Welcome to this edition of Architecture Insights, the newsletter that keeps you in the loop on topics related to Artificial intelligence in Architecture in 5 minutes or less.

In this week’s issue:

What is AI Image Generating Software?

In the rapidly changing world of AI, it is essential to keep up with the latest advancements. For architects and designers, AI-powered image rendering software has the potential to revolutionize the creative industry and the practice of architecture.

Creating images with AI is simple. You input a text prompt, and the AI generates an image based on it.

Here is a set of images that were generated of a brutalist style building on the ocean side:

Like many image generators out there, 1-4 images will be created and you will have the option to choose or edit 1 or more.

While it is unlikely that the generated image is exactly what you want to show your client, or even remotely close, these can act as great starting points for concept design.

We’ll leave it up to you to be the judge of the quality and accuracy of this image, which we generated on Midjourney, a software we will discuss shortly.

How It Works

The AI is trained using millions of data sets provided by software developers. It can be in the form of text or image. The AI is like a massive brain that learns from the data it receives.

For example, if it is shown multiple sets of images depicting brutalist architecture, palm trees, and beaches, it would know what to create when prompted with a similar request.

What are the programs out there & which ones are the best for architects?

As previously mentioned, we utilized Midjourney to create our image. However, we want to clarify that we are not promoting Midjourney or any other software but it’s worth noting that Midjourney is one of the most extensively utilized and developed software solutions for image generation.

Midjourney is often preferred for projects that require high precision and a true-to-life representation. It can generate realistic design aspects such as shadows, reflections, and textures.

Some of you may have already heard of Midjourney, along with DALL-E, Adobe Firefly, Dreamstudio, and others. Each has its strengths and weaknesses.

With the ease of API access, software development is now accessible to many newcomers. However, it is important to research thoroughly before committing to an AI image generator to ensure its viability.

There is no definitive answer to which software to use. We suggest trying as many as possible but keeping in mind our previous advice.

Photoshop vs AI

The question arises: Is Photoshop and manual drafting becoming obsolete? Can we replace hours of rendering with a 30-second prompt for accurate, high-quality images?

This is something only you can determine by experimenting with AI.

As a student, it is important to learn how to use popular software such as Photoshop, SketchUp, and Vectorworks. It is unlikely that you will be able to secure a job without any experience in these programs. Therefore, it is imperative to gain proficiency in them to increase your chances of success in the industry.

For practitioners, the choice depends on accuracy. Our prompt earlier asked for 3 things: a brutalist style building, palm trees, and a beach. The more specific your design needs, the more challenging it is for AI to deliver.

There is no doubt that companies can save a significant amount of money on billable work by using artificial intelligence. However, this raises ethical concerns. Will it make some parts of your job redundant since anyone can use AI? Will it destroy the artistic aspect of architecture? There are more questions than answers in this situation.

At present, software such as Adobe, SketchUp, Vectorworks, and Rhino 3D offers the greatest potential for customizing your project, meaning the full potential of AI is still yet to be recognized in the field of architecture and design and the coming years will dictate a lot in the industry.

Question & Answer

Arty Molinari is a licensed architect out of Flordia who has over 10 years of experience in the field. Hear his thoughts below:

How do you feel about the rise of AI specifically in the field of architecture, are you implementing AI in any way?

I’m on board with it because it is not going away. For architecture it will speed up the way we produce things whether it’s written or graphical. I use it for research and analysis right now. I’m excited about how advanced it will become even though the negative side of it could be quite dangerous from a global cyber standpoint.

In your opinion what are the biggest challenges that the profession will face in the next 10-20 years, how can we prepare for these challenges?

With AI we’re in the biggest transition of the profession since the invention of CAD. The speed at which the profession grows now from a technology standpoint is huge. I personally want to stay on top of the knowledge but I do not want to be a technician anymore. Business is where things happen at the largest scale not design. We all can agree that design sells but without marketing and a solid business strategy and system it won’t go anywhere far. What I’ve seen change in the last decade since graduating college is immense so to go another decade or more into the future is almost unimaginable. However, it’s exciting and I’m going to make sure I’m on top of the wave to take advantage as I’m sure it’s going to be fun and prosperous.

We hope you learned something new in this edition of architecture insights!

Until next Friday,

A.I.