Learning Context: A Unified Framework and Roadmap for Context-Aware AI in Education

TL;DR

This paper proposes a comprehensive framework for making AI in education context-aware by encoding cognitive, affective, and sociocultural factors, enabling personalized, long-term learning experiences at scale.

Contribution

It introduces the Learning Context framework and Model Context Protocol, operationalizing context-awareness in AI educational tools through a scalable, privacy-preserving implementation.

Findings

Implementation within OpenStax platform supports millions of learners.

Enables long-term personalization through durable context encoding.

Ensures privacy and ethical standards in educational AI deployment.

Abstract

We introduce a unified Learning Context (LC) framework designed to transition AI-based education from context-blind mimicry to a principled, holistic understanding of the learner. This white paper provides a multidisciplinary roadmap for making teaching and learning systems context-aware by encoding cognitive, affective, and sociocultural factors over the short, medium, and long term. To realize this vision, we outline concrete steps to operationalize LC theory into an interoperable computational data structure. By leveraging the Model Context Protocol (MCP), we will enable a wide range of AI tools to "warm-start" with durable context and achieve continual, long-term personalization. Finally, we detail our particular LC implementation strategy through the OpenStax digital learning platform ecosystem and SafeInsights R&D infrastructure. Using OpenStax's national reach, we are embedding the LC into authentic educational settings to support millions of learners. All research and pedagogical interventions are conducted within SafeInsights' privacy-preserving data enclaves, ensuring a privacy-first implementation that maintains high ethical standards while reducing equity gaps nationwide.