Expanding Observability via Human-Machine Cooperation

We ask how to use machine learning expand observability, which presently depends on human that informs conceivability. The issue is engaged by considering the question of correspondence between conceived observability counterfactuals and observable, yet so far unobserved or unconceived, states affairs. A possible answer lies in importing out reference frame content could provide means for conceiving further counterfactuals. They allow us define high-fidelity increasing level question. To achieve we propose generative models as providers content. From an applied point view, such a role shows emerging dimension human-machine cooperation.