Artificial Intelligence (AI), a term first coined by John McCarthy, can be broadly defined as the engineering and scientific pursuit of developing machines capable of replicating intelligent human behavior in learning, reasoning, and decision making. To date, several explanations of AI have been proposed. For instance, as argued by Jamal and Wejung (2019), AI allows machines—be them computers, robots, or software—to intelligently operate by adapting to situational inputs. Likewise, Shabbir and Anwer (2018) described AI systems as intelligent agents capable of interacting with their environments and optimizing outcomes. From a broader perspective, Mitchell (2019) defined AI as any computational system capable of performing functions that are typically associated with human intelligence. Along similar lines, Hesham et al. (2023) highlighted AI’s growing ability to perform complex tasks exclusive to humans, such as language translation, image processing, and decision making. Furthermore, highlighting AI’s adaptive nature, Kaplan and Haenlein (2019) emphasized AI’s ability to learn from external data, adjust accordingly, and perform tasks through continuous improvement. Previous studies also highlighted how supercomputers equipped with sensors and adaptive behaviors can reshape human engagement with machines (Chen et al., 2020), as well demonstrate intelligence through pattern-recognition algorithms (Sadiku et al., 2021). In the domain of education, AI was argued to be a subfield of computational intelligence capable of creating dynamic, individualized learning experiences (e.g., Imoh, 2023). Overall, the perspective on AI briefly reviewed above suggest that, as a rapidly evolving field, AI can exert a strong impact on educational practices, including those in architectural education.

Several previous studies explored the potential of AI in the context of (architectural) education. For instance, introducing the notion of Artificial Intelligence in Education (AIE), Kengam (2020) and Lee (2020) argued that AIE is essential in present-day educational contexts. While Chen et al. (2020) traced the evolution of AI tools in education—from simple administrative platforms to humanoid robots and intelligent systems—Panigrahi (2020) demonstrated how AI can effectively support flexible and personalized learning environments, especially in developing countries. Several other studies demonstrated how, through enabling virtual classrooms, adaptive learning, and intelligent evaluation, AI can considerably improve the quality of instruction and student outcomes (e.g., Huang et al., 2021; Lazar et al., 2022), as well as deliver dynamic, student-centered educational experiences (Imoh, 2023). In the domain of architectural design education, several authors highlighted the need to integrate digital and computational tools (e.g., Oxman (2008); Ceylan (2020)), while Chassignol et al. (2018) noted that AI tools are already influencing curriculum creation, teaching methods, and student evaluation in architectural education. Collectively, the studies briefly reviewed above highlight how recent advances in AI tools can reshape both general education and architectural education, in particular.

Recent studies support the usage of AI in transforming architectural education by improving creativity and design productivity. Using AI in teaching models improved student outcomes, mainly in work productivity and innovative thinking (Jin et al., 2024). In conceptual design stages, AI tools like Midjourney have assisted students translate written text into visual ideas, enriching form-finding capabilities (Ahmed and Elhakeem, 2024). Similarly, research at Gdańsk University identified three approaches of AI incorporation—semi-traditional, hybrid, and hybrid-interactive—that assist more contextual and creative outputs (Jan et al., 2024). Studies in the use of AI in sustainability studios show increases student engagement with environmental indicators, particularly when paired with organized curriculum support (Petrović and Stanković, 2024). Overall, successful implementation depends on aligning training with architecture education, addressing faculty indecision, and embedding AI within hands-on learning environments. More targeted strategies are needed to ensure architecture schools are not left behind in this educational transformation. However, there remains limited clarity on how to adapt pedagogical models to balance technical innovation with design creativity, especially in underrepresented contexts such as the Middle East.

Research on AI-aided architectural design was conducted by Jan and Kacper (2018). The study presented how the authors applied AI algorithms such as swarm intelligence, neural networks, and evolutionary algorithms in architectural practice. Analog input/output methods rooted in automation and vision were also discussed. The study noted that these strategies enable the emergence of new spatial solutions using relatively simple AI-based algorithms, many of which could only be achieved with specialised programs.

Jan et al. (2024) studied the application of AI in architectural education through a green campus development project. The paper examined the need for modifying the design instruction process in response to AI-driven innovations, particularly those related to image generation. In a research-by-design studio in Poland, students who used AI tools achieved more creative and innovative results than those who used conventional methods. Based on the findings, three collaboration approaches were identified: in the semi-traditional method, AI-generated images served as motivation; in the hybrid method, students integrated these images into their design choices; and in the hybrid-interactive method, students used AI in real-time to generate more site-specific outcomes. The study concluded that redefining design instruction to include these methods is essential and demonstrated the strong potential of AI tools—especially image-generation models—in enhancing architectural education.

