Designing for Trust: Embedding Moral Responsibility in Generative AI Use in the Classroom
Type
applied research project
Start Date
February 1, 2026
End Date
February 28, 2027
Status
ongoing
Description
This project tackles one of the most pressing yet underexplored challenges in the age of generative AI (GenAI): how to cultivate students’ internal moral responsibility when using tools such as ChatGPT. Current debates around AI in education largely emphasize external controls—policies, detection systems, and assessment redesigns. Our approach is fundamentally different: we focus on students’ personal ethical decision-making, aiming to foster academic integrity and workplace readiness in an AI-driven future.
We begin by investigating why students use GenAI, exploring motivations ranging from curiosity and learning support to efficiency, shortcuts, and potential misconduct. Understanding this spectrum will help us capture not only the benefits but also the risks of AI use when moral responsibility is not explicitly developed.
A distinctive feature of the project is its co-design methodology. Students will be partners, not just subjects, collaborating to build an Ethical Reflection Custom GPT via the OpenAI developer platform. This tool will guide users through prompts and scenarios that encourage self-reflection on fairness, responsibility, and impact before they act on AI-generated content. By embedding reflection into AI use, the tool aims to shift students’ behavior from compliance with external rules to personal accountability.
The project combines research, behavioral innovation, and educational practice. It will deliver: (1) evidence on student motivations and ethical reasoning; (2) a user-friendly Custom GPT prototype; (3) policy and pedagogical recommendations; and (4) capacity-building resources for educators and students.
A UAE–Switzerland partnership underpins this initiative, bringing together the UAE’s rapid AI adoption and ethics-driven national strategy with Switzerland’s strong educational traditions and world-class expertise in pedagogy and ethics. This collaboration ensures both international relevance and local grounding.
Expected outcomes include an open-access ethical prompt library, increased AI literacy and responsibility among students, cross-national policy recommendations, and a hybrid dissemination symposium in the UAE. Ultimately, the project will equip the next generation not just with AI skills, but with the moral compass to use them wisely.
We begin by investigating why students use GenAI, exploring motivations ranging from curiosity and learning support to efficiency, shortcuts, and potential misconduct. Understanding this spectrum will help us capture not only the benefits but also the risks of AI use when moral responsibility is not explicitly developed.
A distinctive feature of the project is its co-design methodology. Students will be partners, not just subjects, collaborating to build an Ethical Reflection Custom GPT via the OpenAI developer platform. This tool will guide users through prompts and scenarios that encourage self-reflection on fairness, responsibility, and impact before they act on AI-generated content. By embedding reflection into AI use, the tool aims to shift students’ behavior from compliance with external rules to personal accountability.
The project combines research, behavioral innovation, and educational practice. It will deliver: (1) evidence on student motivations and ethical reasoning; (2) a user-friendly Custom GPT prototype; (3) policy and pedagogical recommendations; and (4) capacity-building resources for educators and students.
A UAE–Switzerland partnership underpins this initiative, bringing together the UAE’s rapid AI adoption and ethics-driven national strategy with Switzerland’s strong educational traditions and world-class expertise in pedagogy and ethics. This collaboration ensures both international relevance and local grounding.
Expected outcomes include an open-access ethical prompt library, increased AI literacy and responsibility among students, cross-national policy recommendations, and a hybrid dissemination symposium in the UAE. Ultimately, the project will equip the next generation not just with AI skills, but with the moral compass to use them wisely.
Leader contributor(s)
Member contributor(s)
Khan Dr Zeenath Reza
Funder
Method(s)
focus groups; survey
Division(s)