Following her recent article about the potential for text-to-image AI, AFL Architects’ Rita Ochoa asks ChatGPT what it thinks.
I thought it would be interesting to ask ChatGPT, the most talked about AI tool, what it sees as the benefits of text-to-image generation. This is what it said, along with some added insights to highlight some examples from AFL.
Design visualisation
ChatGPT said: “Architects often work with clients who may not have a background in design or architectural drawings. Text-to-image generation allows architects to generate realistic visual representations of their designs based on textual descriptions. This enables clients to better understand and visualise the proposed architectural concepts, helping them make informed decisions and provide feedback.”
Concept exploration
ChatGPT said: “Architects often start with conceptual ideas that are described in text form. Text-to-image generation can assist in rapidly creating visual representations of these concepts, allowing architects to explore various design alternatives and refine their ideas before investing significant time and resources in detailed design development.”
To the right is an example of how the interior of a restaurant may look, allowing the client to assess flow and function alongside its aesthetic and design.
Communication and collaboration
ChatGPT said: “Text-to-image generation can facilitate effective communication and collaboration among architects, clients, and other stakeholders. Instead of relying solely on technical drawings or architectural jargon, architects can generate visual images from text descriptions to convey their ideas more clearly. This helps bridge the gap between architectural expertise and the understanding of non-technical individuals, fostering better collaboration and decision-making.”
Text-to-image generation can also support CPD for either those studying architecture or architectural assistants to learn and develop their skills. Above shows the outcome of a training session AFL held with some of its team using text-to-image technology.
Rapid prototyping
ChatGPT said: “Text-to-image generation can be used for rapid prototyping and design iteration. Architects can generate images from textual descriptions of design elements or features, enabling them to quickly explore and evaluate different design options. This iterative process helps architects refine their designs efficiently and identify potential issues or improvements early in the design phase.”
Accessibility and inclusivity
ChatGPT said: “Text-to-image generation can enhance accessibility in architectural design. People with visual impairments or limited spatial visualisation abilities can benefit from textual descriptions being transformed into visual representations. By providing alternative modalities for understanding architectural designs, text-to-image generation contributes to inclusivity in the field.”
It can also illustrate how spaces that are inclusive for all can look like – for example the main image (top) of a sensory room and one of a wellbeing space (below) – both starting points of ideation that can help depict to clients form and function.
Rita Ochoa is senior associate at AFL Architects
Don’t miss out on BIM and digital construction news: sign up to receive the BIMplus newsletter.