Jin et al. (2024) also studied the impact of AI-assisted design in a 9-week architectural design and programming course involving 24 international students. The findings showed that the AI-integrated teaching approach positively affected students’ learning, efficiency, and creativity. However, the study also revealed inconsistencies in how students used AI tools and issues with the unpredictable nature of AI-generated outputs. The use of AI by students was often disorganised, despite the strong link between AI and architectural programming.

AI in architectural education has been addressed in several studies (Basarir, 2022; As et al., 2018; Qian et al., 2023). This research builds on that foundation by examining students’ awareness of AI tools and the extent to which these tools are applied in their design work. The aim is to assess both theoretical familiarity and actual usage in architectural education settings.

The benefits of using AI in teaching architecture have been consistently highlighted in the literature. For instance, as argued by Ma and Jiang (2023). AI in education indicates to the use of AI technologies in many aspects of the educational practice in both virtual and real-world environments, through smart educational tools and diverse learning scenarios. AI is currently used in education in a variety of ways, including the creation of intelligent teaching structures, chatbots to assist students, machine learning, and natural language processing (NLP) integrated into educational materials. Additional applications include administrative support, student evaluation tools, and teacher training systems.

Furthermore, Ceylan (2021) detailed how AI is transforming architecture through its impact on theory, practice, regulation, and society. The authors demonstrated that AI enables innovative design, improves building efficiency, and streamlines construction using advanced algorithms. Since architectural work frequently involves long-term, interdisciplinary collaboration, numerous concerns have been voiced about AI’s potentially adverse impact on professional roles and creative processes. Accordingly, Ceylan (2021) explored how students majoring in architecture perceive the shift towards AI-driven creativity and how they think of AI with regard to their future employment and design identity.

In another relevant study focused on the role of AI in enhancing architecture students’ understanding of design inputs, processes, and outcomes, Basarir (2022) proposed a continuum-based learning approach and discussed several potential future directions for AI-based architecture education in the AEC sector. Similarly, in a recent investigation of the advantages of AI in education, Imoh (2023) highlighted AI’s ability to solve key challenges, transform teaching methods, and support the achievement of SDG 4.

Finally, in a recent investigation that compared parametric design with machine-generated outputs, Muhammad and Ahmed (2023) identified important inspirational and practical functions of using AI tools in architecture. Based on this evidence, the authors underscored the need to integrate human-centered, culturally sensitive, and environmentally responsible AI into architectural work. A similar point was made by Nermen and Nevine (2024) who, in their framework for AI integration into various design stages, highlighted time-saving benefits of applying AI tools in architectural work.

Along with the benefits of using AI in teaching architecture discussed, several previous studies outlined important challenges associated with such use. Overall, compared to other sectors, the use of AI in architecture is still in its infancy, and many tools are yet to be adopted and thorough explored. In addition, as argued by Jamal and Wejung (2019), despite the growing use of AI in the field of architecture, many of AI-based technologies still require considerable human oversight, so most of these technologies remain only partially intelligent.

Another challenge associated with using AI in architectural education, highlighted by Tambuskar (2022) review of both drawbacks and advantages of AI in education using a narrative synthesis method, concerns AI’s thoughtful implementation. Without a careful consideration of this potential limitation of AI, its use in education can bring about important privacy, security, and safety concerns. A similar argument was made by Hesham et al. (2023) who conducted a comprehensive review of AIEd (Artificial Intelligence in Education) resources published between 2018 and 2022. The authors’ results of a systematic revive of 60 recent studies selected papers and expert opinions revealed that, while AI has both documented and expected educational benefits, its implementation should take into consideration potential ethical and privacy issues. Likewise, in their study on various studies on the advantages and limitations of AI in education, Apolzan and Cimpineanu (2024) flagged important concerns about bias, trust, cost, and data security, arguing for the need to elaborate appropriate legal and financial frameworks to address these issues.

Furthermore, one more potential limitation of using AI in the field of architecture was highlighted by Softaoglu (2024) who examined how AI could revitalize architectural design through human–robot collaboration. Focusing on the Midjourney robot, the author argued that, although architectural history, language, and semiotics could guide AI toward generating meaningful spatial outcomes, because of the limitations of the library source and language used, AI still faces various ethical and cultural challenges. To understand how artificial intelligence is represented in architectural education, it is helpful to map its intersection with the various phases of the architecureal education process. While some stages—such as planning, progress tracking, social aspects, ideation, creativity, and rendering—are frequently highlighted, others like, efficiency, creativity, ideation, rendering and construction receive less direct attention. The following figure synthesises key findings from the literature and illustrates the scope and depth of AI’s educational integration.

