AI in Architecture
ZHA Director Nils Fischer explores how AI is transforming architecture, why experimentation matters, and how technology can empower architects to create better buildings rather than replace them.
Episode Summary: AEC Innovators Circle Podcast, Episode 1
AI in Architecture
Guest
Nils Fischer
Director at ZHA (formerly Zaha Hadid Architects) and one of the founding forces behind CODE, the practice's computational design research group. Over more than two decades, Nils has helped pioneer new approaches to computational design, custom software development and architectural innovation, exploring how technology can expand the possibilities of architecture while maintaining creativity at its core.
Editor's Note: Since this episode was recorded, Zaha Hadid Architects has transitioned its brand name to ZHA across its communications. References in this summary reflect the firm's current branding while preserving the context of the original discussion.
Episode Overview
Artificial Intelligence has become one of the most talked-about topics in architecture, but according to Nils Fischer, the industry's AI journey started long before ChatGPT, Midjourney and today's generative design tools.
In the opening episode of the AEC Innovators Circle Podcast, Nils shares the story behind CODE, ZHA's computational design research group, and explains how the practice has spent more than two decades developing custom software, computational workflows and advanced digital design techniques to solve challenges that conventional architectural tools could not address.
The discussion explores how architectural practices can adopt emerging technologies without losing their creative identity. From custom software and generative AI to data strategy, digital twins, machine learning and large language models, Nils provides a practical perspective on where AI is delivering real value today and where the industry's expectations may be running ahead of reality.
While much of the conversation focuses on technology, the underlying message remains fundamentally human. Nils argues that AI will not replace architects. Instead, it will remove repetitive tasks, increase the number of design options that can be explored and allow architects to spend more time doing what they do best: making informed creative decisions.
Ultimately, this is a discussion about the future of design itself, and how architects can use AI to amplify creativity rather than automate it.
Key Insights
The Best Innovations Start With Real Problems
CODE was not created because ZHA wanted to become a technology company.
It emerged because architects were repeatedly encountering problems that existing software could not solve. The team began creating small tools to handle geometry, automate processes and improve project delivery, eventually building an entire research function around those capabilities.
Key takeaway: Innovation is most valuable when it solves real project challenges.
AI Is Only The Latest Chapter In Digital Design
While AI dominates today's conversations, ZHA has been experimenting with computational design, automation and machine learning concepts for years.
Nils argues that generative AI represents the latest step in a much broader journey rather than a sudden technological revolution.
Key takeaway: AI is an evolution of digital design, not a completely new discipline.
Good Data Creates Better AI
One of the strongest themes throughout the conversation is the importance of data quality.
AI models can only provide meaningful outputs when they are supported by structured, well-managed information. Practices looking to leverage AI successfully must first invest in organising and understanding their own data.
Key takeaway: The quality of your AI outputs is determined by the quality of your data.
Experimentation Must Be Built Into The Business
Rather than relying on a dedicated AI department, ZHA encourages experimentation across the organisation.
People are given time to explore tools, test workflows and share their findings with others. Successful ideas are adopted organically based on their usefulness rather than through top-down mandates.
Key takeaway: Innovation happens when people are given permission to experiment.
Don't Fall In Love With Tools
Technology changes too quickly to build long-term strategies around individual platforms.
Nils recommends focusing on techniques, workflows and problem-solving approaches rather than becoming attached to any single product or vendor.
Key takeaway: Learn principles, not products.
Generative Images Are Only The Beginning
While image-generation tools have attracted significant attention, Nils believes their impact on architecture is often overstated.
The real opportunity lies in AI's ability to understand relationships, structure, hierarchy and design logic rather than simply creating visually appealing imagery.
Key takeaway: The future of AI in architecture is about understanding buildings, not just rendering them.
AI Will Improve Productivity Through Repetition
The biggest immediate gains from AI are likely to come from repetitive, process-driven tasks.
Schedules, specifications, documentation, classification, information management and other administrative activities all present opportunities for automation.
Key takeaway: AI should eliminate repetitive work so architects can focus on design.
Architects Will Become More Important, Not Less
One of the most interesting arguments from the episode is that AI may actually increase the importance of architects.
As machines become capable of generating more options and processing more information, the value shifts towards the people who can evaluate those options, provide context and exercise judgement.
Key takeaway: AI can generate possibilities, but architects still decide which ideas matter.
Better Design Is The Real Opportunity
Rather than focusing solely on efficiency gains, Nils sees the greatest potential in improving design quality.
If repetitive activities can be automated, architects gain more time for iteration, exploration and creative refinement.
Key takeaway: AI has the potential to make architecture better, not just faster.
AI Can Democratise Architectural Knowledge
As urban populations continue to grow, particularly in developing regions, AI may help make architectural knowledge more accessible.
Nils believes AI could help transfer expertise, improve access to design intelligence and support better decision-making in parts of the world where specialist knowledge is less readily available.
Key takeaway: AI could significantly expand the global reach of architectural expertise.
Quote From The Episode
"Mediocrity is now free. But you still have to be able to judge what is truly brilliant."
What AEC Leaders Should Focus On
Based on the discussion, organisations looking to maximise the value of AI should:
- Invest in structured and accessible data
- Create time for experimentation and learning
- Focus on outcomes rather than technology trends
- Automate repetitive and low-value tasks
- Protect intellectual property and sensitive information
- Build AI literacy across the organisation
- Develop strong critical thinking and decision-making skills
- Use AI to enhance creativity rather than replace it
Listen To The Full Episode
In Episode 1 of the AEC Innovators Circle Podcast, Karl Smith sits down with Nils Fischer to discuss AI, computational design, innovation, data strategy, automation and how architecture may evolve in the age of artificial intelligence.
Watch the full episode above and discover why the most successful architects of the future may not be those who resist AI, but those who learn how to work alongside it.
