Overview

This research investigates how students and educators in India are using generative AI tools like ChatGPT in higher education — and how institutions can respond responsibly. Surveys with 71 students and 27 faculty revealed that AI is already a routine academic companion, but one whose use often remains shallow, ethically unclear, and unsupported by policy.

To address these challenges, the paper proposes a four-pillar framework for responsible educational AI:

  1. Pedagogical Alignment – embed AI into coursework that demands reflection, not shortcuts
  2. Transparency and Explainability – teach students to interrogate AI outputs
  3. Student-Centred Co-Design – involve learners in shaping how AI supports them
  4. Institutional Guardrails – establish clear, consistent policies and guidance

Rather than banning or blindly embracing AI, this framework repositions it as a structured partner in learning and a catalyst for redesigning teaching, assessment, and governance.


Research Questions


Methodology