The More Knowledgeable Other

In the AI Story

The MKO's essential function is not the transmission of information but the creation of a developmental relationship. A good teacher notices what the student is avoiding, not just what she asks about. A good teacher senses when apparent mastery conceals deeper confusion. These perceptions are embodied and affect-laden — they depend on the kind of understanding that comes from being a creature who has itself undergone development. The cultural-historical tradition has insisted, since Vygotsky, on the unity of cognition and affect in development: the child does not simply learn to count; she learns to count in the context of a relationship whose emotional quality is constitutive of the learning.

AI performs one dimension of MKO function — cognitive scaffolding — with extraordinary capability. Contemporary systems read user intent, calibrate response sophistication, adjust explanations to apparent comprehension level. What they cannot perform is the social-emotional dimension: recognizing when a learner is ready to be challenged versus when she needs consolidation, creating conditions of trust and emotional safety, sensing the identity-level implications of new capability. The Trivandrum training succeeded not because Claude replaced the human relationships in the room but because Claude handled cognitive scaffolding while the leader's presence, the engineers' relationships with each other, and the shared context of discovery handled the social and emotional scaffolding the transformation also required.

A distinctive feature of AI as MKO is the unbounded asymmetry between learner and system. Traditional MKOs are more capable than learners but remain within the same category of cognitive agent; the student can imagine becoming the teacher. AI systems operate across domains with breadth no individual human can match, producing an asymmetry the learner cannot traverse by becoming like the system. This unbounded asymmetry disrupts the motivational structure Vygotsky's framework assumed — the aspiration toward a reachable level — and creates the specific risk The Orange Pill names: the temptation to outsource cognitive work because the tool will produce plausible outputs regardless of whether the learner engages.

The cross-domain reach of AI MKOs offers something no human MKO can provide: scaffolding across the boundaries of specialization. A single Claude conversation can span statistics, rhetoric, neuroscience, and systems architecture in ways that exceed any human teacher's range. This creates a new kind of zone — not deeper into a single domain but across domains — whose developmental implications remain genuinely open.

Origin

The MKO concept is implicit throughout Vygotsky's writing but rarely named explicitly. Barbara Rogoff's work in the 1980s and 1990s, especially Apprenticeship in Thinking (1990), made the MKO central to cultural-historical pedagogy. The application to AI emerged as soon as large language models became capable of sustained dialogue (roughly 2022–2023), with researchers asking whether systems without consciousness could nonetheless serve MKO functions.

The framework's most fruitful contemporary extension asks not whether AI is an MKO but what kind of MKO it is, and what forms of scaffolding it can and cannot provide.

Key Ideas

Function over substance. What makes an MKO is not what it knows but what it does — calibrate, respond, withdraw — in service of the learner's development.

Cognitive and social-emotional dimensions. The MKO operates in both domains; AI handles the first with unprecedented capability and cannot, as currently constituted, perform the second.

Unbounded asymmetry. The gap between learner and AI MKO cannot be traversed through normal development, which changes the motivational structure the traditional framework assumed.

Cross-domain reach. AI MKOs enable scaffolding across disciplinary boundaries no human teacher can span — a new developmental opportunity and a new failure mode if the learner cannot hold the fragments together.

Complement, not replacement. The developmental optimum combines AI cognitive scaffolding with human social-emotional scaffolding; neither alone suffices for whole-person development.

Debates & Critiques

A central debate concerns whether genuine intersubjectivity is required for MKO function. If yes, AI cannot be a genuine MKO and only simulates the role; any scaffolding it provides lacks the relational substance development requires. If no — if the functional properties of the interaction are what matters — then AI can serve as a genuine MKO when properly deployed. The cultural-historical tradition tends toward the first answer for social-emotional development and the second for cognitive scaffolding, suggesting a division of labor rather than a binary choice.