Recent years have seen a significant change in how images are produced. Artificial intelligence and machine learning have recently become more prevalent, raising concerns about how creative processes can advance technologically. With the help of a dataset of word-image pairs, AI programmes like DALL-E, DALL-E 2, and Midjourney have been trained to create images from text descriptions. The broad range of talents includes anthropomorphizing objects and animals, connecting seemingly unrelated ideas, and altering already existing visuals.
For a wide range of sentences that investigate the compositional structure of language, DALL-E and comparable algorithms can produce believable visuals. Although DALL-E has some of the same capabilities as a 3D rendering engine, the way it receives inputs differs. In contrast to DALL-E, which may frequently “fill in the blanks,” 3D rendering requires that the input be given in exact detail. Additionally, it has the ability to independently control some objects’ characteristics.
Combining disparate notions is one of the most exciting elements. As it is possible for product design and architecture to draw inspiration from seemingly unrelated ideas, this skill may have repercussions for the design and architectural industries. The AI generative models shorten the time between intention and execution, encouraging designers to consider more design options from fresh angles. They provide a simple way to experiment with data and come up with unique approaches to challenging situations.
These technologies, which enhance the chance for serendipity and expand the spectrum of creative capacities beyond the conventional ways, are referred to by some researchers as “Artificial Serendipity.” These tools are already being used experimentally by architects to investigate intricate topics like urban planning and the potential of existing locations. Others are creating structures or merely examining the nature of design trends and technology by fusing architectural terms with modern design clichés, pop culture allusions, and various art forms.
Despite the limits of these models, the area is developing at a never-before-seen pace. Gaudi, a “neural architect” that Apple recently launched, advances this method by generating 3D sceneries from text instructions like “move upstairs” or “walk down the hallway.” Although it is difficult to foresee where these advancements may lead us, their effects are already apparent. These are strong tools that may be used in the domains of architecture and design to quickly investigate, optimise, and test innovative designs.
