The democratization narrative surrounding AI tools emphasizes that 'anyone can build software now,' treating the language interface as a universal equalizer. Benkler's institutional analysis reveals this as a partial truth. The language interface eliminates one barrier (technical training) while preserving another (communicative precision). The capacity to describe complex needs clearly is not evenly distributed. It is cultivated through education, practiced in professional contexts, and reinforced by social environments that reward explicit, structured communication. Working-class speakers, speakers of English as a second language, and speakers from oral cultures often use high-context, implicit communication styles that AI systems — trained predominantly on written, explicit, Western discourse — struggle to interpret accurately.
This inequality is less visible than the coding barrier it replaced, which makes it more insidious. When software creation required programming skills, the exclusion was legible: people knew they couldn't code and accepted that software development was not accessible to them. When software creation requires only description, the exclusion is illegible: people believe they should be able to use the tool, and when they cannot produce acceptable results, they internalize the failure as personal inadequacy rather than recognizing it as a structural inequality in communicative capital. The democratic promise of the language interface is real but incomplete, and the incompleteness is obscured by the rhetoric of universal access.
Benkler's framework suggests that the institutional response is educational: the development of curricula that teach not programming but structured description — how to break complex needs into modular components, how to specify requirements with precision, how to evaluate outputs against intentions. This is a teachable skill, and teaching it would reduce the inequality of articulacy. But it requires recognizing that the language interface is not transparent, that AI-enabled production has skill requirements, and that the distribution of those skills reproduces existing inequalities of educational access and cultural capital unless deliberate institutional interventions equalize them.
The concept is introduced in this simulation as an application of sociolinguistic research on elaborated and restricted codes (Basil Bernstein) to the context of AI-enabled production. Benkler did not address this specific inequality axis in his original framework, but his consistent attention to the distribution of capability and his insistence that technology does not automatically democratize provide the analytical foundation for recognizing that the language interface distributes productive capacity unevenly.
New barrier beneath the removed barrier. Eliminating the need for coding does not eliminate skill requirements; it shifts them to communicative precision, which remains unequally distributed.
Correlation with existing inequalities. The capacity for structured, explicit description correlates with education and class, reproducing social hierarchies in the ostensibly democratized domain of software production.
Invisible exclusion. When the barrier is articulacy rather than technical training, failures are internalized as personal inadequacy rather than recognized as structural inequality.
Educational remedy available. Structured description is teachable, and teaching it would reduce inequality — but only if educational institutions recognize the need and invest in curricula that cultivate communicative precision.