AI chatbot ChatGPT Not Yet a Match for Humans in Accounting, Study Finds
A recent study conducted by researchers at Brigham Young University and 186 other universities reviewed the potential of OpenAI’s AI chatbot, ChatGPT, in accounting exams. While students scored an overall average of 76.7%, ChatGPT scored 47.4%, proving that it still requires improvement in the realm of accounting.
The study found that ChatGPT struggled with mathematical processes required in tax, financial and managerial assessments and had difficulty answering short-answer questions. Higher-order questions were also harder for ChatGPT to answer.
Additionally, ChatGPT generated incorrect or nonsensical answers and sometimes fabricated fake references. Nevertheless, the researchers anticipate that the upcoming GPT-4 will improve significantly, especially on the accounting questions posed in their study.
Despite its impressive performance in other fields, ChatGPT is not yet perfect and using it exclusively is not recommended according to Jessica Wood, a freshman at Brigham Young University.
