Currently, I am pursuing my second master's degree in the Design and Environmental Analysis program at Cornell University. I invite you to join me in my world of design and photography utopia.
Linghao Li ｜李凌昊
PhD in Design
MA Design in D+EA ‘24
MA Graphic Design and Visual Experience ‘22
BFA Visual Communication Design ‘16
NCSU | Cornell University｜SCAD｜TAFA
+ 1 912-391-7213 | ll933@cornell.edu
Behance / Instagram / Linkedin / Mail
Design Works
Photography Works
Research Topic
About Myself
Resume/CV
Building upon my foundation in design theory and art education, I have significantly broadened my research interests. During my time at Cornell University's Human-Centered Design Department, I had the privilege of systematically exploring Pluriversal Design under the guidance of Dr. Renata Leitão. Concurrently, I pursued a minor in Anthropology, mentored by Dr. Viranjini Munasinghe. This diverse academic experience, encompassing Cultural Anthropology, Visual Studies, History of Photography, and Design for Interaction, propelled me into the realm of interdisciplinary research and helped me define my unique research trajectory.
A pivotal moment in solidifying my academic direction came through my studies with Professor Andrew Moisey. His deep insights into visual studies and photography profoundly influenced my approach to understanding how visual culture shapes and reflects societal narratives. His mentorship has been instrumental in affirming my commitment to pursuing visual studies as the core of my future academic career.
Presently, my research focuses on the intersection of visual culture, design, art history, and anthropology. I am particularly fascinated by the history of photography and visual media in early 20th-century Northeast Asia, especially in relation to how these mediums influenced socio-political narratives during periods of colonization.
My passion for these topics drives my commitment to advancing knowledge in these areas. I invite you to explore more about my work and research interests in the following messages. This dedication fuels my pursuit of creating meaningful contributions to the understanding of visual culture and design in East Asia. Welcome to my academic journey.
- International Association of Societies of Design Research Congress
Wangda Zhua, Linghao Lib , Seryung Kimc
ᵃ Hong Kong Polytechnic University, Hong Kong, Hong Kong
ᵇ North Carolina State University, Raleigh, United States
ᶜ Georgia Institute of Technology, Atlanta, United States
*wz334@cornell.edu
In Fall 2024, a three-week Generative AI (GAI) module was introduced in an undergraduate digital media course within a design program at a research-intensive public university in the U.S. This pictorial presents a unique approach to integrating GAI into a design studio through a proposed design education perspective, showcasing student work. Students first learned to use a GAI tool developed by the author, designed with a simplified UI and interactive design provenance, using a Stable Diffusion model. They practiced writing structured prompts to generate novel and appropriate design scenarios. In the second week, students were tasked with designing chairs, creating design scenarios using the AI tool, and translating them into design artifacts using AutoCAD. In the third week, students received feedback and refined their final chair designs. This pictorial explored a GAI design education paradigm, transitioning from scenario space to artifact space.
This pictorial proposed a model of integrating GAI into design education, through a proposed design paradigm from speculative design, which frames design as a progression from scenario space to artifact space (Østvold Ek et al., 2024). In this approach, students begin by generating design scenarios using GAI, crafting structured prompts that guide the GAI's output. These scenarios function as conceptual starting points that inspire the development of applicable artifacts, translated into form through CAD tools.
A Generative AI design tool and a three-week curriculum focused on outdoor chair design were developed and implemented in an undergraduate digital media course with 29 students at a research-intensive public university in the U.S. To assess the creative outcomes, three experienced design instructors evaluated each submitted artifact based on novelty and utility. This pictorial presents four representative student submissions.
By examining this integration of Generative AI into design education, this pictorial explores the evolving relationship between AI, design paradigms, and pedagogy. It raises critical questions about the future of design education, such as: How does AI reshape the design process and the role of designers? Did GAI Enhance Design Creativity? etc.
