The Applicability of Artificial Intelligence Applications in Accounting: The Case of ChatGPT and Claude

Abstract

Digitalization and the use of artificial intelligence are rapidly increasing in every field today. The
accounting profession, which was carried out using traditional methods for many years, has also
transitioned to digitalization with the development of the internet. Manual ledgers have been
replaced by computers with the advancement of technology. The aim of this study is to evaluate
the accounting knowledge level of ChatGPT, developed by OpenAI, and Claude, developed by
Antropic, and to examine the accuracy of their answers. Questions prepared in the fields of
General Accounting, Management Accounting, and Strategic Management Accounting were
asked to both ChatGPT and Claude, and the accuracy of the answers obtained was examined and
compared. When their answers were examined, ChatGPT answered all general accounting
questions correctly except for the verbal question, while Claude answered all general accounting
questions correctly. In the Cost Accounting questions, it is not possible to say anything definitive
for both applications; both answered one of the five questions incorrectly. As a result of their
answers to the strategic management accounting questions, it is observed that ChatGPT is
insufficient, and users may prefer Claude.