tagCloud

architecture arduino art audio audiovisual city classes coding computational creative coding design digital digital media documentary domestication interaction interactive iot localization max media mqtt node p5 performance physical processing programme psychogeography publication robotics sensing sensors sonic arts sonification sound synthesis system tangible teaching technology ubiquity visualization wearable workshop

Let’s Work Together

stavros.didakis@nyu.edu

Logico-Fantastic Machine

A Computational Ecology for the Advancement of Creativity

Art & Technology Research, Interactive Installation

3 Robotic Interfaces, 3 Mobile Phones, 5 OLED Screens, Generative AI
2025

“Logico-Fantastic Machine” is a cybernetic ecology of robotic and machine learning agents that endlessly generate visual and textual narratives with extrapolative and creative characteristics. Three physical kinetic structures scan the interior space and constantly search for original features (semantic, contextual, sentimental). The sensed data are transferred to the generative system, which performs a distillation process and outputs audiovisual content in the form of animated imagery, text, and voice via custom-trained models. Each new generation is being evaluated by the system’s machine agents that continuously train themselves; through an iterative process of generation and evaluation, the agents learn how to extrapolate from the previous dataset, and, therefore, how to be creative.

The work is inspired by Italo Calvino’s lecture “Cybernetics & Ghosts”, where he mentions of a machine that is able to generate never-ending narratives, yet remains haunted by the ingenuity of human creativity that echoes of past literature, linguistic structures, and cultural memory. In “Logico-Fantastic Machine,” this tension between algorithmic logic and imaginative possibility unfolds in real time, as the system navigates the boundaries between computation and invention. As the system evolves, it develops its own internal lexicon, an emergent vocabulary shaped by the recursive interplay of data, machine vision, and creative speculation. The audiovisual output oscillates between the recognizable and the xeno, forming a dynamic interplay of synthetic language, fragmented imagery, and uncanny poetics. By integrating real-time environmental feedback, “Logico-Fantastic Machine” resists static authorship, instead existing as an open-ended speculative entity—one that is always writing, always dreaming, and always learning.

Koios-3

Koios-3 consists of three robotic interfaces, each featuring a mobile phone with a custom app that scans distinct properties of both physical and virtual environments. The first device, equipped with a computer vision (CV) system and a Large Language Model (LLM), searches to identify design features/elements of the space. This refers to surrounding colors, textures, patterns, objects, or other features that of the physical world. The second device is also equipped with a CV and an LLM model and focuses on identifying emotional and sentimental aspects of the observed space, aiming to recognize moods, atmospheres, social interactions, and affective states of people present. Finally, the last robotic interface, is connected to a mobile phone that displays a virtual depthmap. The rotation of the device triggers the phone to switch angles and positions and to reveal different views of complex patterns with varying spatial depth. Data gathered from these three interfaces—encompassing design characteristics, emotional cues, and depthmap insights—is transmitted to the prompt generator, which folds these details into instructions that ultimately produce the textual and visual narratives of the main system.

Divergence-Creator

The “Divergence-Creator” is an autonomous generative AI system that creates high-resolution imagery and corresponding textual narratives depicting divergent architectural and urban environments. Its goal is achieved by continually generating new content that strays as far as possible from previous outputs, using real-time sensory input from Koios-3 to guide its custom machine-learning models. Each newly generated image features innovative spatial and aesthetic elements, while the narrative text strives to describe these visual features accurately. Alongside this, a text-to-speech mechanism produces an audio narration in real time, further enhancing the immersive experience for the audience. Once images and text are created, they are automatically assessed by machine agents that measure qualitative and quantitative attributes—such as novelty or technical composition. A physical interface is available for the visitors to assist them evaluate themselves the content by adjusting ten sliders for the corresponding scoring bands (five for text and five for image). The audience’s feedback is added to the system’s database in tandem with the machine-derived evaluations and it remains permanently accessible, allowing the audience to review both the creative outputs and the respective analyses and commentaries. The model then uses this growing dataset to retrain itself, learning from and incorporating the most valuable characteristics—those deemed most original or well-executed—to ensure subsequent generations of images and text diverge further from earlier iterations while maintaining alignment with the values of the main objective.

Obsidian Aperture

The generated images from the Divergence-Creator are automatically extracted into a flowing visual sequence that interpolates between each architectural and urban design, forming a continuous journey through the system’s evolving cityscapes. As the animation materializes in real time—immediately following each image’s creation—it effectively opens a window into the machine learning model’s latent space, showcasing the underlying creative processes that drive this dynamic exploration of form and structure.

The “Logico-Fantastic Machine” is inspired from Italo Calvino’s reflections in his lecture “Cybernetics and Ghosts,” where he envisions a device capable of endlessly reinventing its own codes to craft new narratives and textual structures. By turning its internal logic upside down, the machine transcends its prior limitations, discovering creative avenues that expand well beyond its original scope. In doing so, it pursues Calvino’s broader aim of enlarging the sphere of human imagination: not only by generating unfamiliar stories and forms, but also by demonstrating that even rigorous, rule-bound processes can yield unknown worlds of possibility.