This study employs a case study methodology to ensure methodological transparency. Beyond the pedagogical framework outlined above, the research systematically documented prompt engineering strategies, iterative workflows, and instructor evaluation protocols. The novelty-utility assessment framework, grounded in creativity theory (Guilford, 1967), enabled comparative analysis across student outcomes. While prioritizing contextual documentation over statistical comparison, this pictorial contributes empirical observations about prompt specificity and creative outcomes that complement quantitative GAI design education research. Future studies could employ mixed methods to enhance generalizability while preserving contextual depth.
The interplay between scenario space and artifact space is a foundational concept in design theory, particularly within the field of human-computer interaction (HCI). Scenario space refers to the breadth of user needs, goals, and contexts in which a system may be employed, while artifact space encompasses the range of possible technical solutions and design alternatives that can address those needs (Morris et al., 2023). The task-artifact cycle, as articulated by Ralph (2015), posits that user tasks and technological artifacts co-evolve in an iterative process: the introduction of new artifacts enables novel user activities, which in turn generate new requirements and inspire further design innovation. Design space analysis extends this perspective by providing a systematic approach to exploring both scenario and artifact spaces, thereby supporting designers in evaluating trade-offs and identifying optimal solutions (Morris et al., 2023). In the context of design education, engaging students with both scenario and artifact spaces fosters critical thinking and creativity, as learners are encouraged to empathize with diverse user contexts and iteratively prototype solutions (Kolko, 2010). The emergence of generative AI tools further amplifies this paradigm by enabling rapid exploration of artifact space, allowing students to visualize and test a broader array of design alternatives in response to complex, real-world scenarios (Cai et al., 2023). This integration of scenario-based inquiry and artifact generation not only aligns with constructivist pedagogies but also prepares students for the dynamic, technology-driven landscape of contemporary design practice (Laurillard, 2013). Thus, the scenario space to artifact space framework provides a robust foundation for cultivating adaptive, reflective, and innovative designers in the age of generative AI.
Design Creativity
Creativity is commonly defined by two key dimensions: novelty and utility (Guilford, 1967). While terminology varies—some scholars use “originality” instead of novelty and “usefulness” instead of utility—the underlying concepts remain consistent. However, debates persist regarding additional elements that contribute to creativity. Simonton (2012) proposed a three-criterion model, adding surprise as a necessary component, based on the U.S. Patent Office evaluation standard. Later, in 2017, researchers suggested a fourth element, aesthetics, challenging traditional definitions of creativity (Acar et al., 2017). Despite these discussions, the fundamental criteria of novelty and utility remain central to evaluating creative outputs.
Design education is essential for fostering creativity, and the design process itself plays a central role in this development (Wang, 2011). Traditionally, design has been modeled as an iterative movement between problem space and solution space, aligning with frameworks like the double diamond model (Rugman & D Cruz, 1993) (Figure 1). Within this context, speculative design, introduced by Dunne and Raby (2013), encourages conjectural thinking and the exploration of alternative futures. This approach is particularly relevant for integrating GAI into design education, as GAI's affordance in generating future scenarios complements the speculative design methodology (Østvold Ek et al., 2024). By structuring the design process as a transition from scenario space (AI-generated speculative concepts) to artifact space (realized designs) (Figure 2), students can engage with AI not just as a tool for execution but as a collaborator in ideation and innovation.
Figure 1. Double Diamond model & Speculative design process.
Figure 2. Up: Conceptual diagram of the tool: 1. Creation space for generating images by prompts and operations; 2. Process space for recording generated images and prompts; 3. Information table for showing the related parameters of the selected images. Bottom: Main User Interface (UI)
The URL to this tool is https://gai-npaz.vercel.app/.
Invitation code: ygdJ2U3y
The URL to this tool’s tutorial is https://youtu.be/-mcwD837p9o
Figure 3. The UI and flow of the tool.
