Harnessing AI Empathy for Human-Centric Architectural Designs

By Sayjel Vijay Patel

Original illustration generated by the author with Midjourney

The best architects have a superpower; the ability to step in the shoes of the occupants of their buildings and spaces. Pretending to be a family, child, or person of determination, uncovers needs the might that otherwise would go unoticed.

Take, for example, the design approach of renowned Dutch architect Rem Koolhaas for the private residence “Maison Bordeaux”. While the project boasts many user-centric concepts, for me, the standout feature is the design of child-height “memory” windows. To an adult, their placement might seem unusual, but for children, they offer a unique and personalized view of the world — one connected to a specific time in their lives. This thoughtful decision that could have been overlooked, illustrates the power of creative role-playing in design.

This practice forms the heart of human-centered architecture — where empathy drives the design of building that are more responsive to the needs and experiences of the people who use them.

Creative Roleplaying meets AI

Today, AI brings a new dimension to this approach. With techniques like role prompting, architects can now use AI to simulate different perspectives — asking it to “pretend” it’s an architect, a safety inspector, or even a worried parent — and generate new design ideas based on those viewpoints. But we can go even further. Artificial empathy adds a deeper layer by enabling AI to simulate emotional perspectives, such as the anxiety of a parent concerned about safety or the excitement of a child exploring a playground [1].

While this empathy is artificial, its potential to inspire human-centered design concepts is profound.

Why it matters to designers

1. Better Spaces, Better Lives. By asking an AI to “think like a child worried about bullying,” it will focus on the design of a playground that provides sight lines for safety or quiet spaces for comfort. This reveals important insights that could be overlooked.
2. Quick Insights, Broad Possibilities. AI provides a fast feedback loop, offering ideas in minutes that might otherwise take months of discussion. While it’s not a substitute for talking to real people, AI provides a low-risk sandbox to test assumptions, and broadening your perspective.
3. Inclusive Design as a Learning Practice. By prompting AI to consider the experiences of someone with mobility challenges or someone sensitive to noise, it provides feedback on find design flaws and opportunities that we might have overlooked. This helps us learn to see differently — improving our empathetic skills as practitioners.

How to Use It Well

1. Start with a Role. Begin by framing the AI’s perspective. Example prompt: “Act as a parent worried about noise in an apartment building.” This establishes a clear context for the AI’s output.
2. Add a Dose of Empathy. Follow up with prompts that focus on emotional considerations. For example: “How might the parent’s fears about noise be eased?” This will lead to suggestions such as aacoustic materials, or designated play areas for children.
3. Mix in Real Feedback. Use AI-generated ideas as a starting point, but validate them with real users — parents, neighbors, or community members — to check if the concepts actually address their needs.
4. Repeat and Refine. Iterate on the AI’s output by tweaking prompts and incorporating new insights. The more context and detail you provide, the more accurate and relevant the suggestions become.

Example of using creative role-playing with ChatGPT to generate design concepts by assigning different roles to explore various aspects of a playground design.

A Final Thought

“Pretend empathy” can’t capture the nuance of firsthand conversations or the depth of real human experiences. Instead these tools can complement — not replace — traditional methods of engaging with the people.

AI is a tool — a powerful one — but it works best as part of a cycle that starts and ends with people.

Just as Rem Koolhaas used child-height windows to reveal a child’s-eye view of the world, these tools help us see hidden perspectives in everything from playgrounds to apartments. Yes, the AI’s empathy is “pretend,” but it can spark creativity and deeper insight.

N O T E S

1 For a deeper exploration of artificial empathy and its role in user-centered design, take a look at the article “Toward Artificial Empathy for Human-Centered Design: A Framework” by Jianxi Luo and Qihao Zhu. It offers insightful perspectives and a detailed framework on this emerging field.

About the author

Sayjel Patel is the CTO and Co-Founder of Digital Blue Foam (DBF), dedicated to transforming the Architecture, Engineering, and Construction (AEC) industry for the digital era.