The work is part of a new, unfolding series, “give me space”, that explores how AI video models construct public life from sparse, ambiguous visual cues. Across vast blue fields and minimal architecture, tiny figures work, shop, wait, cross, gather, fall, dive, and return. The works draw on retro-modern visual languages while exposing the model’s native gestures: its tendency to infer paths from lines, give miniature figures as small as marks legible social roles, distribute varied movement across those figures, and manifest multiple incompatible spatial roles in a single visual field. Any anomalies are absorbed back into routine: after the flood, outage, evacuation, loss, accident, or repair, life is expected to resume. Systems persist. Sound by Barbara Chira with the assistance of AI.