Artificial intelligence for intimate apparel fit and comfort from technical evidence to business governance

Artificial intelligence (AI) for intimate apparel supports body measurement, size and product recommendation, garment simulation, and virtual try-on, but these tasks provide different evidence about fit and comfort. This critical narrative review examines direct bra research alongside adjacent apparel technologies and identifies the validation needed for business use. Recent direct studies include machine-learning models of breast morphology, bra recommendation prototypes, finite-element contact models, dynamic scanning, pressure measurements, and longitudinal wearer research. They establish meaningful component-level progress, while leaving uncertainty about transfer across products and populations and about the causal effects of AI-assisted decisions on customer and financial outcomes. Two-dimensional image quality, three-dimensional geometric agreement, physical support, perceived comfort, and business impact are therefore evaluated separately. An operational transfer test distinguishes methodological relevance from evidence of successful domain transfer. Building on the evidence gaps, the review proposes task-specific evaluation priorities, safeguards for intimate body data, and a staged research agenda with explicit resource and decision requirements. These proposals are analytical contributions, not empirically validated maturity models or deployment guarantees. The central implication is that businesses should match each proposed claim to its measured outcome and scale investment only after the relevant technical, wearer, and operational evidence is established.