A new artificial intelligence system called ChatGPT has passed a Wharton exam, according to a new research paper, showing the potential of the controversial chatbot.
Research by Wharton professor Christian Terwiesch found that AI systems “have demonstrated a remarkable ability to automate certain skills of well-paid knowledge workers in general, and knowledge workers in jobs held by MBA graduates in particular, Including analysts, managers and consultants.”
ChatGPT performed “very well” on the final exam in Operations Management, a core course in the Wharton MBA program, giving correct and “excellent” answers in their interpretation.
“ChatGPT3 is very good at revising its answers based on human cues. In other words, where it initially failed to match a question with the correct solution, Chat GPT3 was able to correct itself after receiving appropriate cues from human experts. Considering this performance , Chat GPT3 will score B to B on the exam,” the study concluded.
The language processing system GPT, or Generative Pre-trained Transformer, was developed by OpenAI to provide human-like dialogue through artificial intelligence.
“The conversational format enables ChatGPT to answer follow-up questions, admit mistakes, challenge incorrect premises, and reject inappropriate requests,” reads the system’s brief on OpenAI.
The new tool quickly caught on and raised concerns about its potential use in school settings, where it could, among other tasks, use information gleaned from the internet to write essays and answer complex questions.
Earlier this month, the New York City Department of Education banned the use of ChatGPT on public school networks and devices over concerns that students could use the tool to harm their education.
Terwiesch’s paper suggests that schools should carefully examine the interplay between AI tools and the educational experience, including testing policies and “curriculum design that focuses on human-AI collaboration.”