French-Japanese robotics AI conference maps gaps toward human-centric physical AI
FrnkNlsn · x · 2026-11-17
A talk at the French-Japanese Conference on Physical and Multimodal AI for Robotics (#FJAIRobot) outlined the gaps between foundation models and human-centric physical AI: understanding, predicting and adapting to humans is the missing piece. Key points include why HRI is hard (variable context-dependent behavior, heterogeneous multimodal data, safety and social norms), where to get data (robot manipulation, human activity, and HRI datasets like HSRI and HRI-SENSE), what's still missing (intentions, emotions, subjective outcomes, failure recovery, longitudinal data), and the research opportunity of building human-aware foundation models for trustworthy, adaptive physical AI.
More from Embodied
- Skild AI open-sources Steel physics solver: up to 13x faster than PhysX with 3x less GPU memory — deepakpathak · 2026-11-17
- XREAL Aura review: the spatial computer Apple should have made — Scobleizer · 2026-11-17
- ScaleVLA scales VLA models to 552B parameters: 10 demos beat 2B backbone trained on 50 — GalaxyGeneralRobotics · 2026-11-17
- Amazon's next delivery runner has four legs: RIVR robot joins the last-mile fleet — TinfoilTricorn · 2026-11-17
- Fudan and Fysics AI unveil UniWAM, hitting 80% success on real-robot mobile manipulation — jiqizhixin · 2026-11-17
- ChipAgents CEO joins NYSE interview on chip AI expansion and chip superintelligence vision — WilliamWangNLP · 2026-11-17