This post examines how two runs of the same Competitive Scenario Planning artifact—using different levels of input detail—produced materially different outputs. By comparing sparse and richer inputs, it shows how SolveBoard’s governed AI artifacts preserve intent, constrain analysis, and make the relationship between inputs and outputs explicit.