Faculty Perceptions on the Use of Generative AI in Engineering Education

Abstract

Generative AI is a type of artificial intelligence capable of creating content like text, images, or music based on user prompts or inputs. Generative AI (Gen AI) models are trained on large amounts of data and use smart algorithms to create things that appear to be made by humans. This study aims to develop an understanding of Generative AIs current and potential impact on engineering education from the perspective of engineering faculty at an R2 institution. In particular, we aim to explore faculty perceptions of Gen AIs usefulness and ease of use following the Technology Acceptance Model (TAM) framework, faculty's intent to use AI vs actual use, how it has been integrated into the classroom (i.e., types of assignments or activities), and any benefits or concerns with adopting Gen AI in engineering education. We employed mixed methods statistical analysis techniques, such as descriptive and inferential statistics, as well as exploratory factor analysis, to identify pattern, trends, relationships, and contrasts with respect to faculty perceptions of using Gen AI in teaching engineering courses. QualtricsTM survey results were tabulated to provide insights into faculty perceptions on the use of Gen AI. Our results show that about 1/2 of faculty survey responses integrate Gen AI content into their teaching materials, about 1/2 of faculty responses use Gen AI frequently or regularly, and about 2/3 feel somewhat to extremely confident in using AI technologies in engineering education; however, most respondents use Gen AI as a text-based alternative for writing assistance (i.e., drafting, content generation, editing, and summarizing) and educational tools (i.e., creating quizzes or explanations). By discovering faculty perceptions towards integrating Gen AI into engineering curricula, this research contributes to ongoing discussions on the role of Gen AI in higher education and provides insight into the extent to which engineering faculty currently embrace Gen AI in engineering education at an R2 institution.

Headnote

Abstract-Generative AI is a type of artificial intelligence capable of creating content like text, images, or music based on user prompts or inputs. Generative AI (Gen AI) models are trained on large amounts of data and use smart algorithms to create things that appear to be made by humans. This study aims to develop an understanding of Generative Ais current and potential impact on engineering education from the perspective of engineering faculty at an R2 institution. In particular, we aim to explore faculty perceptions of Gen Ais usefulness and ease of use following the Technology Acceptance Model (TAM) framework, faculty's intent to use AI vs actual use, how it has been integrated into the classroom (i.e., types of assignments or activities), and any benefits or concerns with adopting Gen AI in engineering education. We employed mixed methods statistical analysis techniques, such as descriptive and inferential statistics, as well as exploratory factor analysis, to identify pattern, trends, relationships, and contrasts with respect to faculty perceptions of using Gen AI in teaching engineering courses. Qualtrics™ survey results were tabulated to provide insights into faculty perceptions on the use of Gen AI. Our results show that about 1/2 of faculty survey responses integrate Gen AI content into their teaching materials, about 1/2 of faculty responses use Gen AI frequently or regularly, and about 2/3 feel somewhat to extremely confident in using AI technologies in engineering education; however, most respondents use Gen AI as a text-based alternative for writing assistance (i.e., drafting, content generation, editing, and summarizing) and educational tools (i.e., creating quizzes or explanations). By discovering faculty perceptions towards integrating Gen AI into engineering curricula, this research contributes to ongoing discussions on the role of Gen AI in higher education and provides insight into the extent to which engineering faculty currently embrace Gen AI in engineering education at an R2 institution.

Keywords-Faculty, Faculty perceptions, Generative AI, Survey, Engineering education

I. Introduction

The field of engineering education is undergoing a paradigm shift in how engineering courses will be developed and taught. The adoption readiness and integration of Gen AI tools such as ChatGPT, Gemini, Copilot, and others have been met with both eagerness and caution largely due to uncertainty regarding its capabilities and potential impact. The integration of Gen AI within academia has sparked many conversations and debates, particularly with concerns about academic integrity, reliability of the output, and whether its adoption in the classroom will lead to a deterioration of critical thinking skills [1]. As institutions continue to understand Gen Ais potential impact (both positive and negative), and develop policies for its use in academia, it is important to learn faculty perspectives on this rapidly changing disruptive technology. This paper investigates how the engineering faculty at Kennesaw State University, an R2 institution, perceives the use of AI in engineering education, using survey data administered during Fall 2024 and Spring 2025. Of the 108 current faculty members in the Southern Polytechnic College of Engineering and Engineering Technology (SPCEET), 35 faculty, or about 1/3, completed the survey.

II. Background

The rampant growth of Gen AI technology has been met with both anticipation and trepidation in higher education. For example, while ChatGPT has sparked widespread discussions in academic communities, particularly concerning its implications for teaching and learning, concerns over academic integrity and institutional support abound. The Technology Acceptance Model (TAM) in Fig. 1, a well-established and widely recognized framework due to its reliability in predicting information technology adoption [2-5], suggests that both technology's perceived usefulness and ease of use are crucial factors in perceptions towards adopting technology, leading to its eventual acceptance and continued use [6].

