The Taxonomy Wall: Why Semantic Expansion Requires Structural Redesign

Expanding an AI system's vocabulary isn't simply about adding words—it requires proper semantic domain organization. When five integration attempts failed with identical patterns, the developer discovered the bottleneck wasn't fine-tuning but a fundamental architectural issue: the domain assignment logic was routing semantically distinct concepts into an overloaded catch-all category. This reveals a critical principle: vocabulary growth in complex AI systems depends on solid foundational taxonomy design, not incremental optimization.