Artificial intelligence and academic integrity in architectural education: challenges, risks, and future directions

Abstract

The rapid adoption of generative artificial intelligence (GenAI) is transforming architectural design practice; however, its implications for academic integrity in architectural education remain insufficiently understood. Limited research has examined how GenAI influences authorship, assessment, and studio culture, while evidence-based institutional guidance remains scarce. This study critically investigates GenAI's influence on academic integrity in architectural education through a systematic review conducted in accordance with the PRISMA guidelines. A total of 32 peer-reviewed publications from 2020 to 2026 were retrieved from the Scopus database and analyzed to examine GenAI adoption, associated integrity risks, and emerging institutional responses in architectural and design education. Thematic analysis revealed a paradigmatic shift toward learning by co-doing , in which students increasingly curate rather than create AI-generated outputs. The most prominent challenge identified was the Efficiency–Feasibility Paradox, whereby greater ideation efficiency often comes at the expense of design feasibility. Additional risks include cognitive offloading, design fixation, process fabrication, and biases arising from predominantly Eurocentric training data. The findings highlight the need for process-oriented assessment, AI verification literacy, hybrid human–AI pedagogies, and institutional policies that promote transparency, equity, and responsible AI use.