The TAM model serves as the theoretical framework for analyzing the survey data collected in this study. Following Ajzen's seminal work with Theory of Planned Behavior [7-8], the TAM model posits that an individual's intention to adopt a particular information technology (IT) is influenced by two distinct factors: (1) perceived usefulness, and (2) perceived ease of use. Whereas perceived usefulness is defined as the extent to which a potential IT user perceives that the use of that IT system will improve one's job performance [9], perceived ease of use is defined as the extent to which a potential IT user perceives that use of the IT system will be free from effort [9]. Additionally, TAM postulates that perceived usefulness has a direct, positive effect on a potential user's perceived intention to adopt and use an IT. Perceived ease of use directly affects perceived usefulness and, thus, has a direct effect on potential intent to use an IT [9].

Engineering faculty members, as key stakeholders in our Southern Polytechnic College of Engineering and Engineering Technology which houses six departments and twelve engineering majors with over 5000 students, play a pivotal role in the adoption and integration of new technologies into engineering education. However, while technology brings new opportunities and benefits to educators in their teaching and research practices, some faculty resist adopting new technologies for various reasons, and this resistance from educators can be a key roadblock to the eventual adoption and use of technology [10].

By utilizing the TAM model, this preliminary study reveals significant insights into the transformation of attitudes of engineering faculty at an R2 institution. A principle factor contributing to the perceived ease of use and usefulness of Gen AI begins with faculty experience with Gen AI technology. Although some faculty have resisted using this technology based on survey results, others have embraced it and use Gen AI frequently (i.e., 1-3 times a week) (30.3%) or regularly (i.e., daily) (21.2%). Fig. 2 shows that about 41% (14/34) of respondents either agree or strongly agree with integrating AIgenerated content into their teaching materials. Fig. 3 illustrates ways in which Gen AI activities have been used to assist engineering faculty - the most prevalent being writing assistance (i.e., drafting, editing, and summarizing), educational tools (i.e., creating quizzes, and explanations), and a search engine.

The transition from introduction of Gen AI to its eventual adoption, in addition to the factors influencing it as predicted by the TAM model, offer valuable insights into how disruptive technologies are integrated into engineering curricula.

III. Methodology

A Qualtrics™ survey was administered via the KSU weekly electronic newsletter, The Weekly Feed, to all engineering faculty within SPCEET between October 2024 and January 2025, yielding a total of 3 5 responses among the 108 fulltime engineering faculty in the college, or a 32.4% response rate. There are 31 questions on the survey consisting of Likertscale, multiple choice, and free response questions. The Likert scale ranged from 1 (Strongly disagree) to 5 (Strongly agree). The four categories of questions include Demographics, Technology Acceptance, AI Integration and Usage, and Perception and Attitudes Towards AI.

The data cleaning process involved converting responses from the Likert scale into numerical form, mapping each level of agreement or disagreement into a corresponding numeric value. Not all questions from the survey instrument were used in this analysis. Moreover, there were some qualitative questions that have been omitted from this analysis due to the nature of those questions. A total of fifteen questions were used in this analysis. The survey questions were thoroughly evaluated to ensure that there was no intentional bias. The full survey instrument can be found in the Appendix. A Chi-square analysis was conducted to explore the association of factors with participants' perception of using Gen AI in engineering education. Exploratory factor analysis (EFA) was used to assess the survey responses with the intention of understanding what differences might exist between engineering faculty's intention to adopt Gen AI in their teaching practices through actual use and perceived ease of use (Factor 1) and the behavioral intention of using AI and perceived usefulness in improving one's job performance by doing so (Factor 2). EFA was chosen for its ability to uncover the fundamental theoretical structure of a relatively large set of variables [11]. This approach enabled us to discover important factors emerging from the faculty responses. The analysis was conducted using the JASP (Jeffs Awesome Statistical Program v. 0.19.3) publicly available statistical software package [12]. The number of factors were derived from a Parallel analysis based on Factor Analysis, an

Oblique rotation using the promax rotation, a base analysis using a correlation matrix, a maximum likelihood factoring method, and a factor size loading above 0.3 to capture any nuanced relationship between measurement items.

IV. Results

The EFA on the Gen AI faculty data set in Table 1 reveals two factors that account for 48.9% of the proportional variance in the rotated solution. These factors focus on faculty perceptions of Gen Ais impact on faculty adoption.

