Dies ist eine Übersichtsseite mit Metadaten zu dieser wissenschaftlichen Arbeit. Der vollständige Artikel ist beim Verlag verfügbar.
Graduate Students' Perceptions of Generative AI's Impact on Critical Thinking Development
0
Zitationen
5
Autoren
2026
Jahr
Abstract
Generative artificial intelligence (AI) tools are becoming increasingly common in graduate education, offering conveniences such as improved access to information, support with organization, and opportunities for personalized learning. At the same time, faculty and students continue to question whether frequent use of these tools may unintentionally weaken critical thinking, a skill central to advanced academic work. This study examined how graduate students perceive generative AI and how they believe it influences their development as critical thinkers. The project focused on students’ patterns of use, their confidence in applying AI ethically, and the benefits and risks they associate with these tools. An exploratory quantitative design was used with a sample of 58 graduate students representing 21 universities and a range of disciplines. The survey included both closed- and open‑ended questions addressing AI use, perceived effects on thinking, and ethical considerations. Quantitative data were summarized descriptively, and qualitative responses were coded thematically. Most students reported using AI sparingly, typically one to two hours per week, and mainly for tasks such as organizing ideas or proofreading. ChatGPT was the most frequently used tool. While a majority (62%) felt confident in their ability to use AI ethically, more than half (58%) believed that reliance on AI could reduce critical thinking or create dependency. Students acknowledged that AI can help clarify concepts and spark ideas but also expressed concerns about diminished creativity and overly passive learning. Reports of academic misconduct were uncommon. Overall, students viewed AI as a helpful supplement rather than a replacement for independent reasoning. The findings highlight the need for thoughtful instructional design that encourages students to critique and question AI outputs while maintaining active engagement in their own learning.
Ähnliche Arbeiten
Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI
2019 · 8.773 Zit.
Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead
2019 · 8.682 Zit.
High-performance medicine: the convergence of human and artificial intelligence
2018 · 8.242 Zit.
BioBERT: a pre-trained biomedical language representation model for biomedical text mining
2019 · 6.898 Zit.
Proceedings of the 19th International Joint Conference on Artificial Intelligence
2005 · 5.781 Zit.