Using generative AI in human resource development: an applied research study

Alexandre Ardichvili, Khalil Dirani, Sami Jabarkhail, Walid El Mansour, Sarah Aboulhosn

Research output: Contribution to journalArticlepeer-review

1 Scopus citations


In this article, we report results of an experimental study investigating the use of ChatGPT in the design of an HRD intervention. Our goal was to identify the benefits and drawbacks of using this technology in HRD practice. The theoretical framework for the study was provided by the Activity Theory, an often used approach in Human-Computer Interaction research. The study participants were HRD doctoral students from two large public universities in the USA. The participants used ChatGPT in developing an HRD intervention plan, based on a supplied organisational scenario. Data for the analysis included the intervention plans, prompts used by participants, outputs received from ChatGPT, and participants’ responses to open-ended questions. The study revealed varied participant experiences with ChatGPT, including differences in time investment and interaction rules. Among key findings were the tendency to interact with ChatGPT as if it were a human consultant, and steps to establish division of labour between the tool and the human agent. ChatGPT was valued for its usefulness in initial planning and brainstorming ideas for HRD interventions, yet its limitations were apparent in generic responses and lack of contextual understanding. Across all participants, the crucial role of human expertise in refining ChatGPT’s output was consistently highlighted.

Original languageEnglish (US)
Pages (from-to)388-409
Number of pages22
JournalHuman Resource Development International
Issue number3
StatePublished - 2024

Bibliographical note

Publisher Copyright:
© 2024 Informa UK Limited, trading as Taylor & Francis Group.


  • ChatGPT
  • Generative AI
  • HRD
  • human-AI interaction


Dive into the research topics of 'Using generative AI in human resource development: an applied research study'. Together they form a unique fingerprint.

Cite this