I assigned students a design challenge:
Design an outdoor chair for a historical museum garden (Figure 4). The chair should accommodate people of all ages and sizes, including museum visitors, neighbours, and city explorers looking for a place to rest, have coffee, read, or relax. When designing, students need to consider usability, comfort, durability, safety, and, most importantly, creativity. They were free to explore any materials and manufacturing techniques.
Figure 4 Historical museum garden.
To guide their exploration, I provided an open-ended prompt guidebook that encouraged them to describe their scenario space across multiple dimensions, such as context, material, colour, and user experience (Figure 5). Instead of limiting their thinking strictly to the concept of a "chair," I offered an example that expanded beyond conventional definitions, prompting them to approach the design with a broader and more imaginative perspective, as depicted in Figure 6.
Figure 5 Up: Open-ended prompt guidebook for students; Bottom: A flexible prompt, selected tokens from the guidebook.
Figure 6 The prompt avoided to mention “chair” to increase the flexibility.
Peers Evaluation
I guided students to evaluate their design scenarios using two key dimensions: utility and novelty. Utility assesses whether the design meets fundamental requirements, such as comfort, safety, durability, feasibility, and suitability for a museum garden setting. Novelty, on the other hand, is more subjective—it considers whether the design feels surprising or unique.
In class, students evaluated each other's designs by adding emoji reactions on a shared Miro board, using thumbs-up for utility and stars for novelty (Figure 7). The top three students who received the most votes shared their insights with the class. One student, for example, explained their approach:
"I started with a key word I wanted and then gradually added elements—like ‘flower chair,’ then ‘add cushion,’ then ‘outdoor background.’ I also used the paint tool to enhance the flower petals. For another design, I started with a ‘tea party setting’ to establish a mood, then added ‘transform to flower chair’ to refine the concept."
Figure 7 A Screenshot for the Miro Board.
Workflow
Once students established a design scenario, the next step was to place their concept into context using Photoshop, allowing them to visualize how their chair would fit within the museum garden. From there, they transitioned to the artifact space, creating technical drawings in AutoCAD to bring their design to life.
Throughout the process, students had the flexibility to refine their ideas. If they realized that adjustments were needed, they could revisit Photoshop to modify the scenario or even return to the Generative AI tool for inspiration. There was no strict linear order—students were encouraged to iterate between scenario and artifact spaces until they were satisfied with their design. At each stage, their decisions were guided by the balance between utility and novelty, ensuring both functionality and creative expression. Figure 8 suggests this iterative movement across tools during the design process. Students spent two more weeks working independently with design, and submitted screenshots for the scenario creation process, a drawing of the design in context, and design techniques drawings.
After three weeks, three experienced design instructors independently evaluated the students' submitted designs, assigning comparative scores based on utility and novelty. Below, we present four representative designs that highlight different approaches and creative outcomes.
Figure 8 Tools used include GAI, Photoshop, and AutoCAD; Three corresponding submission images
Inspired by the garden setting, the student envisioned a chair that metaphorically "blooms," bringing the essence of nature into functional design (Figure 9). Initially, she experimented with various prompts, such as "a chair that looks like a flower," refining the idea until they found a compelling direction. As she progressed, she adjusted her prompts to make the flower petals more geometric, enhancing the design's feasibility for manufacturing while maintaining its artistic vision.
All three experienced design instructors rated this student’s work highly in both utility and novelty.
Figure 9 Submissions of Design-High Utility, High Novelty.
High Utility, Low Novelty
Another student designed a modular garden chair, emphasizing flexibility and adaptability (Figure 10). She experimented extensively, testing numerous prompts and iterating on generated images to refine her concept.
Her initial prompt was highly structured, specifying context, materials, function, and shape, which resulted in predictable outputs with less surprise. In her variations, she reinforced key elements by adding similar prompts to generate new images, leading to a more convergent design process that gradually honed and strengthened her final concept.
All three experienced design instructors rated this student’s work highly in utility but lacked novelty.
