From transcript to insights: Summarizing safety culture interviews with LLMs

MATEC Web of Conferences, vol. 413, pp. 03003

Abstract

The latest developments in AI have the potential to significantly support qualitative analysis of interview transcripts. This study explores the utility of the OpenAI o1-model to assist in efficiently obtaining reliable summarizations of safety culture interviews. Analysis shows that the current approach has the clear potential to significantly improve efficiency of interviewers by providing a concise report that summarizes multiple interviews according to a pre-defined format. However, some hallucinations are present in the generated report. Additional work will aim at reducing their presence, but such hallucinations also emphasizes that LLMs should primarily assist, rather than replace, interviewers in creating a definitive report.

Authors 4

1. Wouter Martinus Petrus Steijn

   Netherlands Organisation for Applied Scientific Research

   Affiliation as printed

   TNO, Work Health Technology, Sylviusweg 71 2333BE Leiden, The Netherlands

2. Janneke van de Loo

   Netherlands Organisation for Applied Scientific Research

   Affiliation as printed

   TNO, Work Health Technology, Sylviusweg 71 2333BE Leiden, The Netherlands

3. Dolf van der Beek

   Netherlands Organisation for Applied Scientific Research

   Affiliation as printed

   TNO, Work Health Technology, Sylviusweg 71 2333BE Leiden, The Netherlands

4. Jop Groeneweg Aachen

   Leiden University · Netherlands Organisation for Applied Scientific Research · Delft University of Technology

   Affiliation as printed

   Leiden University, Social and Behavioural Sciences, Wassenaarseweg 52 2333 AK Leiden, The Netherlands

   TNO, Work Health Technology, Sylviusweg 71 2333BE Leiden, The Netherlands

   TU Delft, Faculty of Technology, Policy and Management, Jaffalaan 5 2628 BX Delft, The Netherlands