Do you need AI champions in your team?
They help, but only with protected time, a clear mandate and leadership backing. Added on top of a full job with no support, champions tend to burn out.
What champions can do
An AI champion is a member of staff who helps colleagues use AI well in their part of the business. Done properly, the role closes the gap between central guidance and day-to-day work. Champions answer quick questions, spot good use cases, flag risks early and feed back what is and is not working. They also give cautious colleagues someone at their own level to ask, which often counts for more than a message from leadership. That peer route is one of the most effective ways of getting sceptical staff on board.
Why champion schemes stall
The common failure is treating the role as a badge rather than a job. Someone keen is named champion, given no extra time and expected to help everyone on top of a full workload. Before long they are either overwhelmed or quietly stop, and colleagues take that as a sign the initiative has ended.
What champions need
Protected time: A set amount of time each week or month, agreed with their manager and honoured when work gets busy.
A clear mandate: A written description of what the role covers, such as running drop-ins, collecting use cases and escalating risks, and what it does not, such as approving new tools.
Visible leadership support: A senior sponsor who meets champions regularly, removes blockers and makes clear to managers that the role matters.
A way to share what they learn: A regular forum or channel where champions compare notes, so good practice in one team reaches the others.
Recognition: Acknowledgement in appraisals or development plans, so the work counts towards their career rather than against it.
If the role changes someone's duties, hours or pay, take advice on your own situation before formalising it.
Choosing the right people
The most technical person in the room is not always the best choice. Colleagues need to trust and approach a champion, so look for these qualities.
Respected practitioners: People whose day-to-day work others rate, such as a senior bookkeeper or an experienced office manager, carry credibility a newer hire may not.
Genuine curiosity: People who like trying new approaches and are honest about what fails.
Good judgement about risk: People who will say when AI is the wrong tool, as well as when it helps.
Patience with questions: People who can explain things without making colleagues feel slow.
Do not pick only enthusiasts. A thoughtful sceptic who has tested the tools carefully can be a highly credible champion, because colleagues know they will not oversell.
How many and where
Aim for coverage across functions rather than a large group. In a smaller organisation, one champion per team or department is usually enough. Champions should reflect the different kinds of work you do, so a sales champion can speak to proposal writing while an operations champion covers scheduling and supplier management.
When you may not need them
In a very small team where everyone works closely together, a formal scheme may add structure without much benefit. A shared channel and a short regular catch-up can do the same job. Champions earn their keep once AI use spreads across several roles and leaders can no longer see all of it directly.
Supporting champions over time
Check in with champions regularly about workload and what they are seeing. They are often first to notice when training has not landed or a tool is causing problems, which makes them a valuable source of evidence on what makes AI training actually stick.
More questions