OpenAI’s chatbot product, ChatGPT, has been not too long ago in comparison with college students by way of efficiency on accounting exams. Whereas the scholars outperformed ChatGPT, researchers nonetheless discovered the chatbot’s efficiency to be spectacular. They even went as far as to say that ChatGPT is a sport changer that can revolutionize the best way folks train and study for the higher.
A bunch of researchers from Brigham Younger College (BYU) and 186 different universities carried out a examine to find out how OpenAI’s know-how would carry out on accounting exams. The outcomes of the examine have been revealed within the journal Points in Accounting Schooling.
In response to the examine, college students scored a median of 76.7 % on the accounting examination, whereas ChatGPT achieved a rating of 47.4 %.
The examine revealed that ChatGPT outperformed the coed common in 11.3 % of the questions, with significantly robust efficiency in accounting info methods (AIS) and auditing. Nevertheless, ChatGPT’s efficiency was weaker in tax, monetary, and managerial assessments. This might probably be attributed to the bot’s problem with the mathematical calculations required for a majority of these questions, in response to the researchers.
Furthermore, the examine discovered that ChatGPT, which makes use of machine studying to supply pure language responses, carried out higher on true/false questions (68.7% appropriate) and multiple-choice questions (59.5%), however struggled with short-answer questions (with appropriate solutions starting from 28.7% to 39.1%).
Moreover, the researchers noticed that higher-level questions have been more difficult for ChatGPT to reply, and it typically offered detailed responses for incorrect solutions or responded inconsistently to the identical query. Regardless of steadily offering explanations for its solutions, ChatGPT sometimes selected the flawed multiple-choice response regardless of providing correct descriptions.
It was additionally discovered that ChatGPT offered false info, together with producing a seemingly legitimate reference for non-existent sources. Moreover, it made errors in mathematical calculations comparable to including as an alternative of subtracting or dividing numbers incorrectly.
To contribute to the continuing debate on the position of fashions like ChatGPT in training, David Wooden, a BYU professor of accounting and the lead creator of the examine, enlisted quite a few professors to check ChatGPT’s efficiency with that of college accounting college students.
The social media pitch by his co-author garnered an enormous response, with 327 co-authors from 186 instructional establishments in 14 international locations contributing 25,181 classroom accounting examination questions. Additionally they concerned undergraduate BYU college students who offered ChatGPT with 2,268 questions from numerous accounting domains of various complexity and kinds, together with AIS, auditing, monetary accounting, managerial accounting, and tax.
(With inputs from PTI)