Neurodivergent Expertise in AI-Augmented Knowledge Work: Cognitive Debt, Reflective Scaffolding, and the Redesign Imperative

Artificial intelligence systems are rapidly embedding in knowledge work environments designed for neurotypical cognitive profiles, creating new barriers for neurodivergent professionals while simultaneously offering transformative accommodation possibilities. This article introduces cognitive debt — accumulated costs to sustained attention, learning transfer, and mental health from AI use patterns that bypass effortful cognitive engagement — and argues that autistic and neurodivergent knowledge workers possess critical expertise for understanding and preventing these dynamics. Recent experimental evidence demonstrates that unrestricted generative AI access significantly impairs learning transfer, the capacity central to expert knowledge work. A threshold transition model predicts that autistic and neurodivergent professionals, navigating narrower sensory tolerance windows and atypical precision-weighting profiles, will exhibit cognitive debt indicators at earlier timepoints than neurotypical colleagues — not as vulnerability but as expertise developed through sustained navigation of mismatched cognitive ecologies. The article reframes neurodivergent knowledge as design specification: AI systems built to support autistic and neurodivergent cognitive flourishing — sensory-aware, pacing-respectful, effort-preserving — advance cognitive sustainability for all workers. Three falsifiable predictions and institutional implications are proposed.