Princeton Memo Directs Faculty on Acceptable Use of ChatGPT and Other AI

In a Jan. 25 memo to the Princeton faculty, Dean of the College Jill Dolan and Dean of the Graduate School Rodney Priestley presented guidance on the use of Artificial Intelligence (AI) and ChatGPT for teaching. They noted that the University does not plan to ban ChatGPT but encouraged professors to consider the impacts this new technology could have on classes. 

ChatGPT (or Chat Generative Pre-Trained Transformer) is an AI tool that can generate human-like concepts and ideas in text form in response to short prompts. Released in November by the AI lab OpenAI, ChatGPT has presented challenges for teachers across the country who fear students may turn to this technology to complete their classwork. 

According to The New York Times, some public-school systems such as New York City and Seattle have banned the use of ChatGPT, and some universities are shifting back to handwritten papers, group work, and in-class assignments to combat these concerns. Additionally, more than 6,000 teachers have signed up for GPTZero, an app created by Princeton student Edward Tian ’23 to detect AI-generated text. 

“Despite the copious handwringing in the media and elsewhere about this chat bot and its implications for higher education, we and our colleagues in the College and the Graduate School remain confident that our undergraduate liberal arts and graduate education programs will remain vital, vibrant, and useful in the years ahead,” Dolan and Priestley wrote in the memo.

They added that Princeton’s Academic Regulations require students to properly cite sources, and that the undergraduate Honor Code requires students to submit original work. They offered a handful of other suggestions for professors including: to be explicit about the policy on the use of AI/ChatGPT in their classes, to explain the value of original work and grading standards, to design assignments that require critical thinking and asks students to demonstrate their process, and to be clear with students about the benefits and limitations of ChatGPT. 

Dolan and Priestley went on to share two examples of beneficial ways to use the AI software: To engage with class readings by asking ChatGPT questions and assessing its responses, and using the software’s generated essay to compare to drafts written by students as a teaching tool. 

“As we know, ChatGPT won’t be the last AI program that challenges authentic teaching and learning. But this challenge is one we think Princeton faculty and instructors can manage with grace and aplomb,” they wrote.