Risk and Ethics Assessment of HRI Applications before Deployment: Large Language Models in Social Robots for Children with Autism
This internal research report examines the necessity of apriori ethical and risk assessment for human-robot interaction applications, specifically focusing on large language models in social robots for children with autism. Evaluating a platform combining social robots with LLMs using EU Guidelines for Trustworthy AI, expert interviews, and literature, the authors identify 68 perceived risks and opportunities and formulate an action plan. The findings highlight essential methodological gaps and a lack of empirical evidence while demonstrating that early integration of ethics improves trustworthiness.
Summary and content fields are drawn from the document text and are pending editorial review.
Who issued it
Who it governs
AI researchers, developers, and practitioners evaluating ethical and risk considerations for deploying social robots and language models in sensitive educational or care contexts.
What the document contains
The document covers the integration of large language models and social robots for supporting caregivers in creating educational materials for children with autism, including a risk analysis identifying 68 risks and opportunities and a proposed action plan.
Binding legal enforcement mechanisms, broad legislative mandates, and comprehensive empirical evidence for long-term deployment risks.
- Ethical statements in AI research often remain superficial and are treated as bureaucratic burdens rather than substantive engagement.
- An apriori ethical and risk assessment is necessary for sensitive applications such as robot-assisted education for children with autism.
- Integrating ethics early in system design improves trustworthiness and societal value while reducing costs.
- Evaluating a combined LLM and social robot platform identified 68 perceived risks and opportunities for children and caregivers.
- Significant methodological gaps and a lack of empirical evidence currently hinder the proper assessment of major risks.
How it scores
Each score is backed by a quote verified against the document text. Editorial review has not yet confirmed them.
Accountability
2/5Clear responsibility for AI outcomes
Industry track papers, which are potentially closest to deployment, rarely conduct substantive ethical analysis, frequently offloading responsibility to 'humans in the loop'.
Abstract
screening score · 2026-10-23
Fairness
1/5Non-discrimination and equitable treatment
often repetitive, and largely limited to issues such as privacy, bias, or annotator compensation.
Abstract
screening score · 2026-10-23
Human Dignity
1/5Respect for the inherent worth of persons
Risk and Ethics Assessment of HRI Applications before Deployment: Large Language Models in Social Robots for Children with Autism
Title
screening score · 2026-10-23
Human Oversight
1/5Humans remain in control of consequential decisions
frequently offloading responsibility to 'humans in the loop'.
Abstract
screening score · 2026-10-23
Privacy
1/5Protection of personal data and private life
often repetitive, and largely limited to issues such as privacy, bias, or annotator compensation.
Abstract
screening score · 2026-10-23
Transparency
0/5Openness about how AI systems work and are used
Abstract
screening score · 2026-10-23
Enforcement
0/5Consequences and compliance mechanisms with real force
Abstract
screening score · 2026-10-23
Impact Assessment
4/5Required assessment of an AI system's risks/impacts before or during use
This article argues for the necessity of an apriori ethical and risk assessment, particularly for sensitive applications such as robot-assisted education for neurodivergent children.
Abstract
screening score · 2026-10-23