AI code tools struggle to capture how bodies actually feel and move
Researchers found that large language models can help designers sketch ideas about physical experiences—from chronic pain to posture—but the AI often gets trapped in oversimplified solutions and produces fragile code that breaks easily. The finding matters for companies building health tech, wearables, and embodied AI tools: LLMs alone aren't ready to reliably translate messy human experiences into working systems.
Originaltitel: Exploring bodily phenomena through code: a research through design inquiry of sketching with llms
<p>How can large language models (LLMs) support critical design explorations for describing, articulating, and responding to bodily phenomena? This study investigates the integration of code-generating LLMs across design explorations in four different self-tracked or bio-sensed domains: learning, chronic illness, posture control, and knee injury. We synthesise our collaborative research through design inquiries into three themes: 1) forced articulation of fuzzy phenomena, 2) difficulty of escaping local optima, and 3) unexpected fragility of generated sketches. These themes, though potentially applicable in other domain contexts, are grounded in our challenges with code-generating LLMs on embodied and embodiment subjects, with a particular focus on working with technology and data relevant to embodied experiences. We propose two implications for design of LLM-based sketching tools anchored on our design inquiry journeys.</p>