Edited by Fabrizio Gay
The 2024 Nobel Prize in Chemistry recognised two fundamental contributions to computational protein design. AlphaFold, developed by Google DeepMind, predicts the three-dimensional and dynamic structure of proteins from their amino acid sequences with accuracy comparable to experimental methods. Rosetta, developed at the David Baker Laboratory, University of Washington, enables de novo protein design by combining physical modelling, optimisation algorithms, and statistical approaches, producing enzymes with unprecedented functions such as pollutant degradation or neutralisation of pathogenic viruses.
These systems are paradigmatic examples of artificial exaptation: AlphaFold maps the space of possible proteins, revealing latent structures and functions, while Rosetta explores their configurations, reusing molecular modules and physical principles in combinations not observed in nature. Although guided by human objectives, they operate according to an adaptive logic similar to evolutionary processes that co-opt pre-existing traits for new functions. They do not create ex novo, but refunctionalise and re-signify existing configurations.
The concept of exaptation, anticipated by Darwinian observations and formalised in biology by Gould and Vrba (1982), denotes the reuse of traits developed for different or non-adaptive functions. Similar dynamics emerge in theories of artefact and cultural evolution: from the technical concretisation of objects (Simondon; Leroi-Gourhan) to anthropological bricolage (Lévi-Strauss; Ingold), from the transformation of the mundane into art (Danto) to processes of translation across semiotic systems (Lotman; Floch; Fontanille). Across these contexts, a shared logic of co-option, reuse, recombination, and re-signification emerges.
Today, exaptation is not merely a theoretical principle but an operational logic, amplified by the capabilities of generative Artificial Intelligence models such as AlphaFold and Large Multimodal Models (LMM), trained on heterogeneous datasets. In these tools, exaptation guides the transformation of informational patterns – images, styles, structures, morphologies – at unprecedented scale and speed, making it central to contemporary creative and design processes.Issue 18 of “XY”, updating the theoretical perspectives proposed by Roberto de Rubertis in Darwin architetto (2012), invites scholars, designers, and researchers to submit contributions exploring artificial exaptation as a generative principle, interpretive lens, or design model.