← All capabilities
Training
Phase: OperateTeach teams to use and supervise AI for their specific job.
Training gives teams the competence to use AI well: how to ask it for things, how to review its outputs and when not to trust it. It’s not a generic "AI for everyone" course; it’s practical skill applied to each function’s real work.
What it includes
- Training per function: what AI does in each team’s specific work.
- How to supervise and correct an agent’s output without being a technical expert.
- Judgement to know when AI is wrong and when to escalate.
- Living playbooks that update with what the team learns.
When you need it
- People use AI, but badly: they accept outputs unchecked or prompt poorly.
- Only the technical team knows how to get value from the tools.
- Everyone reinvents how to use AI with no common criteria.
How it’s measured
- Trained people who reach competent use (not just attendance).
- Quality of supervised outputs from trained vs. untrained teams.
- Reduction in misuse errors after training.
Common mistakes
- A generic AI course that touches nobody’s real work.
- Training once and never updating when the tools change.
- Teaching people to use AI but not to supervise it or spot when it fails.
FAQ
- Isn’t it enough for the tool to be easy to use?
- Ease helps, but using AI well is judgement: knowing how to ask, review and distrust at the right moment. That’s trained. Without training, people either over-delegate or distrust everything.
- Is training an event or something continuous?
- Continuous. Tools and processes change, so training is a living playbook that updates, not a course given once and filed away.
Related capabilities
Missing any of these?
Start with an assessment: which capabilities you have, which you’re missing, and where it pays off most to start.
Talk to us →