Background
This reflection is based on my tutorial discussion following Intervention 4.
The experiment was coordinated by me and involved one traditional designer, one AI-assisted designer, and a group of audience participants.
Due to technical limitations, the two designers could not share their screens simultaneously. As a result, I had to take on a dual role — both coordinator and narrator — constantly switching between the designers’ processes, explaining the progress to the audience, and maintaining engagement.
Although this technical issue increased the experiment’s complexity, it also revealed a crucial insight: in the process of AI collaboration, the human mediator plays an indispensable role. It is precisely this human presence that makes technology more comprehensible, relational, and emotionally engaging.
Summary of Tutor Feedback
My tutor first acknowledged the clarity of my methodology and the logical structure of my report.
She noted that the distinction between Phase 1 (Individual Creation) and Phase 2 (Collaborative Co-Creation) was clear, effectively demonstrating the research transition from “observation” to “participation.”
She suggested that I further strengthen the critical depth of my analysis in the following areas:
- Introduce more diverse — even opposing — perspectives, incorporating critical or skeptical views of AI-generated design;
- Enhance triangulation by cross-referencing interviews, behavioural and emotional data, and visual analysis;
- Reflect on the evolving relationship between myself and AI, considering how the creative process reshaped my own understanding.
She also advised me to include videos, visual materials, and on-site documentation in my blog for transparency, while keeping the report version more analytical and concise.
Learning Reflection
Through this intervention, I realized that AI not only transforms the design process but also reshapes the audience’s role — from passive observers to active co-creators.
In the first phase, the audience was astonished by AI’s speed and visual diversity. They frequently asked questions and expressed curiosity.
In the second phase, when both designers began responding in real time to audience feedback — such as adjustments to lighting, posture, or skin tone — the atmosphere became increasingly interactive.
Participants began to show emotional engagement and even developed a sense of authorship, feeling that they had “participated in creation.”
This phenomenon reminded me of Attribution Theory (Heider, 1958), which suggests that people tend to assign higher value and authenticity to outcomes they perceive as stemming from human effort and intention.
Although the visuals were generated by AI, once audiences recognized the designer’s thought process and judgment within them, their trust in the final result increased.
Thus, this experiment not only compared AI and human creation visually, but also revealed the psychological mechanisms linking automation and trust.
In our tutorial discussion, my tutor and I also explored the concept of empowerment.
She encouraged me to consider: Who is being empowered by AI?
I gradually realized that this empowerment operates on multiple levels:
- For designers, AI enhances efficiency and expands creative boundaries;
- For audiences, AI opens new pathways for participation and expression;
- For brands, it enables enhanced creative expression while reducing production time and cost.
However, empowerment also implies a redistribution of creative control.
Methodological Reflection
At present, I have collected data from around ten participants, primarily qualitative in nature.
My tutor emphasized that in Action Research, the depth of data is more important than its quantity.
Therefore, in future analysis, I plan to apply Content Analysis to identify patterns within my qualitative data.
Although my original plan was to conduct full triangulation, the limited sample size led me to adopt a partial triangulation approach — combining verbal responses with non-verbal emotional cues such as laughter, nods, and eye contact.
While non-quantitative, this method revealed subtle emotional nuances that numerical data alone could not capture.
Interestingly, the audience repeatedly raised questions about authorship and copyright during the experiment.
I plan to invite a copyright lawyer and an AI project management expert for follow-up interviews to expand the discussion on legal and ethical responsibility in AI-assisted creation.
Conclusion
I realized that the true meaning of innovation lies not only in efficiency, but in how technology redefines human relationships.
The AI-driven collaborative process allowed me to see that when human ethics and emotion remain the guiding principles, authentic warmth can coexist with automation.
Ultimately, this reflection helped me rediscover the core of my project — it is not only about AI design, but about re-understanding the conditions and meanings of human co-creation in the algorithmic age.
Leave a Reply