidence levels
Good facilitation lets confident experimenters stretch without leaving newer users behind.
End with a next-week workflow
Every participant should leave knowing which real task they will improve, how they will measure usefulness and where they must involve another human.