Online Archive of University of Virginia Scholarship
Leveraging LLMs for Enhanced Sustainability Reporting: An Application for Analyzing, Comparing, and Visualizing ESG Reports; Automated Interpretation and the Politics of Transparency: How LLM-Generated Summaries Shape the Meaning and Accountability of ESG Disclosures9 views
Author
Magee, Fiona, School of Engineering and Applied Science, University of Virginia
Advisors
JACQUES, RICHARD, EN-Engineering and Society, University of Virginia
Abstract
Corporate sustainability reporting has become increasingly complex, with institutions like Deutsche Bank releasing lengthy disclosures that are difficult for stakeholders to analyze efficiently. To address the need for a faster, more transparent method of extracting insights from these reports, I developed a web-based application that leverages large language models (LLMs) to summarize disclosures, identify trends over time, and enable comparisons with peer
institutions. Users can interact with the data through text-based summaries, visual analytics, and an integrated query tool, reducing manual review effort and making ESG reporting more accessible. The prototype demonstrates how AI-driven analysis can enhance clarity, improve decision-making, and lay the groundwork for scalable, cross-company benchmarking in sustainability practices.
Degree
BS (Bachelor of Science)
Language
English
Rights
All rights reserved by the author (no additional license for public reuse)
Magee, Fiona. Leveraging LLMs for Enhanced Sustainability Reporting: An Application for Analyzing, Comparing, and Visualizing ESG Reports; Automated Interpretation and the Politics of Transparency: How LLM-Generated Summaries Shape the Meaning and Accountability of ESG Disclosures. University of Virginia, School of Engineering and Applied Science, BS (Bachelor of Science), 2026-05-08, https://doi.org/10.18130/jcxh-2w78.