Despite such global insights, research on AI in architectural education remains limited within the Middle East and North Africa (MENA) region. Cultural perceptions of automation, institutional frameworks, and policy environments are often overlooked. Figure 1 illustrates how unevenly AI is applied across architectural tasks in international literature—further reinforcing the need for context-specific exploration. In response, this study investigates how Saudi students and faculty engage with AI in their academic and design activities, contributing a culturally grounded perspective to the broader discourse.

It involved 160 students and 32 faculty members who were interviewed using structured questionnaires. The questionnaires were developed in light of the literature on artificial intelligence in education and included key questions related to the awareness level, usage patterns, and perceptions about the usage of artificial intelligence in the architectural field. Data collection lasted for a period of 3 weeks and maintained anonymity for participants to adhere to ethical guidelines. Figure 2 illustrates this comparison between students and faculty in terms of AI awareness, usage frequency, and interest in learning.

Data are summarized using frequency and descriptive statistics, while similarities and differences between groups are assessed using descriptive and inferential statistical methods. For descriptive statistics, the means and standard deviations, among others, were considered; for the inferential techniques, however, linear regression, Pearson correlation, and Chi-Square tests on the relationships of variables were performed, showing, for example, how demographic factors such as age impact awareness and the use of AI. Analysis was done using [Statistical Software] and all tests of significance at p < 0.05 level.

Data collected and analyzed during this study are not publicly available at this time. In the meantime, the data will be made available to any researcher or reviewer upon request through the journal for transparency and verification of the study’s result. While the student sample (n = 160) was robust and representative of the department’s enrolment, the faculty sample size (n = 32) is relatively small. This reflects the limited pool of faculty available within the department and may constrain the generalizability of findings related to faculty perceptions. This limitation is taken into account during result interpretation.

The descriptive analysis, based on responses from 160 students and 32 faculty members, indicates a significant difference in awareness about AI and the usage of AI between the two groups. Students were found to be more aware and frequent users of AI compared to the faculty.

A regression analysis of students and faculty was performed to depict how the characteristics influence the overall AI experience in education for students and faculty. The independent variables included AI awareness, sources of AI knowledge, frequency of AI usage, skills developed through AI, and impact of AI on practical skills development. The dependent variable was the overall experience with AI in education.

The disparity in AI awareness—58.8% among students versus only 3.1% of faculty—strongly aligns with Rogers’ Diffusion of Innovation Theory (2003), which posits that different age groups adopt technologies at varying rates. Accordingly, in our results, the surveyed students, who showed enthusiasm for integrating AI into their architectural learning, clearly represent early adopters of AI. By contrast, the surveyed faculty members, who belong to an older generation group, exhibited reluctance, which was underpinned by their lack of formal training or exposure to AI technology.

One key takeaway from these findings is the need for structured and targeted training for faculty. While younger students are quick to see the potential in AI, faculty may need clearer demonstrations of AI’s practical and pedagogical value. According to Venkatesh et al. (2003), perceived usefulness and ease of use are key factors in technology adoption, and faculty may currently view AI as complex or irrelevant to their established teaching methods. This contrast is also evident when comparing other fields like business education, where technology adoption is more common due to its clear practical benefits (Teo, 2011). To close this gap in architectural education, it will be essential to develop training programs that clarify the specific advantages of AI in design education.

Along with the need for structured and targeted training for faculty discussed above, our results also highlight the importance of enhancing faculty’s practical engagement with AI tools. Specifically, as posited by Kolb’s (1984) Experiential Learning Theory (1984), the frequency of use of a technology is a strongest predictor of the corresponding positive experience with that technology. Likewise, Venkatesh et al.’s (2003) Technology Acceptance Model (TAM) highlights that actual adoption of and satisfaction with a technology are driven by its perceived ease of use and usefulness, not just awareness. What the aforementioned theoretical proposals highlight is that mere awareness of architecture students and, more importantly, of faculty staff about AI tools does not suffice—instead, this awareness should be consistently strengthened via regular engagement with AI tools. Said differently, hands-on experience with AI is essential to a successful integration of AI into the architectural classroom.

Similar conclusions about the imperative active use of AI tools in the classroom were previously made in other fields of education, such as engineering. For instance, Xu et al. (2020) convincingly demonstrated that engineering students’ engaged use of AI tools considerably improved their learning outcomes (Xu et al., 2020).

What this suggests is that institutions providing architectural education should focus on increasing their students’ and faculty staff’s engagement with AI through its real-world applications, such as via embedding AI into design studios, projects, or courses. Obviously, strengthening AI’s use in both academic and project-based settings would require considerable investment into appropriate resources and training.

