Independent research & editorial platform
Somatic AI research for embodied movement and human–AI co-creation.
Founded by Hanna Zhu, Somatic-AI Lab connects somatic practice, movement computing and generative systems through source-linked research, field synthesis and practitioner guidance.
This week · Pulse
Mon / Tue / Wed drops — news scan, community field notes, and a plain-language explainer. Updates weekly (Mon–Wed). Public categories: news-digest, community-news, popular-explainer.
Motion AI this week: shared interpretation, controlled noise and off-screen consequences
Research first released from 3–9 November 2026 places motion control at three different levels: a shared language–movement backbone, the noise used to initialise video generation, and the evolving state of a world beyond the camera. Möbius, Householder Motion Embedding and EvoWorld make those levels…
Community signals: when a robot should wait, and how researchers make room to meet
The verified 4–10 November 2026 scan connects a new social-robotics preprint with two official CHI community reports: one asks when a system should withhold movement, while the others document formats for people to discuss and develop research together. These are different kinds of evidence—an exper…
If an AI can describe its own dance, has it checked that the dance is right?
An AI that both generates movement and describes movement can make a creative conversation easier, but agreement between its words and its output is not an independent check. Both may reflect the same learned assumptions. A new unified motion–language model, Möbius, makes this question timely: makin…
This month · Long-form
Motion AI's new alignment stack: time, expression, scene and execution
The most consequential motion-AI work published from 20 September to 31 October 2026 does not converge on one larger generator. It builds an alignment stack: temporal representations that remain readable before a phrase finishes, control objectives that preserve expressive dynamics, joint video–moti…
Body image is not body schema: what AI motion systems keep confusing
Current AI motion systems are increasingly good at representing what a body looks like and where its joints are, yet those achievements should not be mistaken for a body schema: the largely pre-reflective organisation that makes situated action possible. The distinction between body image and body s…
Productive hesitation: micro-phenomenological inquiry meets selective prediction
Confidence label: Plausible. Micro-phenomenological inquiry and selective prediction share a productive problem: how to avoid turning insufficient evidence into a premature answer. They do not share a mechanism or a theory of experience. The connection proposed here is methodological: a movement-rea…
The clarification budget: testing when a movement AI should ask instead of act
Confidence label: Speculative. This brief proposes a non-contact movement experiment in which an AI has a limited budget for asking clarification questions. The question is whether well-timed requests preserve a performer's agency better than either continuous guessing or indiscriminate pausing. No …
An expressive-fidelity audit for body-to-visual systems
This guide is for creative technologists building a live body→signal→visual pipeline in Python or TouchDesigner. It provides a repeatable audit for a common failure: the system tracks a performer accurately but erases how the movement is organised. The workflow was checked against current documentat…
Sarah Fdili Alaoui: designing computation from inside dance practice
Sarah Fdili Alaoui is a professor at the Creative Computing Institute, University of the Arts London, and a choreographer, dancer and Laban Movement Analyst. Her public record matters to embodied AI because it offers a sustained alternative to technology-first dance research: movement knowledge, cho…