The current Socially Assistive Robotics (SAR) paradigm is afflicted by a categorical error: the assumption that social acceptance is directly proportional to the accumulation of anthropomorphic mimetic signals. This paper introduces the Phenomenological Subtraction framework, which reverses the traditional design vector from “additive” (the accumulation of social cues) to “subtractive” (instrumental transparency). Synthesising technical transparency (Heidegger), intentional spatiality (Husserl) …
Read moreThe current Socially Assistive Robotics (SAR) paradigm is afflicted by a categorical error: the assumption that social acceptance is directly proportional to the accumulation of anthropomorphic mimetic signals. This paper introduces the Phenomenological Subtraction framework, which reverses the traditional design vector from “additive” (the accumulation of social cues) to “subtractive” (instrumental transparency). Synthesising technical transparency (Heidegger), intentional spatiality (Husserl) and resistance to optimisation (Dostoevsky), the framework presents a model in which an artificial agent’s efficacy is determined by its capacity to withdraw. By integrating the concepts of ontological honesty and sensory attenuation, the model provides solutions for safeguarding “ontological privacy” and ensuring compliance with the transparency requirements of the EU AI Act. Ultimately, the paper posits that algorithmic silence represents the pinnacle of ethics and aesthetics in human-machine interaction. Keywords: AI Ethics, Social Robotics, Phenomenology, EU AI Act, Human-Robot Interaction (HRI), Subtractive Design.