Figure 10 Submissions of Design-High Utility, Low Novelty
Low Utility, High Novelty
This student generated only a few AI variations but started with a highly open-ended prompt, focusing on experiential qualities rather than function (Figure 11). Instead of explicitly mentioning a chair, they described attributes like smooth, sculptural, soft, elegant, and light. While this approach led to a unique and artistic form, the final design did not integrate well with the historical garden setting and posed challenges in terms of maintenance and practicality.
Two instructors rated this design high in novelty but low in utility, while the third gave it low scores in both categories.
Figure 11 Submissions of Design-Low Utility, High Novelty
Low Utility, Low Novelty
The student initially envisioned a breathable chair for the community, but their prompts focused heavily on describing a chair made from natural materials (Figure 12). As a result, the AI-generated designs leaned toward personal, individual-use chairs rather than a seating concept suited for communal spaces. He did not change his mind as he did not come to the one-on-one meeting with the instructor in the middle of this design.
All instructors rated this design low in both utility and novelty.
Figure 12 Submissions of Design-Low Utility, Low Novelty
Did GAI Enhance Design Creativity?
GAI has the potential to enhance creativity, particularly by accelerating the early ideation process and allowing rapid concept iterations. However, the study also showed that the creativity of outcomes (i.e., utility and novelty) depended heavily on how students engage with the AI tool, which is aligned with previous research (Zhu et al., 2024; Zhu et al., 2025). Some students leveraged AI to generate highly novel and unexpected forms, using open-ended prompts to explore unconventional design ideas. Others approached AI as a refinement tool, crafting structured prompts that led to more predictable but functional designs. Those who crafted more open-ended, or less specific, prompts often generated highly novel results, while high utility was only consistently achieved when it was explicitly addressed within the prompt (e.g., student mentions “chair” in 5.1 and “bench” in 5.2). This points to a potential strategy to enhance design creativity: being deliberately vague about the visual form while being specific about function. In other words, students might benefit from creating what could be called a “verbal moodboard” —a way of describing the atmosphere, emotion, or feeling they want the design to evoke, without fixing its exact shape. This kind of prompt gives the AI room to explore a range of forms, often leading to more novel results, while still offering enough direction to support utility. Thus, while GAI supports creativity by enabling faster and broader exploration, its effectiveness ultimately depends on the designer’s capacity to engage with it critically. The most creative outcomes emerge when students combine AI-generated inspiration with critical design thinking, rather than solely relying on AI outputs.
Did We Teach Design Creativity?
Rather than directly teaching creativity as a fixed skill, this study created the conditions for students to explore and express creativity through their interaction with GAI. For example, the structured scenario-to-artifact process encouraged students to explore, evaluate, and refine their ideas iteratively, reinforcing key aspects of design thinking. The novelty and utility framework helped students assess creativity and reflect on their work. Peer evaluations using Miro further fostered a collaborative and reflective learning environment. More importantly, the study revealed that prompt design itself became a teachable skill. However, teaching creativity is not the same as enabling creativity—although the course provided tools and frameworks to help students navigate the creative process, some naturally embraced open-ended exploration, whereas others relied on more conventional design strategies. This suggests that creativity in this context was shaped as much by students’ individual design tendencies as by the tools they were given. Helping students learn to balance divergent (exploratory) and convergent (refining) thinking may be key to developing creative fluency in AI-assisted design.
Teaching Design Creativity by AI?
The findings suggest that AI may be an effective tool for teaching design creativity, but it should be positioned as a collaborator rather than a replacement for human ideation. However, students must critically evaluate and refine AI-generated ideas to align with user needs and design constraints.. Future AI-assisted design education could focus on developing better interfaces for AI-human collaboration, guiding students in prompt engineering, and incorporating more structured feedback loops to encourage deeper engagement with design thinking. Moreover, design education should move towards assessment based on process, not just outcome. As AI becomes more capable of generating high-quality outputs with minimal effort, judging students solely on the final product becomes less meaningful—not reflecting the student’s understanding, creativity, or growth. Focusing on how students engage with AI, such as prompts, iterations, and reflections, reveals far more about their creative development. Ultimately, the goal is not just to teach students how to use AI but to help them develop creative confidence and decision-making skills in an AI-augmented design process.
