[Paper Review] AI Misuse in Education Is a Measurement Problem: Toward a Learning Visibility Framework

Eduardo Davalos, Yike Zhang|arXiv (Cornell University)|Mar 8, 2026

Artificial Intelligence in Healthcare and Education0 citations

TL;DR

The paper reframes AI misuse in education as a measurement problem and proposes the Learning Visibility Framework to make learning processes observable alongside outcomes for ethical AI integration.

ABSTRACT

The rapid integration of conversational AI systems into educational settings has intensified ethical concerns about academic integrity, fairness, and students' cognitive development. Institutional responses have largely centered on AI detection tools and restrictive policies, yet such approaches have proven unreliable and ethically contentious. This paper reframes AI misuse in education not primarily as a detection problem, but as a measurement problem rooted in the loss of visibility into the learning process. When AI enters the assessment loop, educators often retain access to final outputs but lose valuable insight into how those outputs were produced. Drawing on research in cognitive offloading, learning analytics, and multimodal timeline reconstruction, we propose the Learning Visibility Framework, grounded in three principles: clear specification and modeling of acceptable AI use, recognition of learning processes as assessable evidence alongside outcomes, and the establishment of transparent timelines of student activity. Rather than promoting surveillance, the framework emphasizes transparency and shared evidence as foundations for ethical AI integration in classroom settings. By shifting focus from adversarial detection toward process visibility, this work offers a principled pathway for aligning AI use with educational values while preserving trust and transparency between students and educators

Motivation & Objective

• Clarify why AI misuse in education is more a visibility/measurement issue than a detection problem.
• Propose a framework to make learning processes observable and trackable in AI-enabled assessments.
• Define principles to guide transparent, ethical AI use in classrooms without surveillance-centric approaches.
• Link AI tooling with pedagogical design to preserve trust and learning depth.

Proposed method

• Synthesize evidence from cognitive offloading, learning analytics, and multimodal timeline reconstruction.
• Introduce three core principles (P1-P3) of the Learning Visibility Framework.
• Argue for timeline-based, process-aware assessment and human-centered interpretation of data.
• Discuss design considerations, privacy, and potential future work for implementing visibility-based approaches.
• Relate framework to existing multimodal learning analytics and timeline reconstruction techniques.

Experimental results

Research questions

• RQ1 How can AI use in assessments be clearly specified and modeled to distinguish valid versus invalid use?
• RQ2 How can learning processes and outcomes both serve as evidence in assessment, beyond final products?
• RQ3 How can a transparent timeline of student actions reconstruct learning sequences in AI-assisted environments?
• RQ4 What are the ethical and practical considerations for implementing visibility-based assessment in education?

Key findings

• AI misuse is more effectively addressed by increasing visibility into learning processes rather than relying on detection tools alone.
• Process-based evidence (e.g., revision histories, intermediate steps) can illuminate engagement and reasoning beyond final submissions.
• A transparent timeline of student actions helps contextualize AI use and supports formative dialogue and accountability.
• Privacy, data volume, and potential circumvention are key challenges to be addressed in visibility-centered designs.
• The framework emphasizes transparency, shared evidence, and pedagogical alignment over surveillance or punitive detection.