Designers’ Place In An AI Empowered World
FORMA Design strives to stay current on continuing technological advances and how they relate to the design industry. This article describes the current state of Artificial Intelligence (AI) technologies and what benefits and constraints the technologies have in relation to the real estate development design industry.
Designers’ Place In An AI Empowered World
A topic surrounded by controversy within many industries is Artificial Intelligence. In some industries, the technology can boldly be compared to man first discovering fire according to Google CEO Sundar Pichai, or bring about the apocalypse. What does AI mean for architects and landscape architects and the future of the design industry? In many ways AI can prove to be helpful to workflows and eliminate many of the common mundane tasks designers and planners often experience; controversially some results created by AI have striking resemblances to the best work humans can offer.
To understand the evolving landscape of Artificial Intelligence we need to first grasp what Artificial Intelligence means by definition. Artificial Intelligence, as defined by Columbia University, is “the field of developing computers and robots that are capable of behaving in ways that both mimic and go beyond human capabilities”. Programs that utilize AI are able to autonomously analyze and contextualize given data without any human intervention. An example of the previously mentioned process could be GPS navigation. Instead of pulling out a map and manually navigating yourself based on your understanding of where you are, GPS automatically examines your location and logically devises the most efficient way of getting to your destination. This results in automation of a human task done at the speed of a computer in real time. These kinds of tasks can be easily recognized in many functions of our everyday lives and date back to 2011 with virtual assistants like Siri and Alexa.
Starting in April of 2022 the emergence of mainstream and ready to use generative AI and machine learning began. Generative AI defined by McKinsey & Company describes algorithms “that can be used to create new content, including audio, code, images, text, simulations, and videos”. Due to the vast and seemingly terrifying uses possible with generative AI, this niche of technology has drastically changed the way people can create. Machine learning uses “algorithms to automatically learn insights and recognize patterns from data, applying that learning to make increasingly better decisions” (Columbia University). The most prevalent example of this groundbreaking type of AI is ChatGPT. Released in November 2022, ChatGPT, similar to its image generating counterpart DALL-E 2, is able to generate text based on a large data set. From the process of machine learning, ChatGPT is able to write poems, essays, code, etc. and is also able to solve simple problems just based on a text prompt.
The dynamic of machine learning to generative AI has been applied to the field of Architecture and Landscape Architecture in a variety of ways to tackle an array of issues / tasks, such as concept generation, dimension drafting and annotation, schematic design, rendering, development planning, and design optimization. Released in May of 2023, Autodesk Forma (no affiliation with FORMA Design, Inc.) is one of the most promising examples of AI pushing the industry forward and maximizing efficiency for project development from start to finish. The software is able to generate topography maps and 3D terrain models of any location in a matter of minutes, a process previously orchestrated by tedious GIS processes and 3D software. The software also allows the introduction of contextual data pertaining to the desired location and quick generation of buildings with desired dimensions, floor counts, and floor plans. The most powerful application of this tool comes in the visualization and numerical data about proposed designs calculated accurately in real time. Autodesk Forma visually demonstrates daylight potential, sun exposure, wind comfort, microclimate, and impact of landscape interventions. Based on the ease and speed of data analyzed by AI systems within Autodesk Forma, planners and designers can quickly come up with informed decisions about efficiency and impact of their designs.
Closely related to Autodesk Forma, Sidewalk Labs, developed by Google, can be used by planners and developers to evaluate previous development options based on cost and quality of life data. Delve, a program under the Sidewalk Labs suite generates data driven options for developers and planners which outperform benchmarks. In use cases for the Sidewalk Labs suite, developers such as Quintain, based in the U.K., saw over 10% increase in metrics including profit on cost, residential units, usable residential area, and open space. While extremely powerful, these tools make complex workflows streamlined and comfortable to approach and negate the need for anyone who specializes in manually collecting and evaluating data to make informed decisions for urban projects. The computer does it all at almost no cost and does it faster with a higher degree of accuracy and possibilities.
Another emerging tool revolutionizing the technical aspects of project development is BricsCAD. BricsCAD is a tool similar to Sidewalk Labs which uses generative AI on many scales to provide designers with endless possibilities based on their desired parameters. BricsCAD can “conceptualize 3D building models based on user-defined parameters such as number of floors, building size, and the materials to be used” (Architizer). At a smaller scale BricsCAD can automatically implement 3D details into general conceptual models and can also automate the process of categorizing CAD components in structures, annotation, and replicate architectural and structural details in 2D drawings which are also automatically generated in real time. The software automatically completes detailed information throughout a model generated in 2D and 3D. The automation software like BricsCAD provides drastically reduced times spent generating and manually created drawings and modeling in general, making workflow and design phases more efficient.
Perhaps the most recognized type of tool that AI brings to the world is generative AI tools which have been popularized by programs such as Midjourney, DALL-E, and Photoshop’s Generative Fill. The misconception is these tools potentially do all the rendering and conceptual work for you in a heartbeat just by typing words. Unfortunately, no matter how perfect tools like these may seem, they are simply not intelligent enough to replace designers. They are simply a tool which we can use, not a tool that will eliminate us. A close comparison for the uses of these tools is rendering software such as Lumion or V-Ray. A designer may input a pre-created model into a rendering tool and get a basic result; however, to fully display the designer’s intention, the image may need to be touched up in Photoshop or the smaller components in that image may need to be altered. This is where the designer’s strength comes in to play. AI currently cannot replace the precise emotions and nuance that a human thinks or feels.
Generative AI tools struggle to comprehend context within images, or understand inputted designs and gritty specifics based on the input of words. Best stated by Patrick Hebron, a user experience designer at Adobe and graduate professor from New York University, “AI has limited purview into the nature and proclivities of human experience”. While generative AI tools can do amazing things such as creating plant assets for renderings and stylize images however the user sees fit, their functionalities stop there at asset generation, concept formulation, and basic alterations in Photoshop’s case.
Practicing design professionals should look at AI powered tools with optimism, not something to fear. AI powered tools empower individuals and smaller teams to accelerate their practice by encouraging and positioning computers and software to excel. AI tools allow designers to make intelligent and data driven decisions that offer “possibilities or evidence to support the human decision-making process” (Hebron). Instead of being detrimental to human creativity or replacing it, AI empowers the design process and allow humans to be the best at what they do. The overlapping of human processes and AI processes gives birth to workflows unprecedented in efficiency and empowerment for individuals and smaller entities.
Articles sourced:
15 Top AI Tools for Architects and Designers – Architizer Journal
Embracing artificial intelligence in architecture – AIA
What is ChatGPT, DALL-E, and generative AI? | McKinsey
The History of Artificial Intelligence, Machine Learning and Deep Learning (algotive.ai)
Artificial Intelligence (AI) vs. Machine Learning | Columbia AI
FORMA Design, Inc.
151 Kalmus Drive, Suite H5
Costa Mesa, CA 92626
714.673.6200
info@formacompanies.com