- Reference
Cai, A., Rick, S. R., Heyman, J. L., Zhang, Y., Filipowicz, A., Hong, M., Klenk, M., & Malone, T. (2023). DesignAID: Using Generative AI and Semantic Diversity for Design Inspiration. Proceedings of The ACM Collective Intelligence Conference, 1–11. https://doi.org/10.1145/3582269.3615596
Dunne, A., & Raby, F. (2013). Speculative Everything: Design, Fiction, and Social Dreaming. The MIT Press.
Fariz, M., Lazuardy, S., & Anggraini, D. (2022). Modern Front End Web Architectures with React.Js and Next.Js.
Guilford, J. P. (1967). Creativity: Yesterday, Today and Tomorrow. The Journal of Creative Behavior, 1(1), 3–14. https://doi.org/10.1002/j.2162-6057.1967.tb00002.x
Kar, A. K., Varsha, P. S., & Rajan, S. (2023). Unravelling the Impact of Generative Artificial Intelligence (GAI) in Industrial Applications: A Review of Scientific and Grey Literature. Global Journal of Flexible Systems Management, 24(4), 659–689. https://doi.org/10.1007/s40171-023-00356-x
Khawas, C., & Shah, P. (2018). Application of Firebase in Android App Development-A Study. International Journal of Computer Applications, 179(46), 49–53.
Kolko, J. (2010). Abductive Thinking and Sensemaking: The Drivers of Design Synthesis. Design Issues, 26(1), 15–28. https://doi.org/10.1162/desi.2010.26.1.15
Laurillard, D. (2013). Teaching as a Design Science: Building Pedagogical Patterns for Learning and Technology. Routledge. https://doi.org/10.4324/9780203125083
Morris, M. R., Cai, C. J., Holbrook, J., Kulkarni, C., & Terry, M. (2023, April 15). The Design Space of Generative Models. 36th Conference on Neural Information Processing Systems (NeurIPS 2022). https://doi.org/10.48550/arXiv.2304.10547
Østvold Ek, M., Paulsen, M., & Trondsen, J. K. (2024). Speculative Design Through the Lens of Ai. DS 131: Proceedings of the International Conference on Engineering and Product Design Education (E&PDE 2024), 627–632. https://doi.org/10.35199/EPDE.2024.106
Ralph, P. (2015). The Sensemaking-Coevolution-Implementation Theory of software design. Science of Computer Programming, 101, 21–41. https://doi.org/10.1016/j.scico.2014.11.007
Rugman, A. M., & D Cruz, J. R. (1993). The “double diamond” model of international competitiveness: The Canadian experience. Management International Review, 33(2), 17.
Simonton, D. K. (2012). Teaching Creativity: Current Findings, Trends, and Controversies in the Psychology of Creativity. Teaching of Psychology, 39(3), 217–222. https://doi.org/10.1177/0098628312450444
Wang, T. (2011). Designing for Designing: Information and Communication Technologies (ICTs) and Professional Education. International Journal of Art & Design Education, 30(2), 188–199.
https://doi.org/10.1111/j.1476-8070.2011.01675.x
Zhu, W., Guo, R., Wang, Y., & Xing, W. (2024). Designing with AI: Integrating Image-Generative AI into Conceptual Design in a CAD Class. ResearchGate. 2024 ASEE Annual Conference & Exposition. https://doi.org/10.18260/1-2--47146
Zhu, W., Xing, W., Kim, E. M., Li, C., Wang, Y., Yang, Y., & Liu, Z. (2025). Integrating image-generative AI into conceptual design in computer-aided design education. Educational Technology & Society, 28(3), 166-183.