A few years ago, the launch of Artificial Intelligence (AI) disrupted all kinds of jobs. ChatGPT, by OpenAI, launched in 2022 is created to assist and engage with humans naturally. A few months later, Perplexity entered the market. It was created to assist by summarising already available information on the web, quickly and efficiently. Further, in 2023 there were a lot of concerns around the safety and ethical concerns around Artificial Intelligence, that is when Claude was launched by Atrophic, it claimed to combine qualities from both Chat GPT and Perplexity. Moreover, it took a step further by claiming to be a more reliable, responsible, and transparent AI platform.
Today people are more open towards using AI and may look at it as that multi-talented assistant who assists in all kinds of tasks. For instance consider the case of an Architect, Jane, who works in an architectural firm whose focus is Modern and sustainable buildings, and gets a brief from a client, a tech company. They want to build a new headquarters, which focuses on sustainability and energy conservation. Based on this brief and a long 2-hour discussion, Jane wants to summarise it and streamline the entire process. Next, she wants to know some layouts, elements, and new building materials that will help her achieve the vision set by the tech company. Further, when the preliminary drawings are ready, and Jane is ready for the second meeting with the client, she asks AI to review them to ensure the design meets the project’s goals.
Therefore, to summarize, she needs assistance in the following ways and AI could be by any architect for the same :
- Synthesis from brief
- Idea generation
- Image Generation
- Review
But the question of which one to use remains. This article will do a side-by-side comparison by feeding the same prompts to all three (ChatGPT, Claude, and, Perplexity) and the responses and the results will be presented in the following format :
- The Prompt
- ChatGPT’s Response
- Claude’s Response
- Perplexity’s Response
- Winner
Synthesis from brief
ChatGPT’s Response
Claude’s Response
Perplexity’s Response
Winner: Perplexity
Reason: In this initial phase, it is important to have references to a variety of material to read through and visual data to refer to. Perplexity’s resources are a boon at this stage. Claude and ChatGPT gave a response that felt limited.
Idea generation
ChatGPT’s Response
Claude’s Response
Perplexity’s Response
Winner: Tie
Reason: All three platforms gave almost similar responses, the points, the way it was broken down and the length of the answer were identical in this case.
Image Generation
ChatGPT’s Response
Claude’s Response
Perplexity’s Response
Winner: Claude
Reason: A basic diagram explaining key elements to include will help architects remember the central idea of this project. ChatGPT produced an external render, which could be interpreted in many ways, and most importantly it is of little use without floorplans.
Review
Claude’s Response
Perplexity’s Response
Winner: Claude
Reason: This is a huge one, architects rely on their seniors and team members for reviews but what Claude did could be helpful for architects while iterating designs. The response of ChatGPT felt repetitive, i.e. it was blurting out the same information which it gave in the first two parts. Perplexity’s answer was almost identical to Claude’s in this case but Claude’s had an edge with the way it broke down and explained the areas of improvements.
Final takeaways
Overall winner: Claude
It was reliable throughout this entire research and could be used for either or all four parts (Synthesis from brief, Idea generation, Image Generation, and Review) where an architect needs assistance most commonly. Perplexity is advantageous in the initial research phase, with its references, links, and image stock. It could be used for pitches and presentations that have options for design styles and concepts. Finally, ChatGPT is used by most people due to familiarity, it gives an overall picture of the project, which aligns with the vision. But follow-up questions started giving repetitive answers which felt unreliable.
The future of Architectural Design is collaborative with Artificial Intelligence. Just like the invention of AutoCAD in 1982 and other Design tools like Sketchup, and Vray, the list goes on, to help ease the workload but did not replace architects’ hand sketches, block models, and visualization abilities. Although the AI platforms discussed here are not specific to the architecture industry. These AI platforms are getting integrated to optimize the workflow, assist with research, and review design.
To make the AI architecture industry-specific look like software like Revit integrated with AI could give humanity more energy-efficient buildings, which is a global need for today. AI with 3D modeling tools, architects would most space-efficient layouts. It took a while for people to accept and begin using Computer Aided Tools, similarly in a few years AI would be part of every architect’s workflow. In the end, it is the Architect who is the driver of the boat, steering the wheel and AI is the compass, guiding, and assisting.
References:
Ma, X. (2024). Perplexity vs. ChatGPT for designers — A detailed side-by-side comparison. [online] Medium. Available at: https://medium.com/design-bootcamp/perplexity-vs-chatgpt-for-designers-a-detailed-side-by-side-comparison-f270e07f6470 [Accessed 9 Apr. 2025].
