Abstract: Generative AI (Gen AI), which refers to computational techniques that are capable of generating seemingly new, meaningful content such as text, images, or audio from training data (Feuerriegel et al., 2024), has greatly impacted education. In design education, the most distinctive culture is often to focus primarily on time-consuming graphic elements that may present many potential solutions. Gen AI interventions can help studio-based students quickly create more options to find better solutions to real or hypothetical design problems. Its application has presented a technological revolution by changing the approaches that people often work and communicate. Therefore, it is crucial to incorporate Gen AI in studio-based design education to help students build competitive skills that will enhance their employability. In this research, Gen AI mainly includes Photoshop AI and XIPU AI Chatbot. The paper seeks to investigate the sophisticated dynamics between them and the design studio within higher education by implementing a pilot project. This project will integrate Photoshop AI and chatbot into a second-year design course at a transnational university in China. The central research question is "How can the integration of Photoshop AI and chatbots into a second-year design course impact the studio-based design culture, and what challenges and solutions emerge in the process of developing and deploying these applications within the educational context?” Using action research as the underpinning methodology, qualitative data will be collected through classroom observations and student interviews. The final outcome of this study would be a comprehensive understanding of how generative AI can be effectively integrated into studio-based design education to enhance student learning and creative outcomes and how this integration affects the studio-based design culture.

Keywords: generative AI (Gen AI), Photoshop AI (PS AI), XIPU AI Chatbot, studio-based culture, design education

Case Highlights:

• A combined use of Photoshop AI and XIPU AI chatbot to enhance student's ability in graphic expression, communication, spatial analysis, and interpretation
• AI-enhanced authentic assessment aligned with design studio culture, promoting real-world relevance and improving students’ employability
• Active student participation in reimagining the assessment process, ensuring it meets their learning needs
• An innovation guided by action research methodolog

1         Introduction

1.1         Background

According to Feuerriegel et al. (2024), Generative AI refers to computational techniques that are capable of generating seemingly new, meaningful content such as text, images, or audio from training data. The content has greatly impacted education, especially in HE (Higher Education). Many scholars have started to conduct Gen AI from the perspective of business and private life to optimise process and decision-making (Brynjolfsson and McAfee, 2016; Burstrom et al., 2021; Moussawi et al., 2021). The latest advancements have primarily focused on machine learning and deep learning to generally extend traditional, data-driven AI tasks, such as predictions, classifications, or recommendations toward the generation of unique, realistic, and creative content. However, previous Gen AI research cannot specifically encompass the studio-based culture, which distinctively features time-consuming graphic expressions of spatial analysis and interpretation. This case study investigates how Gen AI interventions can assist studio-based students in rapidly generating diverse design options, thereby facilitating more efficient problem-solving for real or hypothetical design challenges.

This case study also aligns with XJTLU Three-Year Strategic Planning for Academic Affairs 2024-2026, ​, particularly ​Strategic Area Four (SAF): Educational Innovation, Enhancement, and Reform. SAF emphasises initiatives such as curriculum review and enhancement, including the ​AI-Enhanced Module Project​ across all schools at XJTLU. The case study focuses on the ​Skills for Planning and Design Practices (SPDP)​​ module, exploring how advanced Gen AI tools—such as ​Photoshop AI (PS AI) and AI-powered chatbots—can enhance studio-based learning. This initiative has been developed through extensive collaboration with colleagues from the ​Educational Development Unit (EDU) and the Management Information Technology and System Office (MITS)​​ at XJTLU.

1.2         Case objectives

This case research has three objectives.

• To help students understand skills in spatial analysis and interpretation by employing Gen AI;
• To identify challenges during the application process of Gen AI;
• To help students better develop skills of Gen AI for graphic expression and communication

1.3         Central question

The central question is ‘How can the integration of PS AI and XIPU AI Chatbot into a year 2 design module impact the studio-based design culture, and what challenges and solutions emerge in the process of developing and deploying these applications within the higher educational context?”

1.4         Research framework

The entire practice can be conducted based on the summarised research framework as shown in Figure 1.

Figure 1: Research Framework

2         Case description

This case study explores the integration of Photoshop AI (PS AI) and the XIPU AI chatbot into teaching activities and assessments within the Skills for Planning and Design Practices (SPDP) module. The innovation aims to enhance the authenticity of learning tasks and align with industry expectations, thereby improving students' employability. The approach is informed by the spiral action research methodology proposed by Kemmis and McTaggart (2004), which focuses on iterative stages of planning, acting, observing, and reflecting. This methodology provides a structured framework for improving educational practices (Clark et al., 2020). Figure 1 illustrates the key stages of this case study using Kemmis and McTaggart's spiral action research model.

Figure 1: An illustration of stages of the case using the spiral action research model (Kemmis & McTaggart, 2004)

2.1         Planning Phase

In the initial planning stage, a key problem was identified: students struggled to complete an assessment task that required using Photoshop to create a collage of landscapes derived from movie screenshots (see Appendix 1 for detailed requirements). The steep learning curve of Photoshop often hindered students’ ability to achieve proficiency within the module's timeframe. Simultaneously, industry trends and university strategies increasingly emphasized integrating AI tools into teaching.