In Table 2, Perceived Ease of Use - Training to use generative AI tools is simple and effective has the least positive correlation (r = 0.494), while Actual System Use - regular use of generative AI tools in my lesson planning and delivery has the highest positive correlation (r = 0.882) with the factors in Factor 1. The perception that AI will positively impact the future of engineering practice (r = -0.332) and faculty confidence levels with using AI technologies (r = -0.321) are cross-loaded between both factors (i.e., they are influenced by multiple underlying constructs) and negatively correlates with Factor 1. Whereas Factor 1 is specifically related to actual use of Gen AI and its perceived ease of use and adoption, Factor 2 specifically concerns behavioral intentions of using AI and its perceived usefulness in improving one's job performance. Here, the extent to which AI is integrated into engineering curriculum/coursework has the least positive correlation (r = 0.370). In contrast, the faculty's behavioral intention to use AI tools in future courses has the highest correlation (r = 0.902) with the factors in Factor 2.

V. Discussion

A breakdown of survey questions and results relevant to Factors 1 and 2 under the Technology Acceptance Model

category follows:

Perceived Usefulness

Q: Generative AI tools improve the quality of my teaching and materials. About 63% (22/35) of faculty responses "agreed" to "strongly agreed" whereas about 37% (13/35) "disagreed" to "strongly disagreed."

Q: Generative AI supports my teaching by providing diverse examples and explanations. The results are mixed with about 49% of faculty responses "agreed" to "strongly agreed" while about 51% either "disagreed" or "strongly disagreed."

Perceived Ease of Use

Q: Integrating generative AI into my teaching is straightforward. About 57% (20/35) of faculty responses either "agreed" or "strongly agreed," 29% were "neutral," and 14% either "disagreed" or "strongly disagreed."

Q: Training to use generative AI tools is simple and effective. About 66% (23/35) of faculty responses either "agreed" or "strongly agreed," 23% were "neutral," and 31% "disagreed" or "strongly disagreed."

Behavioral Intention to Use

Q: I plan to incorporate generative AI tools in my future courses. About 63% (22/35) "agreed" or "strongly agreed," 29% were "neutral," and 9% "disagreed."

Q: I would encourage colleagues to use generative AI in their teaching. About 65% (22/34) of faculty responses either "agreed" or "strongly agreed," 32% were "neutral," and 3% "disagreed."

Actual System Use

Q: I regularly use generative AI tools in my lesson planning and delivery. About 34% (12/35) "agreed" or "strongly agreed," 20% were "neutral," and 43% "disagreed" or "strongly disagreed."

Q: I integrate Ai-generated content into my teaching materials. About 40% (14/35) "agreed" or "strongly agreed," 29% were "neutral," and 31% "disagreed" or "strongly disagreed."

AI Integration and Usage

Q: On a scale from 1 (not at all) to 5 (extensively), how much is AI integrated into your curriculum/coursework? About 9% (3/32) of faculty responses use AI "very much" or "extensively," 31% use AI "moderately," and 60% use AI "slightly" or "not at all."

Q: How confident do you feel using AI technologies? About 64% (21/33) of faculty responses feel "somewhat confident" or "extremely confident," 30% feel "neutral," and 6% feel "somewhat unconfident" or "not confident at all."

Whereas about one-third of faculty survey responses indicate "neutral" to "not confident at all" in using AI, their behavioral intent suggests that 63% (22/35) of faculty plan to use Gen AI in their future courses and 65% (22/34) would encourage their colleagues to use Gen AI in their teaching. To summarize, of the 40% who have integrated AI into their curriculum or coursework, 64% of those faculty feel "somewhat" to "extremely confident" in using AL

Regarding the confidence level of faculty using Gen AI, 21.9% of respondents were "Extremely confident" with AI tools, and 40.6% were "Somewhat confident," while 31.3% are "Neutral." Of the responses that use Gen AI either frequently or regularly (daily), the distribution of responses by engineering department are as follows: Civil and Environmental Engineering (29.4%), Mechanical Engineering (23.6%), Industrial & Systems Engineering (17.6%), Electrical & Computer Engineering (17.6%), Mechatronics (5.9%), and Engineering Technology (5.9%).

Survey comments include:

* More execution practice, more intentional integration, and more guidance from the university is needed.

* I know it can do a lot more than what Eve done so far.

* I think we are in the development stages, but it is maturing

* fast every day.

* I use it every day and am efficient and effective in generating the right questions for answers I want using AI.

* It can help give a new approach to questions.

* It is becoming more like Google.

* I use Gen AI for help if I'm stuck.

Results for the perception that AI will positively impact the future of engineering practice are illustrated in Table 3.

As shown in Table 3, coupled with survey results, even neutral or low-confidence users often believe that AI will have a positive future impact. High-confidence users (especially "Extremely confident") nearly always associate AI with a strong positive future impact.

Demographic Information

Among the survey responses, 24% identified as female faculty and 76% identified as male. Fig. 3 illustrates a wide range in academic experience ranging from 1-55 years with the median of around 8 years. About 64% (21/33) had between 1 and 10 years of faculty experience.

