Authority over assistance boundaries is highly centralized in system logic in LLM-based programming support systems¶

Claim · All claims
Evidence · 1 study · review r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? In the reviewed corpus, 69 of 90 studies (76.7%) were coded as System only, with learner-side runtime authority in 9 studies (10.0%) and instructor-side in 11 studies (12.2%). → Minsun Kim 2026

Evidence¶

Minsun Kim 2026¶

Minsun Kim, S. Moonwara A. Monisha, Zihan Wu, David H. Smith IV. (2026). Exploring the Design Space of LLM-Based Programming Support in CS Education: A Scoping Review through the Lens of Assistance Governance. https://arxiv.org/abs/2607.21257 doi:10.48550/arxiv.2607.21257

q2 · i? · review · r2

Qualitative synthesis of the 90-system corpus coding Authority into four categories (system only, system + learner, system + instructor, system + learner + instructor), counting runtime features such as "selecting hint levels, enabling or disabling solution access". Authority was the most concentrated PEA dimension.

"Specifically, 69 studies (76.7%) were coded as System only. Systems with instructor-side runtime authority appeared in 11 studies (12.2%), and systems with learner-side runtime authority appeared in 9 studies (10.0%)."

Discussion¶

Related Claims¶

• Prompt-Based Constraint is the dominant enforcement mechanism in LLM-based programming support systems — related
• Policy in reviewed systems is commonly expressed as a set of assistance boundaries, concentrated on response constraints, scaffolding, and personalization — related
• Policy and enforcement counts were not strongly associated across systems — related
• The reviewed corpus mostly targets introductory learners in short-term studies, with supervision level structurally tied to deployment context — related
• Policy–enforcement pairings are dominated by prompt-based constraints, and policy themes differ in enforcement breadth — related
• Teachers retain ultimate control over AI systems because professional responsibility cannot be delegated — reports the opposite