Human-proximity behavior enters model evaluation
DeepMind reported stronger safety-instruction and proximity behavior, including stopping a humanoid when a person approached its operating area.
Technology Domain: safety
Human-Robot Teamwork Gets a Requirements Taxonomy
Researchers consolidate fragmented human-robot teamwork requirements into six categories and twenty-one subcategories.
Trustworthy Embodied Intelligence Gets a Graded Systems Framework
A systems framework links robust task execution, hardware safeguards, evidence and deployment governance.
MulRobBench Extends Safety Evaluation to Multi-Robot Decisions
A benchmark evaluates decision-level safety and reliability across multi-robot and aerial scenarios.
World-Model Security Research Expands the Threat Surface Around Embodied AI
Researchers mapped attacks across sensing, world modelling, planning, control and human interaction in embodied-AI systems.
