Paper Number
ECIS2026-1496
Paper Type
CRP
Abstract
Artificial intelligence and machine learning increasingly shape sectors, yet their effectiveness depends on access to large, representative datasets - often limited by regulatory and organisational constraints. Federated learning (FL) promises an alternative by enabling multiple actors to collaboratively train models without sharing raw data. While the technical literature on FL is extensive, its ethics remain underexplored within the Information Systems discipline. This study therefore considers what ethical principles remain unaddressed in FL, despite promises of privacy-preserving data processing. Following a qualitative, inductive approach, we combine a systematic literature review of 27 papers with 7 semi-structured interviews with practitioners implementing FL. Our findings reveal that while privacy emerges as the dominant focus, considerations related to accountability, fairness and transparency received comparatively limited attention. We contribute to research by advancing an empirically grounded theoretical understanding of ethics in FL, proposing an agenda for future research, and outlining implications for practice.
Recommended Citation
Jones, Pippa and Barbereau, Tom, "Ethics In Federated Learning" (2026). ECIS 2026 Proceedings. 3.
https://aisel.aisnet.org/ecis2026/resp_AI/resp_AI/3
Ethics In Federated Learning
Artificial intelligence and machine learning increasingly shape sectors, yet their effectiveness depends on access to large, representative datasets - often limited by regulatory and organisational constraints. Federated learning (FL) promises an alternative by enabling multiple actors to collaboratively train models without sharing raw data. While the technical literature on FL is extensive, its ethics remain underexplored within the Information Systems discipline. This study therefore considers what ethical principles remain unaddressed in FL, despite promises of privacy-preserving data processing. Following a qualitative, inductive approach, we combine a systematic literature review of 27 papers with 7 semi-structured interviews with practitioners implementing FL. Our findings reveal that while privacy emerges as the dominant focus, considerations related to accountability, fairness and transparency received comparatively limited attention. We contribute to research by advancing an empirically grounded theoretical understanding of ethics in FL, proposing an agenda for future research, and outlining implications for practice.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.