Aim To investigate the ethical implications of AI-driven natural language generation systems (NLG), such as ChatGPT, in university education at the University of Jordan from the perspective of academic staff. Specific objectives include assessing awareness levels, identifying ethical concerns, and exploring institutional support and training needs. Methods A cross-sectional survey of 274 academic staff members (111 females and 163 males) from diverse disciplines was conducted. Participants provided demographic data and responded to questions on knowledge, awareness, ethical concerns, and institutional support related to AI in education. Statistical analyses included descriptive statistics, Chi-square tests, Mann–Whitney U tests, and regression analyses. Results Varied awareness and usage of AI-driven NLG among academic staff were found, with 46.7% using AI for university tasks, primarily in research (46.4%) and teaching (33.9%). While 62% demonstrated moderate to high knowledge of AI in academia, which varied by academic discipline (p < .001, Table <InternalRef RefID="Tab3">3</InternalRef>). Fewer than one-third were familiar with current ethical guidelines for AI in education (Table <InternalRef RefID="Tab2">2</InternalRef>). Ethical concerns about AI use by students were prevalent (70% of academic staff), especially regarding plagiarism and exam cheating (Table <InternalRef RefID="Tab2">2</InternalRef>). Editing and language corrections (70.8%) were deemed ethically acceptable uses of AI for students (Table <InternalRef RefID="Tab2">2</InternalRef>), with ethical concerns over effortless assignments, sole reliance on AI for research, and exam cheating. Regression analyses revealed that factors such as gender, age, academic rank, teaching experience, and discipline influenced attitudes and knowledge of AI (p < .05, Tables <InternalRef RefID="Tab5">5</InternalRef> and <InternalRef RefID="Tab6">6</InternalRef>). Conclusion This study reveals that a significant proportion of academic staff at the University of Jordan use AI and have moderate to high knowledge of it. Although there are gaps in their understanding of AI ethics, they expect students to maintain high ethical standards. This underscores the need for comprehensive ethical guidelines, strong institutional support, and initiatives such as training programs and workshops on ethical AI use to support their role in guiding students in Jordanian academia.

AI in education: enhancing learning potential and addressing ethical considerations among academic staff—a cross-sectional study at the University of Jordan

期刊论文
出版模式：

开放获取
发表日期：

2025年5月27日

查看全文

Marwa M. Alnsour,
Latifa Qouzah,
Sanaa Aljamani,
Rasha A. Alamoush,
Mahmoud K. AL-Omiri

摘要

Aim

To investigate the ethical implications of AI-driven natural language generation systems (NLG), such as ChatGPT, in university education at the University of Jordan from the perspective of academic staff. Specific objectives include assessing awareness levels, identifying ethical concerns, and exploring institutional support and training needs.

Methods

A cross-sectional survey of 274 academic staff members (111 females and 163 males) from diverse disciplines was conducted. Participants provided demographic data and responded to questions on knowledge, awareness, ethical concerns, and institutional support related to AI in education. Statistical analyses included descriptive statistics, Chi-square tests, Mann–Whitney U tests, and regression analyses.

Results

Varied awareness and usage of AI-driven NLG among academic staff were found, with 46.7% using AI for university tasks, primarily in research (46.4%) and teaching (33.9%). While 62% demonstrated moderate to high knowledge of AI in academia, which varied by academic discipline (p < .001, Table 3). Fewer than one-third were familiar with current ethical guidelines for AI in education (Table 2). Ethical concerns about AI use by students were prevalent (70% of academic staff), especially regarding plagiarism and exam cheating (Table 2). Editing and language corrections (70.8%) were deemed ethically acceptable uses of AI for students (Table 2), with ethical concerns over effortless assignments, sole reliance on AI for research, and exam cheating. Regression analyses revealed that factors such as gender, age, academic rank, teaching experience, and discipline influenced attitudes and knowledge of AI (p < .05, Tables 5 and 6).

Conclusion

This study reveals that a significant proportion of academic staff at the University of Jordan use AI and have moderate to high knowledge of it. Although there are gaps in their understanding of AI ethics, they expect students to maintain high ethical standards. This underscores the need for comprehensive ethical guidelines, strong institutional support, and initiatives such as training programs and workshops on ethical AI use to support their role in guiding students in Jordanian academia.