Research Questions the Way AI and Robotics Talk About the Human Body

An article by Nobuchika Yamaki published in the AI & SOCIETY journal argues that the way AI and robotics describe the human body still relies too heavily on outdated terminology. The dominant concepts in the field remain control, optimization, prediction, and representation — the idea that the body functions like a system to be measured, modeled, and adjusted towards a goal.

According to Yamaki, these concepts once provided clarity and made design manageable in an engineering sense. However, he argues that they increasingly fail to correspond to what bodies actually do as people learn and move in everyday life. The central concern of the article is not only that these metaphors are persistent but that they remain in use largely without critical questioning — as if their suitability for describing 'embodied life' has already been resolved.

The writing progresses by examining what movement and skill acquisition look like when viewed closely from the level of human experience and action. Yamaki emphasizes that no one experiences their own body as a control system. According to him, skill does not primarily arise from executing internal instructions nor from systematically reducing error step by step.

The article highlights the impact of language and modeling: the concepts used to describe the body also guide what assumptions are considered self-evident in AI and robotics. Yamaki calls for closer attention to what movement and learning truly are before attempting to fit them into familiar, perhaps already distorted models.

Source: The movement we still do not know how to model, AI & SOCIETY.