Practical guide
AI fluency for OD managers: clarify decision ownership
Map a fictional cross-team handoff to expose conflicting decision rights before proposing a training or organizational change.
Organizational development managers can practice AI fluency by examining how a work system distributes decisions. This fictional cross-team exercise concerns an unclear approval handoff. It creates an ownership map and a bounded change proposal, not a diagnosis of employee attitudes or a recommendation to reorganize real people.
Map the decision rather than drawing an organization chart
An ownership map should show who can make the decision that blocks the work.
The fictional design team prepares an internal announcement and believes operations approves the release. Operations believes it checks logistics while the programme owner approves release. The programme owner expects design to send routine announcements independently. The packet supplies these conflicting descriptions without resolving them.
Ask AI to extract each stated responsibility and keep the contradictions visible. A neat chart with one approver chosen by the assistant would invent authority. Record the exact decision: permission to release this category of announcement. Separate preparation, factual review and release approval instead of giving all three the same label.
Distinguish the work-system gap from a learning gap
Repeated waiting does not by itself establish that people need communication training.
In the case, an announcement waits while each team assumes another owns approval. That behavior is consistent with conflicting role expectations. The packet does not establish unwillingness, poor teamwork or a lack of skill. Document the handoff mismatch before proposing an intervention.
Ask what evidence would show the problem persists after responsibilities are clarified. This creates a way to distinguish an unclear process from a different constraint. AI can propose hypotheses, but do not let it diagnose organizational culture from one example or describe a proposed cause as an observed fact.
Draft a responsibility agreement for an owner to approve
A proposed map becomes useful when its decisions and exceptions can be challenged.
For discussion, propose that design prepares, operations checks logistical facts and the programme owner authorizes release. Mark the arrangement as a proposal. Include the type of announcement it covers and which exceptions require a different route. Do not present the proposal as current policy.
Add what each handoff must contain: version, requested decision, unresolved facts and required timing. If the approver is unavailable, the map should identify an unresolved backup question rather than automatically transferring authority. A faster process is not a sufficient reason to invent permission.
Walk a difficult case through the proposed map
A diagram that works only for routine cases may hide its most important gap.
Introduce an announcement with a late logistics change after approval. Ask who checks the changed fact, whether the earlier approval still applies and which version may be sent. The packet does not settle those decisions; the walkthrough should surface them for the owner.
Now introduce an urgent request while the programme owner is absent. Record the missing escalation route. Do not interpret urgency as permission for any available person to release the announcement. The purpose is to identify specific operating decisions needed before the proposed arrangement can be used.
Evaluate the handoff before claiming organizational improvement
A small trial can check whether the new instructions are understandable.
Deliver the original responsibility conflict, proposed map, two exception walkthroughs and open owner decisions. A later trial could inspect whether participants identify the same approval route and whether the necessary information reaches it. Describe that as a proposed test, not an achieved improvement in culture or productivity.
Review whether the worker retained uncertainty about authority and separated the proposed process from current practice. The HR partner guide focuses on a manager conversation about reported barriers. This exercise focuses on cross-team decision rights and whether a work-system change addresses the particular handoff failure.
Sources and scope
NIST: AI risk management. Skills England: workplace AI foundations.
These references provide background, not validation or endorsement of this exercise. The case details, calculations and suggested review questions are original instructional material. Use them to discuss observable work, not to infer customer outcomes, professional credentials or performance in every setting. Before adapting the exercise, confirm the relevant facts, approved tools, data permissions and decision owners. If you change the case, revisit the expected answers and checks as well. These examples describe practice tasks, not a promise that a particular product includes the fictional features.
Sources: [1] [2]
Sources
- 1.AI RMF Core · NIST
- 2.AI foundation skills for work benchmark · Skills England