Home Robot Safety Is All About Relationships

· Source: IEEE Spectrum · Field: Technology & Digital — Robotics & Autonomous Systems, Artificial Intelligence & Machine Learning, Compliance & Risk Management · Depth: Intermediate, short

Summary

The International Organization for Standardization (ISO) is updating its 12-year-old safety requirements for personal care robots, specifically ISO 13482. While the proposed revision addresses hazard identification, risk assessment, and various use scenarios, technology policy researcher Jae-Seong Lee argues it falls short by not setting limits, proposing testing methods, or establishing enforcement mechanisms for the complexities of human-robot collaboration. Lee emphasizes that safety in domestic robotics is not a fixed property of the machine but emerges from the bidirectional human-robot relationship, which is particularly challenging in uncontrolled home environments with diverse users like older adults and children. Unlike industrial robots, domestic humanoids must adapt to clutter, pets, and fluctuating human behavior, making traditional operating envelopes insufficient. The current standard acknowledges these relational hazards but lacks enforceable rules, delaying critical decisions about safe relational behavior and whose perspectives define "normal" or "safe" judgment.

Key takeaway

For robotics engineers developing domestic humanoids, you must move beyond machine-centric safety metrics. Your design brief should shift from merely assessing robot outputs to ensuring safe engagement across the full range of human states and uncontrolled environments. Prioritize system-level relational assurance, actively embedding diverse user perspectives, especially those of older adults, into your safety frameworks to prevent deeply entrenched, unexamined assumptions from hardening in early product deployments.

Key insights

Domestic robot safety is relational, emerging from human-robot interaction, not just machine properties.

Principles

In practice

Topics

Best for: CTO, VP of Engineering/Data, Director of AI/ML, Robotics Engineer, Policy Maker, AI Ethicist

Related on AIssential

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by IEEE Spectrum.