Digital Grotesque II   (2017)

The resulting architecture presents a complex, rich formal language that is simultaneously disorienting, intriguing, and evocative, yet deliberately non-prescriptive. It inhabits a space between the man-made and the natural, between order and chaos, offering unexpected moments of surprise.

Evolving Autonomy: From Generation to Evaluation

This project marks a significant evolution from its predecessor, Digital Grotesque I . In the first iteration, the computational system primarily served as a form generator, with the architect consistently evaluating each proposal, adjusting parameters, and synthesizing forms into new iterations. This workflow, while productive, still mandated continuous human oversight and decisive intervention at every stage of the design process.

Digital Grotesque II advances this human-computer collaboration by endowing the computational system with not only generative capacity but also evaluative autonomy. Rather than merely producing options for human assessment, the system learns to evaluate its own creations and optimize them accordingly. This represents a profound progression: from the computer as a sophisticated form generator to a more autonomous designer, capable of iterative self-improvement.

Question of Purpose: Optimizing a Grotto

Yet this enhanced autonomy immediately raises a fundamental question: if the computer can now optimize designs through iterative learning, what should it optimize toward? Unlike conventional architectural structures with clear functional requirements - maximizing structural efficiency, minimizing material use, or optimizing environmental performance - a grotto serves no explicit functional purpose. Its value lies entirely in the experience it creates for those who encounter it. This absence of conventional optimization targets opens space for a different kind of architectural intelligence.

Geometric Properties for Engagement

Rather than claiming to directly measure aesthetic experience itself, the project identifies three geometric properties that demonstrate positive correlation with viewer engagement and interest:

• Experience-ability: changes in the form’s appearance or visibility through a shift of the viewer’s position or perspective.
• Compressibility: the degree of geometric information that cannot be simplified through symmetry groups and repetition.
• Depth-complexity: measuring the topological alternation between void and solid within a structure, cultivating an immersive experience

These metrics enabled the computer to learn relationships between geometric properties and human response. Through evaluation of design variations by online volunteers, the system developed predictive models linking form and geometric properties to emotional impact. The computer could then generate and iteratively optimize proposals specifically to maximize curiosity and sustained engagement.

Evolving Role for the Architect

In this process, the architect's role transforms from direct designer to curator - steering the computational exploration, evaluating proposals, and synthesizing the most compelling solutions into a singular form. The human provides aesthetic judgment and contextual understanding while the computer provides generative capacity that far exceeds individual imagination.

However, this approach raises important questions about the nature of aesthetic experience itself. The reduction of human response to quantifiable metrics, while enabling computational optimization, necessarily simplifies the complex cultural, symbolic, and personal dimensions through which architecture creates meaning. The challenge lies in harnessing computational creativity while preserving the deeper human values that make architecture culturally significant.

Though generated through autonomous computational processes, the form evokes analogies across multiple domains - from architectural traditions to natural formations. This suggests that the computer, while creating the unprecedented, may have inadvertently uncovered patterns that resonate with deeper human perceptual structures and archetypes. It raises the question about the ultimate origin of the form.

Digital Grotesque II offers a glimpse into the ongoing evolution of computational design, prompting us to consider the emergence of machines that can assess and refine their own creations based on aesthetic criteria.