Designing for Difference - Intersectionality and Adaptive AI Mentorship in STEM Pathways: Operationalising Social Realities Through Personality and Identity-Aware Systems

Abstract

This paper presents the development and evaluation of an AI mentoring system designed with intersectionality at its core. Using HEXACO-based personality profiling, demographic mapping, and mentoring preference models, the system dynamically generates three distinct mentor personas for each user. We analyse how different user identities—across gender, class, race, and discipline—affect their mentor choices and interactions. By operationalising intersectionality into AI architecture, we challenge universalist AI approaches and offer an adaptive model that acknowledges and responds to users’ lived realities. The paper contributes a framework for inclusive AI design that centres on difference, not abstraction.

Presenters

Timipado Imomotebegha
Student, PhD Applied Human Computer Interaction, Loughborough University, United Kingdom

Details

Presentation Type

Paper Presentation in a Themed Session

Theme

Social Realities

KEYWORDS

AI MENTORSHIP, INTERSECTIONALITY, PERSONALITY PROFILING, ADAPTIVE SYSTEMS