Human Factors in AI-Assisted Interpreter Training: A Design Science Approach to Human-Centred Educational Technology

The educational value of artificial intelligence (AI) depends on technological capabilities and human–AI interaction. This is particularly relevant in interpreter training, which involves high cognitive demands, contextual decisions, and time-sensitive activities. This study examines how Human-Centred Artificial Intelligence (HCAI) principles can guide AI-assisted educational environments for interpreter training. Using Design Science Research (DSR), an environment integrating AI capabilities across the training cycle was developed and evaluated. Drawing on literature concerning Computer-Assisted Interpreting, HCAI, human–AI interaction, technology acceptance, and interpreting cognition, five design principles were derived: learner agency and human control, transparency and calibrated trust, protection of cognitive resources, contextualised assistance, and AI as a support for learning. A technology acceptance study involving 42 professional interpreters provided contextual evidence on technology use, adoption barriers, and professional practice, informing the specification of these principles. The principles were operationalised in the artefact, and its interaction design was examined through formative heuristic evaluation. Professional evidence therefore informs the design without establishing learner-centred validation or educational effectiveness. The contribution is a human-centred, professionally informed framework linking literature, contextual evidence, design requirements, and implementation to align AI capabilities with educational objectives and the cognitive and interactional demands of interpreter training through explicit and traceable design decisions.