Leveraging artificial intelligence (AI) for data collection and analysis: the case of and application in corporate social responsibility (CSR) through sport research

Purpose – This paper aims to provide industry insights by utilizing advanced natural language processing (NLP) techniques to extract and analyze how companies listed on the Qatar Stock Exchange (QSE) report their corporate social responsibility (CSR) initiatives related to sport. Furthermore, it compares these findings to mainstream CSR outcomes within the sport management literature, offering a nuanced perspective by leveraging a longitudinal dataset and focusing on industries operating in the Middle East. Design/methodology/approach – This study automates information retrieval by integrating the generative pre-trained transformer (GPT-4), a large language model from OpenAI, with LangChain. We gathered annual reports for all companies listed on the QSE from 2006 to 2022, which spans seven sectors and 46 companies. The reports, available in PDF format, were sourced directly from the companies’ official websites and systematically organized by year to streamline the analysis process. Findings – Between 2006 and 2022, companies listed on the QSE reported a total of 672 “CSR through sport” initiatives. This overall upward trend indicates a growing focus on CSR within the sports sector. Originality/value – This research contributes by introducing a collaborative human–AI framework that provides an innovative method for analyzing how listed companies incorporate “CSR through sport” in their official, publicly accessible communications.