Can ChatGPT improve the readability of financial reporting by the public sector entities? The case of Lisbon municipality
The readability of financial reporting from the users’ perspective is an essential element to ensure its usefulness and, therefore, its relevance. Through the lens of Cognitive Load Theory and plain language principles, this quasi-experimental research aims to assess whether an artificial intelligen...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-12-01
|
| Series: | Cogent Social Sciences |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/23311886.2025.2524004 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The readability of financial reporting from the users’ perspective is an essential element to ensure its usefulness and, therefore, its relevance. Through the lens of Cognitive Load Theory and plain language principles, this quasi-experimental research aims to assess whether an artificial intelligence tool, specifically ChatGPT in its GPT-4 model, can be effective in enhancing the readability of the Lisbon municipality’s annual reports and accounts for the year ending in 2023. For this purpose, six surveys, each one including two versions of six different excerpts collected from that source (an original version and a modified one), were distributed among twelve (one survey by two of them) experts on public sector financial matters. The findings globally indicate higher readability levels of the modified version of the excerpts, that is, the one produced by ChatGPT from the original version and following a prompt proposed by researchers. Respondents argued that benefits mostly relate to the text’s form and content. Although exploratory, this study may contribute to the literature by highlighting the potential benefits of using ChatGPT to enhance the readability and overall usefulness of public financial reports for a broader audience, thereby benefiting various stakeholders, including preparers, users, regulators, policymakers, and auditors. |
|---|---|
| ISSN: | 2331-1886 |