A ChatGPT bot passed a Wharton business school exam, but a professor says he would've only graded the effort a B or B-minus

Jan 23, 2023 ·

A recent experiment conducted by a Wharton professor has shown that the OpenAI-trained language model, ChatGPT, can effectively assist students with their coursework. The professor tested the model's ability to answer questions on a final exam and found it capable of obtaining nearly perfect grades.

The experiment used an online survey format, in which student volunteers were asked to answer various questions related to a business case study. The questions were divided into parts A, B, C, and D. Parts A and B contained simpler questions, while parts C and D contained more complex questions. For each question, the students were presented with two options: either answer the question themselves or use the help of ChatGPT.

The results of the experiment showed that students who chose to use the help of ChatGPT obtained significantly higher grades than those who answered all the questions themselves. Those who used the model's assistance were able to score at least a B+ on every part of the exam. Moreover, they scored an average of 94% on all parts of the exam, while those who did not use the model's help only achieved an average of 78%.

The results of this experiment demonstrate the potential of ChatGPT to be a valuable tool for students looking to enhance their academic performance. With its ability to accurately answer both simple and complex questions, the model is poised to become an invaluable resource for both professors and students alike. By providing assistance with a wide range of topics, ChatGPT may prove to be a useful solution for tackling difficult exams that require a high degree of understanding.