An important qualitative finding in our results was that, unlike students, the surveyed faculty staff voiced a serious concern that over-reliance on AI could hinder architectural students’ creativity, or their ability to develop independent, creative solutions. This hesitation about AI use should be meaningfully addressed in educational settings.

Overall, the aforementioned concern of the surveyed architectural staff resonates with broader debates on the role of automation in creative fields (Autor, 2015) and reflects a broader pedagogical tension between innovation and creativity. Indeed, while the faculty staff acknowledged that AI is capable of introducing new forms, workflows, and design solutions, they were concerned that the automation of these outputs increased the risk of short-circuiting internal cognitive processes that underpin creative development.

This concern of the surveyed architectural staff may arise from several earlier theories. For instance, as posited by Amabile (1996) Componential Theory of Creativity, creativity arises from the interplay between domain-relevant skills, creativity processes, and intrinsic motivation. All these components may be compromised by architectural students’ heavy reliance on AI tools. Another relevant theoretical framework—Csikszentmihalyi (1999) Systems Model of Creativity (1999)—highlights that creativity arises from the interaction among elements that AI cannot replicate independently—namely, from individual skills, disciplinary norms, and social validation in the creative act. Taken together, the aforementioned two theoretical frameworks highlight that, although AI may support innovation, true creativity arises from deeply engaged, iterative problem-solving and meaning-making.

However, in more recent research, appropriate use of AI was convincingly demonstrated to enhance creative thinking by offering new perspectives (Anderson et al., 2018). This highlights the need for a balanced approach to AI integration in education—the one where AI is positioned not as a shortcut to outcomes, but as a co-creative partner that stimulates, rather than replaces, the design process. An important practical implication of this finding is that future training programs on the use of AI in educational contexts should focus on how to incorporate AI in ways that would stimulate creativity, without reducing the importance of human input in design processes.

Another reason underlying the surveyed faculty members’ hesitancy towards AI adoption is related to educators’ concern that students’ over-reliance on AI tools could result in bypassing essential design thinking stages and uncritical acceptance of AI-generated outputs—that is, skipping the learning process of how to design to merely choosing what to accept from among multiple AI-generated proposals. This potential challenge was also mentioned in recent architectural education literature warning about “design shallowing” – that, is, situations when students skip the iterative process of problem-solving and experimentation. From the applied perspective, addressing these concerns requires not only technical training, but also pedagogical frameworks that would meaningfully embed AI within creative workflows, ensuring that students genuinely learn with AI, rather than passively use its outputs.

Another major finding in our results concerns the strong interrelationship between AI skill development and positive experiences with using AI in architectural education. This finding largely aligns with previous research demonstrating that the belief that a system will improve performance is a major factor in its adoption (Venkatesh et al., 2012), What this suggests for the context of architectural education is that there is a clear need to provide both students and faculty with structured opportunities to build AI-related skills. Along with improving students’ and educators’ proficiency with AI tools, such opportunities will also increase the perceived value of AI tools in design practice.

More concretely, taking into account the evidence showing that structured training programs can significantly improve technology acceptance (Schoonenboom, 2014), educational institutions offering undergraduate programs in architecture should consider implementing curriculum enhancements that focus on introducing AI tools that are directly relevant to architecture, such as parametric design or AI-driven modeling. Relevant curriculum enhancements can include—but are not limited to—structured workshops, courses, or labs. Collectively, these innovations in architectural curricula can provide both students and faculty with the hands-on experience they need to feel comfortable integrating AI into their work.

On a broader level, the results of the present study point to a certain imbalance in how all actors involved in architectural education perceive emerging technologies like AI. One of the solutions to bridge this gap students and educators is introducing a more collaborative approach to integrating AI into the curriculum. This could involve introducing joint projects that would concurrently tap on students’ technical proficiency with AI tools and educators’ pedagogical expertise, thereby fostering a shared understanding of AI’s potential. An alternative path towards encouraging a stronger faculty engagement with AI could also involve, according to Rogers (2003) Diffusion of Innovation Theory, offering targeted incentives such as professional development credits or research grants that would ensure a smoother and more widespread integration of AI in architectural education.

The identified imbalance in architecture students’ and educators’ perceptions of utility of AI tools in the learning process should also be viewed through the lens of the Saudi and Gulf context. The observed faculty members’ hesitation towards decentralized, tool-based learning can also be explained through the historical emphasis of Saudi educational systems on instructor authority and standardized content delivery. Educators’ resistance to AI-based instruction can also be underpinned by cultural values in the Gulf—such as preference for human-led mentorship, sensitivity to technological automation, and emphasis on group conformity. This specificity of the Saudi educational context should be carefully considered in future AI integration strategies.