The Sustainable Reporting of a Higher Education Institution: A Pilot Project Where AI-Guided Document Analysis Facilitates the Environmental, Social, and Governance Reporting

Abstract

Higher Education Institutions (HEIs) around the world are in the process of incorporating sustainability into their operations. New directives and legislations such as the EU Green Deal, the AI-Act, the Corporate Social Responsibility Directive, and the attention for ESG policies-are examples of top-down approaches. However, there are also bottom-up initiatives emerging. A significant risk here is that these efforts may rely on a few individuals and that sustainability may be perceived as a trend, only gaining focus during the beginning. Therefore, after an enthusiastic start, sustainability should be embedded into the policies governing all activities of the HEI. Besides the inside-out approach, attention should also be given to the outside-in approach. Specifically, what does the external world believe the HEI should focus on regarding sustainability? Transparency around progress towards achieving the Sustainable Development Goals (SDGs) is crucial, and this is best done in accordance with European legislation on Environmental, Social, and Governance (ESG) reporting. Although HEIs in Belgium are not yet required to report on ESG, the University of Applied Science and Arts (PXL UAS) intends to start in an innovative way. Previous research on corporate social responsibility (CSR) investments has established its long-term financial benefits including fostering innovation. Similar 523 524 A. Paesen et al. trends are anticipated for ESG, reflecting the growing importance of sustainable investing. However, to stimulate this trend, one needs adequate solutions that help transform their operations efficiently. The main obstacle for businesses blocking them to follow up on their ESG reporting findings has been identified and indeed lies in the cost of the extensive data gathering process from various sources, which includes the business itself and its supply-and value chain. The developed tool in this paper by ResultX (in a co-creation with all the partners) aims to greatly improve the efficiency and accuracy of the reporting process by utilizing advanced AI techniques, including Natural Language Processing (NLP), Large Language Models (LLMs), Retrieval Augmented Generation (RAG) technology, and more. Implementing these techniques will enable to report findings more easily and save time and money that can be spent on incorporating impactful ESG strategies backed by data, acting upon them, resulting in increased resilience and long-term financial benefits. The PXL UAS deployed this newly developed tool in a pilot project on the one hand to report their achievements within the SDG ambassador track and on the other hand to also check them against the legislation in force.