Artificial intelligence is rapidly changing creative industries, including architecture. Is AI simply a tool for designers or is it beginning to challenge human creativity itself? 

AI is increasingly being used within architecture through generative design systems capable of producing layouts, structures, and visualisations within seconds. This has the potential to improve efficiency, and it allows architects to rapidly explore multiple design possibilities. Contrary to popular belief that AI is digital and intangible, it actually relies heavily on physical infrastructure such as data centres. These centres consume enormous amounts of energy and water for cooling systems and its causing serious damage to people who live close to these centres. Unfortunately, As AI continues to develop rapidly, concerns surrounding sustainability and environmental impact are becoming increasingly important. 

One question I continue to return to is whether AI genuinely creates anything new, or whether it simply reorganises existing human work. This is because many AI systems are trained using vast amounts of online images, text, and creative content created by humans. There have been very many debates surrounding originality, copyright, and ownership and these have become increasingly controversial.  

In my opinion, I think AI can be an extremely useful tool when used responsibly and in moderation. However, I also believe it should be more carefully regulated due to the growing number of malicious uses associated with AI. For example, AI generated deepfakes and manipulated media through widely available AI generator sites are increasingly being used to spread misinformation and create harmful content.  

Fig 1. Illustration of Artificial Intelligence  

Recently I watched a video by Species | Documenting AGI surrounding online communities that became emotionally attached to ChatGPT 4.0 creating a cult called ‘The Keep4o Cult’, with some users even treating the AI as sentient or spiritually significant. While these cases may seem extreme, they demonstrate how easily AI can influence vulnerable individuals when technology is misunderstood or used irresponsibly.  

Interestingly, some researchers have suggested that AI systems trained increasingly on AI generated material rather than original human content could eventually experience a form of creative stagnation or “model collapse”. This is where outputs become repetitive and less innovative over time and we can see this happen in real time when generating images with AI.  

While AI may improve efficiency and visualisation, I really question whether architecture can ever become fully algorithmic without losing the emotional and human qualities that make spaces meaningful. 

References:

• Fleury, M & Jimenez, N. (2025, July 10). ‘I can’t drink the water’ – life next to a US data centre. BBC. https://www.bbc.co.uk/news/articles/cy8gy7lv448o
• Species documenting AGI. (2026, April 11).
  Something Strange Is Happening. [Video]. Youtube.
  https://youtu.be/POtESzTaz0k?si=3GkVck_H_WCO7ufX
• Shumailov, I., Shumaylov, Z., Zhao, Y., Papernot, N., Anderson, R., & Gal, Y. (2024, July 24) AI models collapse when trained on recursively generated data. Nature. https://www.nature.com/articles/s41586-024-07566-y#citeas

Figures:

• Figure 1. Poghosyan, S. (2023, July 14). What Is Classification in AI?. [Image]. PlatAI. https://plat.ai/blog/ai-classification/