Pitt Professors choose how to use AI in class, causing mixed reactions

During the first weeks of the semester, Pitt students and faculty have contended with varying rules and ethical standards surrounding the use of AI. 

As AI and large language models rapidly evolve, higher education has adapted, with universities across the country setting individual ethical frameworks. A survey conducted by the Digital Education Council found that 86% of students use AI in their studies, with 54% using it weekly.

At Pitt, professors can choose how to integrate or restrict AI in the classroom. 

The University Center for Teaching and Learning provides suggested syllabus language for professors who want to encourage AI use, limit AI to specific assignments or restrict it entirely. On campus, organizations such as the Hub for AI and Data Science Leadership and the Learning Sciences and Emergent Technologies Hub also provide resources and host events for community members to learn how to use AI ethically and effectively.

John Radzilowicz, director of pedagogy, practice and assessment for the Teaching Center, advises faculty to make informed decisions about how AI can be useful in individual classes. 

“Students worry sometimes they get mixed signals,” Radzilowicz said. “You go to one class and a faculty member says you’re going to be able to use AI, and here are the rules, and another class says if you use AI, you’re going to fail because it’s an academic integrity violation.”

He also encourages faculty to consult their students and co-create AI policies that work for individual classes and account for how AI is being used in their respective disciplines.

“There are great opportunities here if we are careful and thoughtful about how we use AI,” Radzilowicz said. 

Radzilowicz recommends a “conductor model,” in which students prepare for the modern world by learning to thoughtfully “conduct” AI — similar to how a conductor guides an orchestra — rather than passively using it to replace their work. 

Overall, Radzilowicz believes that AI guidelines should serve the students, whether that be teaching effective AI use or restricting it to build foundational skills such as reading and writing.

“What I say to faculty is, ‘You have a responsibility to the University, you have a responsibility to your discipline — but you have an obligation to your students,’” Radzilowicz said. 

Jeff Aziz, an assistant dean in the Dietrich School of Arts and Sciences, an academic integrity officer for undergraduates and English faculty member, said different disciplines have distinct concerns about AI use that need to be addressed individually. The Academic Integrity Code was amended in 2024 to specifically include language about AI, which Aziz said remains general to accommodate different disciplines.

“When you get too specific, then you’re dictating to programs what they should promulgate,” Aziz said.

Aziz said professors should learn about AI to understand how it can affect academics.

“I have become conversant in generative AI because I feel that someone in a job like my own should know very well how it works,” Aziz said. “I felt that I needed to be AI literate in order to be a good academic integrity dean.”

Aziz said he does not believe classrooms in the future should have AI integrated into the curriculum, but he acknowledges that the technology is here to stay. Although AI is allowed in some classrooms, he said ethical concerns arise when it is used dishonestly. 

“There is no AI detection software that we consider reliable enough to use for academic integrity purposes,” Aziz said. “We solve these problems the old fashioned way.”

Maeve Underwood, a junior social work student, said she believes AI should be restricted by professors in the classroom to uphold academic standards.

“I really like one professor who asked us how we were already using [AI] and then walked us through the ethical guidelines,” Underwood said.

Underwood is also concerned about the environmental and social effects of AI, including data centers displacing and worsening environmental conditions in disadvantaged communities.

Anna Velasco, a sophomore political science student, said she has seen an increase this year in professors allowing students to use AI in select circumstances, though most of her professors restrict it.

“Pretty much every single one of my syllabi have said ‘Don’t use AI,’” Velasco said. “I’ve had a couple of professors that are like ‘I know you’re going to use it, so here’s what I hope you will do.’”

While some professors advise students to use AI to brainstorm, Valesco believes doing so can displace any creative work that a student can do.

“I can’t think of a time using AI where there isn’t any negative effects or risks,” Velasco said.

Grace Riggle, a junior psychology student, uses the AI model Claude to make application-based questions to study.

“If I need it in a pinch I’ll use it, but I try not to,” Riggle said. “It seems like [professors] are weary about AI, and they just want you to use it in the right way.”