Generative AI in Creative Education – Opportunities and Risks

Novel Artificial Intelligence (AI) tools, such as ChatGPT, promise to boost human creativity in different domains, such as fashion design or creative writing. However, little is known about how these tools really impact students in creative domains (e.g., from Academy of Arts) as well as their creative process. Hence, in our study, we, Livia Müller, Hanieh Aslani, and Prof. Dr. Thiemo Wambsganss from the Human-Centered AI Learning Systems (HAIS) Lab at BFH Business School, and Prof. Dr. Jimmy Schmidt, conducted several empirical workshops with students and educators at the Bern Academy of Arts (HKB) to explore how these creative practitioners experience AI, what they think about it, and how they work with it in their creativity processes. Our results raise interesting questions about teaching practices, ownership, and originality of creative work.

Creative Education and Generative AI

In recent years, a growing number of creativity support tools have emerged that enhance human creativity in different domains, such as fashion design (Davis et al., 2025), product development (Kim et al., 2025), or writing (Qin et al., 2025). These systems can improve performance and ideation outcomes in the creative process, especially with the integration of AI. However, in the creative education context (meaning creative practitioners at art schools), these tools raise novel questions, as learning in creative education is grounded in experimentation, reflection, and process thinking, for example, in visual arts, product design, or creative writing.

Exploring Perspectives in Creative Education

Figure 1: Overview of the study methodology, showing data collection through educator interviews and student workshops, followed by thematic analysis and interpretive framing

Hence, we conducted a study at BFH to explore how students and educators at HKB perceive, experience, and adapt to AI in their creative workflows and learning environments. We considered the capabilities of AI and creativity support tools, as well as the educational settings in which they are embedded. Through seven interviews with educators and three participatory workshops with 36 students from HKB, we examined how these groups understand questions of authorship, user control, and teaching practices in the light of AI. We followed a qualitative approach (Blandford et al., 2016) in our research as illustrated in the research design above in Figure 1. The data collection consisted of two complementary parts: (1) 7 Semi-structured interviews with educators, focusing on pedagogical practice and institutional challenges, lasting between 33 and 55 minutes (M = 43.2, SD = 10.6; and (2) participatory workshops with 36 students (aged 18 to 30, M = 22.2, SD = 2.2), focusing on creative processes, challenges, and how AI might support them in the future.

AI raises questions about teaching practices in creative education

The introduction of AI into creative education raises concerns among educators about the impact on existing teaching roles and educational values. As institutional conditions and familiarity with AI among educators differ, students often adopt AI tools more quickly. However, students report challenges to this, as they have to learn tools on the fly without sufficient guidance. This may lead to behavior where students rely too much on AI and critical reflection gets lost in the creative process. Educators also raise concerns that AI tools often encourage speed over reflection in creative work: students move quickly through AI-generated suggestions, no longer trust their own ideas, and diminish critical reflection in the process, important values in creative education (“Students stopped trusting their ideas (…) no reflection, no pause.”). Figure 2 illustrates the different thematic clusters from the workshop activities, showing the different topics raised by educators and students.

Figure 2: Thematic clusters from workshop activities, showing the different topics raised by educators and students

Mixed feelings about originality and ownership of creative work with AI

Both students and educators also raised concerns about originality and ownership of creative work with AI. Some expressed that AI only reproduces existing patterns and work, which raises the question of what still truly belongs to them when AI is involved. The output of AI is often associated with uniformity and predictability rather than creative novelty, whereas educators try to encourage students to think outside of the box (“[AI] works with what is probable. . . and we try to encourage people to think outside the box”). The students, however, are aware of these problems; they state that AI suggestions always need to be filtered through their own judgment to avoid losing originality. Further, both groups have raised ethical concerns with the use of AI, such as uncredited training data, biases in the training data, and the environmental impact of AI usage, leading to discomfort (“I cannot ignore this whole political issue that comes from the use of GPT or AI (…) none of this is really solved.”). Rather than rejecting AI, the students called for clear guidance and careful use to protect originality, ownership, and diversity in creative work (“offer guidance without replacing authorship or experimentation”).

What this means for Educators

For teachers and educators in creative education, our study offers some important takeaways:

• AI creativity tools can be valuable for early-stage work such as brainstorming and ideation. However, current tools optimized for speed and results may undermine students’ iterative and critical thinking processes in the creative process.
• The authorship and contribution of students and AI should be transparent in order to assess where student input, AI suggestions, and collaboration intersect.
• Reflective practices should be embraced in the creative process to encourage students to explain their reasoning behind using AI tools.

Recommendations for Students in using AI-tools

For students, our research provides the following insights:

• AI tools can help overcome creative blocks and generate ideas quickly. However, it is important to reflect on design choices in order to develop creative skills and critical reasoning.
• Creative contribution is important and should be made visible in creative work. The contributions of both students and AI should be articulated to understand their respective roles in the creative process.

Bibliography

Blandford, A., Furniss, D., & Makri, S. (2016). Qualitative HCI research: Going behind the scenes. Morgan & Claypool.

Davis, R. L., Mwaita, K. F., Müller, L., Tozadore, D. C., Novikova, A., Käser, T., & Wambsganss, T. (2025). SketchAI: A ‘Sketch-First’ Approach to Incorporating Generative AI into Fashion Design. Proceedings of the Extended Abstracts of the CHI Conference on Human Factors in Computing Systems, 1–7. https://doi.org/10.1145/3706599.3719782

Kim, H.-J., Kim, J., Jeong, S., Lee, M., Choo, J., & Kim, S.-H. (2025). ShoeGenAI: A Creativity Support Tool for High-Feasible Shoe Product Design. Proceedings of the Extended Abstracts of the CHI Conference on Human Factors in Computing Systems, 1–11. https://doi.org/10.1145/3706599.3721204

Qin, H. X., Zhu, G., Fan, M., & Hui, P. (2025). Toward Personalizable AI Node Graph Creative Writing Support: Insights on Preferences for Generative AI Features and Information Presentation Across Story Writing Processes. Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems, 1–30. https://doi.org/10.1145/3706598.3713569