To address these challenges, the team decided to incorporate PS AI into the assessment. Standard Photoshop (PS) provides a comprehensive suite of tools for image editing, manipulation, and design, but it relies heavily on the user’s manual skills and creative input to achieve results. In contrast, Photoshop AI (PS AI) incorporates artificial intelligence–driven features such as Generative Fill, Generative Expand, and AI-assisted object selection. These tools allow users to automatically generate, remove, or extend image content in ways that would be time-consuming or technically complex with traditional methods. By combining user prompts with AI-powered outputs, PS AI streamlines creative workflows, reduces repetitive manual tasks, and opens up new possibilities for rapid ideation and industry-aligned design practices. An illustration of the PS AI interface is shown in Figure 2. The interface includes a textbox where users can describe their image editing requirements.

Figure 2: Illustration of PS AI (Adobe, 2025)

In addition to transitioning from standard Photoshop (PS) to its AI-augmented tools, the assessment was redesigned as an AI-enhanced task to better align with contemporary industry practices. The revised assignment maintained its focus on creativity and analysis generative AI as a powerful tool within the workflow. An implementation plan was finalized in May 2024, targeting the CW1 assignment in week 5.

The integration of PS AI fundamentally shifted the assessment's focus from technical execution to strategic conceptualisation and critical analysis. In a traditional Photoshop task, a significant portion of a student's effort and grade is often tied to their manual dexterity with the software—their skill with the clone stamp, brush tool, and layer masks to create a seamless composite. The introduction of Generative AI disrupts this model. The tool automates the technical challenge of photorealistic compositing, allowing students to generate multiple visual iterations from a text prompt in minutes.

This automation forces a pedagogical pivot. The assessment is no longer about whether a student can technically assemble a collage, but rather how effectively they can direct the AI to produce a result that embodies a sophisticated conceptual idea. The core intellectual work moves "upstream" to the quality of the student's initial analysis of the film's spatial characteristics and the clarity of their creative direction (the prompts they write). The critical thinking is then tested "downstream" in their written reflection, where they must justify their selections and evaluate the AI's output against their original intent.

Therefore, the assessment criteria had to evolve accordingly. By evaluating the "suitability of visual content" and the "alignment between the collage and critical statement," the assessment now prioritises skills that are increasingly vital in the industry: creative direction, prompt engineering, and the ability to critically evaluate AI-generated content.

The AI-enhanced assessment is detailed below:

Objective of the assessment:

1. Developing graphic expression and communication skills.
2. Articulating the identity of a place.
3. Representing urban contexts by reflecting spatial characters.
4. Interpreting and further representing contextual analysis and detailed site analysis.

Steps required to complete the task:

1. Select a movie from a curated list in the CW1 brief.
2. Capture screenshots of urban scenes with distinctive spatial characteristics.
3. Use PS AI to create a collage incorporating elements from these screenshots.
4. Write a short (50–100 word) critical analysis reflecting on the photo collage.
5. Optionally, use PS AI for contextualizing existing street scenes or historical developments in CW2.

Assessments focused on four criteria: the quality of the AI-generated collage, suitability of visual content, graphic style, and alignment between the collage and critical statement.

To support students, two key technical measures were planned:

1. Ensuring PS AI software availability in computer labs (EB441 and EB369) during the summer as PS AI can recognize the network exit and users must connect to campus wired and wireless networks.
2. Offering training sessions on PS AI at the School/Department level during the first semester of AY24-25.

2.2         Act and Observe Phase

Students were introduced to the AI-enhanced assessment during the module introduction and CW1 brief. In week 2, the MITS officer provided guidance on enrolling in PS AI subscriptions. However, several challenges emerged during this phase:

• Enrollment Issues: Students encountered problems registering for PS AI due to incorrect regional settings (registration required Adobe ID region for the US, but some students chose other regions), which required intervention from Adobe's customer service.
• AI literacy: Students’ lack of skills in using the tool and creating effective prompts leads to errors or suboptimal outputs. For instance, while it successfully added a moon to an image, more complex edits were not achievable.
• PS AI’s limitation: There are instances where PS AI either disregards certain prompts or fails to process them correctly. This issue is not only indicative of the tool's current capabilities but also underscores the importance of students' AI literacy. Crafting high-quality prompts is essential for effective interaction with AI tools, and students need to develop this skill to maximize the utility of PS AI.

By week 5, fewer than half the students used PS AI for their collages, and the lack of AI literacy hindered their ability to meet assessment expectations. Some students, however, were able to create submissions that could meet the assessment’s requirements (see appendix II). These findings highlighted the need for alternative solutions and additional support mechanisms.

2.3         Reflection Phase

Key reflections identified the mentioned primary challenges that hinder students’ ability to complete that assessment. The teaching team recognized the need for timely, scalable support to address these issues, as immediate assistance from instructors was often unavailable. There is also a pressing need to enhance students' AI literacy to ensure they can navigate and benefit from PS AI.

