Oral Examinations Make a Comeback in US Higher Education to Counter AI Threat
A growing number of American college instructors are reviving oral examinations, an ancient assessment method, to address what they describe as an artificial intelligence crisis in higher education. This shift comes as educators observe students submitting flawless written work generated by AI tools while demonstrating little understanding of the material when questioned directly.
The AI-Generated Assignment Problem
"You won't be able to AI your way through an oral exam," states Chris Schaffer, a biomedical engineering professor at Cornell University who introduced what he calls an "oral defense" last semester. Across US campuses, instructors are noticing troubling patterns: take-home essays and written assignments return in perfect form, but when students must explain their reasoning verbally, they often cannot.
Emily Hammer, an associate professor of Middle Eastern Languages and Cultures at the University of Pennsylvania, now pairs oral exams with written papers in her seminars. "It comes across as if we're trying to prevent cheating," Hammer acknowledges. "That's not why we're doing this. We're doing this because students are actually losing skills, losing cognitive capacity and creativity."
Traditional Methods Meet Modern Challenges
Oral examinations represent a significant departure from standard American undergraduate assessment practices, though they have long been part of European university traditions like England's Oxbridge tutorial system. The COVID-19 pandemic initially spurred interest in oral exams as educators sought alternatives to online assessments vulnerable to cheating. Since ChatGPT's 2022 launch, this interest has intensified dramatically.
Bruce Lenthall, executive director of Penn's Center for Teaching and Learning, describes "a massive shift toward in-person assessments" at the Ivy League institution. Several universities have begun running faculty workshops on implementing oral examinations effectively.
Innovative Approaches: Using AI to Assess AI Understanding
Some educators are taking a "fight fire with fire" approach. At New York University's Stern School of Business, professor Panos Ipeirotis developed an AI-powered oral exam for his AI product management course. Students interact with a voice-cloned chatbot that asks increasingly detailed questions about their final projects.
"We wanted to check: Do you know what your team did? Were you a free rider? Did you outsource everything to AI?" explains Ipeirotis, who designed the tool with ElevenLabs, a company specializing in generative AI voice agents. The system provides clues, criticism, and positive feedback during the examination.
Student Reactions and Educational Benefits
Feedback from students has been mixed. Business major Andrea Liu found the AI chatbot's voice surprisingly human but noted awkward pauses and confusing multiple questions. "It felt kind of awkward to be talking to what was pretty much a blank screen," says the 21-year-old, though she acknowledges the necessity: "There is no perfect world where AI exists and kids are not abusing it."
Despite initial nervousness, many students report appreciating the oral format. Cornell junior Olivia Piserchia, a biomedical engineering major, initially found oral defenses nerve-wracking but came to value the one-on-one interaction. "Having that live check-in holds you accountable," she observes. "It's a lot harder to look people in the eyes and say out loud, 'I don't know this.' And that makes you realize, 'I should study this.'"
Addressing Concerns About Student Anxiety
Skeptics note that oral exams can disadvantage shy or anxious students. Carolyn Aslan, who leads Cornell's oral exam training, suggests that clarifying the format ahead of time and beginning with straightforward questions can help. "Sometimes it's actually good to get that quiet student one-on-one, and you finally get to hear from them," Aslan notes. "Sometimes that is the breakthrough."
At Cornell, examples of oral assessment implementation include a religious studies professor who holds 30-minute "final conversations" instead of traditional exams, and an engineering course where the professor conducts four-minute mock interviews with each of 180 students.
The Broader Educational Implications
Educators across disciplines express concern that students who bypass the mental struggle required for genuine problem-solving may fail to develop skills necessary for advanced coursework and professional careers. Huihui Qi, an engineering professor at the University of California, San Diego, launched a three-year study during the pandemic on scaling oral examinations, with several universities subsequently inviting her to share findings.
Clay Shirky, NYU's vice provost for AI and technology in education, reports increasing faculty requests for office hours, presentations, and cold-calling in class. "Instructors are saying, 'I need to look my students in the eye and ask, "Do you know this material?"'" Shirky explains.
As generative AI becomes increasingly sophisticated, the long-term impact on critical thinking remains uncertain. What is clear is that American higher education is undergoing a significant reassessment of how to measure genuine student learning in an age when written work can be produced with unprecedented ease but potentially limited understanding.
