Human-in-the-Loop Control of Objective Drift in LLM-Assisted Computer Science Education

TL;DR

This paper presents a human-in-the-loop control framework for managing objective drift in AI-assisted computer science education, emphasizing stable pedagogical strategies over tool-specific prompts.

Contribution

It introduces a theory-driven, control-based curriculum design that trains students to specify acceptance criteria and manage AI output drift, enhancing robustness across evolving AI tools.

Findings

Sensitivity analysis shows detectable effects for different instructional strategies.

Structured planning improves students' ability to control AI output.

Introducing deliberate drift aids in diagnosing and recovering from specification violations.

Abstract

Large language models (LLMs) are increasingly embedded in computer science education through AI-assisted programming tools, yet such workflows often exhibit objective drift, in which locally plausible outputs diverge from stated task specifications. Existing instructional responses frequently emphasize tool-specific prompting practices, limiting durability as AI platforms evolve. This paper adopts a human-centered stance, treating human-in-the-loop (HITL) control as a stable educational problem rather than a transitional step toward AI autonomy. Drawing on systems engineering and control-theoretic concepts, we frame objectives and world models as operational artifacts that students configure to stabilize AI-assisted work. We propose a pilot undergraduate CS laboratory curriculum that explicitly separates planning from execution and trains students to specify acceptance criteria and architectural constraints prior to code generation. In selected labs, the curriculum also introduces deliberate, concept-aligned drift to support diagnosis and recovery from specification violations. We report a sensitivity power analysis for a three-arm pilot design comparing unstructured AI use, structured planning, and structured planning with injected drift, establishing detectable effect sizes under realistic section-level constraints. The contribution is a theory-driven, methodologically explicit foundation for HITL pedagogy that renders control competencies teachable across evolving AI tools.