 2.4         Revised Plan Phase

Based on the reflections and identified issues, the teaching team decided to create a customized chatbot using the XIPU AI Chatbot function on the Learning Mall page of the module. The chatbot was trained with comprehensive solutions to address frequently asked questions and detailed guides for using Photoshop AI. This allows students to interact with the chatbot at any time to receive step-by-step instructions and resolve their issues and questions on registration and creating effective prompts. See Figure 2 for an example of how the chatbot can assist students in producing high-quality work to fulfil the assessment requirements. In addition, the teaching team is also proactively seeking updates to PS AI and exploring alternative AI-enhanced professional design tools to empower students in creating the required assessment outputs.

Figure 3: An illustration of how XIPU AI Chatbot can assist students in creating AI-enhanced work

2.5         Act and Observe Phase

The teaching team has introduced PS AI and XIPU AI Chatbot to a year 2 theoretical module, titled Issues and Practice of Planning in China II (IPPC II). Students are expected to complete photo collage on posters by using PS AI and XIPU AI Chatbot. At the beginning of Semester 2, AY24-45, all students smoothly registered PS AI with the aid of XIPU AI Chatbot. In addition, XIPU AI has combined MidJourney (MJ) and DallE. Therefore, the EDU colleague was invited to introduce how to better use the two Gen AI tools for generating posters. In terms of submitted posters, most students have used Gen AI tools (see Appendix III). The feedback from students were positive.

 2.6         Future Plans

Some students reported that they want to use some other Gen AI tools due to many emerging options. The team will actively collect feedback on the revised approach. In addition, in response to the growing discussions and debates on the copyright issue of AI-generated content (Yuchen, 2024), the teaching team will incorporate discussions on ethical and legal considerations surrounding AI-generated graphics.

3         Discussion and Conclusion

This research represents a significant innovation in teaching practices by integrating cutting-edge AI tools into authentic assessments. By combining PS AI and the XIPU AI chatbot, the initiative not only addresses the challenges students face in mastering complex software but also aligns their learning experience with industry expectations in the first round of “plan, act and observe phase, revised plan” phase. Then, the further act and observe phase can verify the positive practice in a theoretical module and display its wider application. The action research methodology ensures a systematic approach to refining the innovation, with iterative feedback and collaboration driving its success. Moreover, this approach is highly transferrable from design modules to other teaching contexts. AI-enhanced authentic assessments and targeted support tools like chatbots can be adapted across disciplines to foster deeper learning, improve student engagement, and enhance employability skills.

Acknowledgement:

We would like to thank UPD year 2 students: Yanping Cong, Yixin Chen, Yichen Zhuang, Zixuan Liang, Ruihan Ju, Anyu Zhao, Siyuan Yinn, Jinladuo Guan for sharing the created photo collage and poster.

Reference:

Adobe (2025). Adobe Firefly. The ultimate creative AI solution. Retrieved on Septebmer 24th, 2025, from https://www.adobe.com/products/firefly.html
Brynjolfsson, E., McAfee, A. (2016), The second machine age: work, progress, and prosperity in a time of brilliant technologies, W.W. Norton & Company.
Burstrom, T., Parida, V., Lahti, T., and Wincent, J. (2021). AI-enabled business-model innovation and transformation in industrial ecosystems: a framework, model and outline for further research, Journal of Business Research, 127, 85-95, https://doi.org/10.1016/j.jbusres.2021.01.016
Clark, J. S., Porath, S., Thiele, J., & Jobe, M. (2020). What is action research for classroom teachers. Action Research.
Feuerriegel, S., Hartmann, J., Janiesch, C., Zschech, P., (2024). Generative AI, Business & Information Systems Engineering, Vol. 66, P111-126.
Kemmis, S., & McTaggart, R. (2000). Participatory action research. In N. Denzin & Y. Lincoln, (eds.), Handbook of Qualitative Research. London: Sage.
Moussawi, S., Koufaris, M., Benbunan-Fich, R. (2021). How perceptions of intelligence and anthropomorphism affect adoption of personal intelligent agents, Electronic Markets, 31(2), 343-364, https://doi.org/10.1007/s12525-020-00411-w
Yuchen, L. U. (2024). AI-Generated Content and Its Legal Status Under Copyright Law. Journal of Education, Humanities and Social Sciences, 35, 218-225.

Appendix I: Detailed requirements of photo collage

Exercise 4 (20%) photo collage that expresses the identity of the place

The characters of the living space 4 – Representation of urban context

Choose one movie recommended from below, make a collage of the spatial elements of urban context in the movie. A short statement must be included to help you describe your expression.

The recommended movies are:

·        A Street Cat Named Bob (2016)

·        Blade Runner (1982)

·        Decalcomanie (2019)

·        Farewell to My Concubine (1993)

·        Inception (2010)

·        La La Land (2017)

·        L’auberge Espagnole (2002)

·        Roman Holiday (1953)

·        Trueman’s World (1998)

Appendix II: Student samples of photo collage by using PS AI

Appendix III: Student samples of posters by using PS AI, MJ/DallE