A prompt asks an AI assistant to "create the flow metrics" for a library consortium digital-services ART that has never collected flow time, WIP, or throughput. What should the RTE do instead?
Select an answer to reveal the explanation.
Short Explanation
If the train never timed a Feature, the chatbot cannot magically know flow time. Made-up metrics are fan fiction with decimals. Ground the prompt in numbers you actually have, or go collect them.
Full Explanation
Prompts for measurement work must be grounded in data the ART has collected. Generative models will fabricate plausible statistics if asked to create metrics, and those figures must not become baselines or public dashboards. Copying another city's unpublished numbers is equally ungrounded. The RTE either uses real measurements or starts collecting them.