Abstract:

As artificial intelligence (AI) becomes increasingly accessible in educational settings, traditional mathematical tasks risk becoming obsolete or misaligned with meaningful learning objectives. This "misalignment" can be viewed from different perspectives, such as the potential automation of traditional tasks, the shift in cognitive demands, disconnection from real-world applications, and new assessment challenges for conceptual understanding and strategic thinking. In this sense, mathematical tasks can be shifted from a procedural to a conceptual understanding, from drill to real-world modeling, from isolated problems to interdisciplinary work, from memorization to critical evaluation, and even from closed to open-ended inquiry, all supported by teacher guidance and AI tools in learning environments.

This article introduces a practical methodology for re-imagining mathematics task design through a "Human-in-the-Loop, AI-Augmented" lens. Centered around the AEDR (Anticipate, Elevate, Diversify, and Reflect) framework, this approach guides educators in leveraging AI not as a replacement for student thinking but as a catalyst for deeper conceptual understanding, critical reasoning, creativity, and collaboration. Mathematical examples serve to illustrate how task design can be adapted to challenge both AI systems and human learners in complementary ways. The article positions educators as essential curators and facilitators in this evolving landscape, ensuring that AI serves pedagogy rather than dictating it.

Keywords:

Mathematics Education, Task Design, Large Language Models, Conceptual Understanding, Human-in-the-Loop, AI-Augmented Learning, AEDR Framework, Critical Thinking, Pedagogical Innovation.