A parks department is clarifying AI operating-model roles. What should model stewards be responsible for?
Select an answer to reveal the explanation.
Short Explanation
Model stewards aren’t the logo committee. They watch for drift and file change requests when the model starts wandering off the trail.
Full Explanation
In a parks AI operating model, model stewards own ongoing fitness of assigned models: monitoring for drift or performance degradation, investigating anomalies, and filing change requests when retraining, data pipeline fixes, or control updates are needed. Stewardship keeps civic AI from silently decaying as seasons, visitation patterns, or upstream data sources change—whether the model supports irrigation optimization, reservation demand forecasting, or resident-facing assistants.
Making stewards responsible only for designing park logos unrelated to model behavior fails role definition—visual branding is not model lifecycle ownership. Ignoring performance changes until residents complain publicly fails by concept: public complaint is a lagging, harmful detector compared with proactive monitoring. Deleting monitoring so no one can see model drift fails governance outright by destroying the observability stewardship requires.
Exam caveat: stewards may not personally retrain models; their duty is to monitor, escalate, and initiate controlled change through the approved lifecycle—not to bypass change management. Operational check: assign stewards in the inventory, verify monitoring dashboards and alert routes exist, and sample recent drift alerts to confirm change requests were filed and tracked to closure.