As shown in Fig. 4, most faculty responses came from Associate Professors, Adjunct Professors, and Lecturers.

Fig. 5 shows Industrial and Systems Engineering (ISYE), Mechanical Engineering (ME), and Electrical and Computer Engineering (ECE) departments with the most faculty responses.

In Fig. 6, most participant responses are in the 45-54 age range.

Fig. 7 shows a breakdown of race/ethnicity among survey participants.

VI. Limitations

This study is limited to self-reported data from a relatively limited sample size of n = 35 faculty survey responses from the college of engineering at a single R2 institution, which can come with a particular set of biases and varying levels of familiarity with Gen AI. Hence, the results of this study cannot be generalized to the larger population given the limitation in context and institution.

VII. Conclusions and Future Work

Insights from this preliminary study at Kennesaw State University indicate that most engineering faculty that responded to the survey are cautiously optimistic about Ais educational impact. However, concerns about privacy, ethical use, technical readiness, and institutional support persist. While more than 50% of respondents use Gen AI frequently or daily, almost half of the respondents use Gen AI occasionally, rarely, or not at all.

Current faculty use of Gen AI is largely for writing assignments (i.e., drafting, editing, summarizing, checking grammar, writing codes, or creating meeting notes or action items) or text-based (i.e., writing assistance, chatbots, or content generation such as quiz questions).

A majority (56.3%) of faculty participants believe Gen AI would positively impact engineering practice "a lot," while only 6.2% of respondents expressed skepticism.

The primary concerns of using Gen AI include reliability and accuracy of AI tools (24.7%), data privacy and security (17.3%), ethical considerations (16.0%), and lack of technical expertise (14.8%).

Our engineering faculty identified the following Top 3 primary benefits of using Gen AI: (1) AI automates repetitive tasks, (2) AI provides customized learning support for students, and (3) AI encourages innovation.

Future research should aim to include a broader range of institutions and stakeholders to strengthen the robustness and broad applicability of the results. Longitudinal studies can be conducted to track evolving attitudes and policy shifts using the TAM framework to identify factors that influence their attitudes towards potential integration of AI in their course development, teaching, and research practices. As such, our research team will conduct a sequel study next year. This paper offers initial insights into how an R2 institution's (Kennesaw State University) engineering faculty view the use of Gen AI in engineering education.

References

References

[1] C.K. Lo, "What is the impact of ChatGPT on education? a rapid review of the literature," Education Sciences, vol. 13, no. 4, p. 410, 2023.

[2] V. Venkatesh, "Determinants of perceived ease of use: Integrating control, intrinsic motivation, and emotion into the Technology Acceptance Model," Information Systems Research, vol. 11, pp. 342-365, 2000.

[3] V. Venkatesh, "A model of the antecedents of perceived ease of use: Development and test," Decision Sciences, vol. 27, pp. 451-483, 1996.

[4] R. Agarwal and J Prasad, "Are individual differences germane to the acceptance of new information technologies?" Decision Sciences, vol. 30, pp. 361-391, 1999.

[5] P.J. Hu, P.Y.K. Chau, O.R.L. Sheng, and K.Y. Tam, "Examining the Technology Acceptance Model using physician acceptance of telemedicine technology," Journal of Management Information Systems, vol. 16, pp. 91112, 1999.

[6] J. Miller, and O. Khera, "Digital library adoption and the Technology Acceptance Model," The Electronic Journal of Information Systems in Developing Countries, vol. 40, no. 6, pp. 1-19, 2010.

[7] I. Ajzen and M. Fishbein, "Understanding attitudes and predicting social behavior f Englewood Cliffs, NJ: Prentice-Hall, 1980.

[8] R.S. Keyser, L. Li, E. Kimber, B. Berman, G. Kerbage, M. Valero, Y J. Li, and A.M. Ukuku, "Exploring the blood donation problem in the U.S. and the use of emergent and disruptive technologies to attract new donors," International Journal of Research in Industrial Engineering, vol. 12, no. 2, pp. 256-280, 2025.

[9] J.Y.L. Thong, W. Hong, and K.Y. Tam, "Understanding user acceptance of digital libraries: What are the roles of interface characteristics, organizational context, and individual differences?", International Journal of HumanComputer Studies, vol. 57, pp.215-242, 2002.

[10] K. Watty, J. McKay, and L. Ngo, "Innovators or inhibitors? Accounting faculty resistance to new educational technologies in higher education," Journal of Accounting Education, vol. 36, pp. 1-15, 2016.

[11] B. Williams, A. Onsman, and T. Brown, "Exploratory factor analysis: A fivestep guide for novices," Australasian journal of paramedicine, vol. 8, pp. 1-13, 2010.

[12] JASP Team, "JASP (Version 0.19.3) [Computer software]," 2025. [Online]. Available: https://jasp-stats.org/