Privacy & Safety Challenges of On-Body Interaction Techniques
Authors
Paper Title
Privacy & Safety Challenges of On-Body Interaction Techniques
Publication Info
- Topic area: Privacy and safety in on-body computing systems.
- Keywords: On-body interaction, privacy, safety, wearable devices, human-computer interaction, bystander privacy, data inference, physical harm, psychological harm, design guidelines.
Background and Problem
- Problem / challenge: Privacy and safety challenges specific to on-body computing systems remain understudied, particularly the interplay between these domains.
- Significance: As computing technologies move closer to the body, risks such as data over-collection, psychological harm, and threats to autonomy become amplified, demanding proactive design approaches.
- Motivation and related work: Prior research has addressed privacy and safety in traditional computing platforms and specific technologies like VR and AR. However, these frameworks do not fully account for the unique risks posed by on-body systems, such as continuous body contact, novel sensor modalities, and direct bodily actuation.
Solution
- Proposed approach: A two-round interview study with experts in HCI, privacy, and safety, using speculative scenarios and adversarial role-playing to identify challenges and derive actionable design guidelines.
- Novelty:
- Empirically grounded characterization of privacy and safety challenges specific to on-body interaction techniques.
- Identification of the deep interconnection between privacy and safety risks.
- Development of eight actionable design guidelines to mitigate identified risks.
- Procedure and key techniques:
- Conducted a literature review to classify interaction techniques based on observability and bandwidth.
- Interviewed 15 experts in HCI and privacy/safety using scenarios and adversarial personas.
- Applied thematic analysis to identify challenges and synthesized guidelines through affinity diagramming and scenario-based validation.
Results
- Concrete findings:
- Privacy risks include data over-collection, inference of sensitive information, and bystander privacy violations.
- Safety risks encompass physical harm (e.g., collisions, hardware failures), psychological harm (e.g., anxiety, manipulation), and threats to autonomy (e.g., conditioning, external control).
- Privacy breaches often directly lead to safety risks, highlighting their interdependence.
- Advantage over baselines: Provides a systematic exploration of privacy and safety risks specific to on-body systems, addressing gaps in existing frameworks.
- Experiments / evaluation:
- Two rounds of interviews with experts (N=15), including adversarial role-playing exercises.
- Interaction techniques categorized across observability and bandwidth dimensions.
- Validation of guidelines through scenario-based exercises.
- Limitations and future work:
- Findings are based on qualitative data from a targeted sample of experts (N=15), limiting generalizability.
- Scenarios used were abstract, and risks identified are speculative.
- Guidelines require validation in real-world deployments and in-the-wild studies.
Summary
This paper investigates privacy and safety challenges in on-body interaction techniques, revealing their deep interconnection and expanded threat surface. Through expert interviews, it identifies risks such as data over-collection, psychological harm, and threats to autonomy, and proposes eight actionable design guidelines to mitigate these challenges. The findings emphasize the need for proactive, interdisciplinary approaches to ensure trustworthy on-body systems, while future work should validate these guidelines in practical applications.
Research Questions / Practical Problems
Question signals indexed for this paper.
- 71%
Mind the Gap: Mapping Wearer–Bystander Privacy Tensions and Context-Adaptive Pathways for Camera Glasses
CHI '26· Privacy by Design & User Control +2
- 71%
BAIT: Visual-illusion-inspired Privacy Preservation for Mobile Data Visualization
CHI '26· Uncertainty Visualization +2
Based on Jaccard similarity of research subtopics & professions (≥60%)