Could ChatGPT get an Engineering Degree? Evaluating Higher Education Vulnerability to AI Assistants

TL;DR

This paper assesses the vulnerability of higher education assessments to AI assistants by measuring GPT-3.5 and GPT-4's ability to answer university-level STEM questions, revealing significant risks to current evaluation methods.

Contribution

It provides a novel dataset of assessment questions and evaluates AI performance, highlighting the need to revise assessment strategies in higher education due to AI capabilities.

Findings

GPT-4 answers 65.8% of questions correctly

AI can pass core assessments in various degree programs

Assessment design must be revised to address AI capabilities

Abstract

AI assistants are being increasingly used by students enrolled in higher education institutions. While these tools provide opportunities for improved teaching and education, they also pose significant challenges for assessment and learning outcomes. We conceptualize these challenges through the lens of vulnerability, the potential for university assessments and learning outcomes to be impacted by student use of generative AI. We investigate the potential scale of this vulnerability by measuring the degree to which AI assistants can complete assessment questions in standard university-level STEM courses. Specifically, we compile a novel dataset of textual assessment questions from 50 courses at EPFL and evaluate whether two AI assistants, GPT-3.5 and GPT-4 can adequately answer these questions. We use eight prompting strategies to produce responses and find that GPT-4 answers an average of 65.8% of questions correctly, and can even produce the correct answer across at least one prompting strategy for 85.1% of questions. When grouping courses in our dataset by degree program, these systems already pass non-project assessments of large numbers of core courses in various degree programs, posing risks to higher education accreditation that will be amplified as these models improve. Our results call for revising program-level assessment design in higher education in light of advances in generative AI.