Depth of field (DOF) is the range of distance in a scene that appears acceptably sharp in the rendered image. A shallow depth of field means only a narrow slice is sharp and everything in front of and behind that slice is blurred; a deep depth of field means everything from foreground to background is sharp.
In architectural rendering, deep depth of field is usually the default — you want the room to be in focus from foreground to back wall. Shallow depth of field is used selectively to draw attention to a single object (a chair, a vase, a person), often in lifestyle-coded renders or close-up product shots. Architectural marketing renders almost never use extreme shallow DOF because it makes the architecture itself harder to read.
AI rendering replicates DOF accurately when prompted. Phrases like 'deep depth of field' or 'everything in focus' produce the architectural-standard look; 'shallow depth of field on the foreground chair' or 'background slightly out of focus' produces the lifestyle look. The trick is being explicit about what should be sharp — without direction, the AI defaults vary by training distribution.
See also