News

Panel Dives Into Artificial Intelligence Usage in Classes

Using Artificial Intelligence (AI) in the classroom is an issue that has grown increasingly relevant, with over 56 percent of college students in the United States reportedly using AI to help complete assignments, according to a college survey. Last Friday, students and faculty gathered in Paresky Commons for a panel discussion regarding Andover’s stance on the use of AI in classes. 

 

Hosted by the Dean of Studies Office, the panel included students Sarah Pan ’24, Reed Manning ’25, and Frank Hu ’26, and faculty included Caroline Odden, Dean of Studies, Nicholas Zufelt, Instructor in Computer Science, and David Fox, Instructor in English. Altogether, the panelists  answered a plethora of questions concerning current AI use, general attitudes towards AI at Andover, and ways that AI could effectively be used for education in the future. 

 

Pan, who recently gave a CaMD scholar presentation on AI alignment, spoke about Andover’s current AI policies. She expressed hopes for further experimentation with AI as an educational tool. 

 

“Personally, my experiences with AI in the classroom have been limited. Sometimes when I’m scrolling through the syllabus of the class, it will say [that] the AI policy [is] that AI is explicitly prohibited unless said otherwise by the teacher… I feel that AI in education is a much bigger thing than just coming up with some sort of experimental policy that may or may not work. It’s definitely a work in progress that has to be innovated on, not just by people here, but [also by] people around the world because it’s such a transformative thing,” said Pan. 

 

Zufelt commented on the possibility of amending the AI policy on campus to be more open-minded. Zufelt discussed how, when his students show him their own experiments with AI, it pushes him further as an educator. 

 

“I do think that one policy shift that would be a culture shift would be if there was a little asterisk on our AI policy (which [currently] says [no AI] unless you’re told otherwise) that said, ‘But, if you think you have a cool idea go talk to your teacher about it!’… That to me, seems like a bit more of a catalyst because I’ve had a couple of times where students have come to me and said ‘Check out this cool thing I did!’… and I think that pushes me [into the future] faster,” said Zufelt. 

 

Manning mentioned how AI could act as a tool to boost student learning as opposed to being a source of academic dishonesty. Manning pointed out examples of artificial intelligence being used without violating any academic integrity standards, highlighting student-teacher transparency as part of that process.

 

“Going to some sort of AI and asking what counter-arguments there are and using that to strengthen [a] paper is a good way to become a better writer. I think there are also really good studying opportunities… Getting example problems over and over that can be checked and corrected without using any extra hours for any teacher could be really good. There are a lot of ways that learning could be really enhanced without necessarily sacrificing some form of academic integrity as long as there is communication between the student and the teacher,” said Manning. 

 

However, Odden highlighted how the learning taking place at Andover is more complex than just delivering results. She noted the presence of students’ unique personal perspectives in their work, something that AI can’t replace in the classroom.

 

“Part of being in an English class is thinking about your own identity and experiences that others don’t have. Part of being in a history class is thinking about what came before you and where we are now and where we might be going. Part of being in a science class is that you learn how the world works. There is a lot that we are doing here that is identity formation and trying to help you figure out who you are. We might be able to find ways that AI will enhance that, but we need to remember it’s not about becoming producers, and sometimes I think that the way we think about AI and the way it will replace [doing] work [will be] a little bit different from the way AI might replace what we need to learn in school,” said Odden.