++ Next Call for Entries, 1 Sept. ➜ Set Up Profile ++ Nominees Announced ➜ View

++ Next Call for Entries, 1 Sept. ➜ Set Up Profile ++ Nominees Announced ➜ View

Designers

Suryun Hyeon

Year

2025

Category

New Talent

Country

Korea, Republic

School

Korea University, Samsung Design Membership

Teacher

Soyoung Jeong

Please accept functional cookies to watch this video.

Three questions to the project team

What was the particular challenge of the project from a UX point of view?
The biggest challenge in the project was designing an interface that allows users to visually understand and modify the reasoning process of generative AI. Unlike conventional AI tools that only show the final output, we aimed to tackle the issue of opacity by making the AI's logic transparent and interactive. It was not just about giving users control over the results, but about helping them build an intuitive and cognitive connection with the AI’s internal process. The difficulty lay in creating a system that is both informative and non-overwhelming, enabling users to engage in a collaborative flow with the AI, without requiring technical expertise. Balancing complexity with clarity was our key design focus throughout the project.

What was your personal highlight in the development process? Was there an aha!-moment, was there a low point?
The most memorable moment was when I proposed the idea of visually revealing the AI’s reasoning before allowing users to modify the output. Rather than adjusting results through sliders or buttons alone, I focused on showing how the AI 'thinks' through intuitive visual structures. During a light user testing session, one participant said, 'Now I finally understand why the AI gave me this result,' and that feedback was incredibly rewarding. On the other hand, a challenging moment came when I tried to present too much information on a single screen—users began to feel overwhelmed rather than empowered. That experience reminded me to prioritize narrative flow over sheer data, leading me to simplify the visuals and refine the interaction.

Where do you see yourself and the project in the next five years?
In five years, I hope Moldable will grow into a flexible interaction system used across various AI services, helping more users understand and shape how AI works. Personally, I aim to deepen my role as a designer who builds intuitive, transparent interfaces that bridge people and AI.

XXX

Please first confirm that you are a human being.