AI ClaimsRapid review

Stewart Brand · 1994

How Buildings Learn

What Happens After They're Built

Cover via Open Library

Rough AI truth score

83/100

Brand's central insight—that buildings are changed continuously by occupants, maintenance, markets, and technologies—has become foundational to adaptability research. His layered model is a valuable heuristic, though its time scales and preference for low-road buildings are not universal performance laws.

Based on three central claims · medium-high confidence

The three claims

01Supported

Buildings continue to change after completion as occupants, technologies, regulations, maintenance, and uses evolve.

Building-adaptability research treats change over time as a core design and lifecycle problem, documenting conversion, replacement, maintenance, and shifting user needs. Brand's illustrated cases capture a widespread process rather than a rare exception.

02Mostly supported

Building components change at different rates, so separating site, structure, skin, services, space plan, and stuff can enable adaptation.

Later literature repeatedly uses layered models to reason about service life, replacement, and design for change. Boundaries and time scales overlap, assemblies interact, and regulation or finance may dominate technical separability, making the six layers a heuristic rather than a fixed ontology.

03Mostly supported

Simple, inexpensive, loosely specified low-road buildings reliably adapt better than prestigious architect-designed buildings.

Slack space, accessible services, simple assemblies, and tolerant ownership can make change cheaper, and Brand supplies persuasive cases. High-road buildings can also adapt through stewardship and resources, while cheap construction can be brittle, unsafe, inefficient, or demolished early.

Other claims worth checking
  • Maintenance deserves more design attention than novelty.
  • Post-occupancy change is a better test than publication photographs.

What this number means. It is an AI-generated first-pass judgment of three central factual or causal claims—not a rating, exhaustive fact-check, or human peer review. Claim credits are 100% for supported, 75% for mostly supported, 50% for mixed, and 25% for weak, then averaged and rounded. Lower confidence means the score should move more as better evidence arrives.

Method three-central-claims/0.1.0 · checked 2026-09-01 · 3/3 selected claims assessed · method and source audit