Federated learning is not a cure-all for data ethics

Marieke Bak; Vince I. Madai; Leo Anthony Celi; Georgios A. Kaissis; Ronald Cornet; Menno Maris; Daniel Rueckert; Alena Buyx; Stuart McLennan

doi:10.1038/s42256-024-00813-x

Federated learning is not a cure-all for data ethics

Marieke Bak
, Vince I. Madai
, Leo Anthony Celi
, Georgios A. Kaissis
, Ronald Cornet
, Menno Maris
, Daniel Rueckert
, Alena Buyx
, Stuart McLennan

Technical University of Munich
University of Amsterdam
Charite Universitätsmedizin Berlin
Birmingham City University
Massachusetts Institute of Technology
Harvard Medical School
Harvard T.H. Chan School of Public Health
Imperial College London
Institute of Machine Learning in Biomedical Imaging
Amsterdam UMC
University of Basel

Research output: Contribution to journal › Comment/debate

18 Scopus citations

Abstract

Although federated learning is often seen as a promising solution to allow AI innovation while addressing privacy concerns, we argue that this technology does not fix all underlying data ethics concerns. Benefiting from federated learning in digital health requires acknowledgement of its limitations.

Original language	English
Pages (from-to)	370-372
Number of pages	3
Journal	Nature Machine Intelligence
Volume	6
Issue number	4
DOIs	https://doi.org/10.1038/s42256-024-00813-x
State	Published - Apr 2024

Access to Document

10.1038/s42256-024-00813-x

Cite this

@article{e891de7198464c179062c59edff43ec2,

title = "Federated learning is not a cure-all for data ethics",

abstract = "Although federated learning is often seen as a promising solution to allow AI innovation while addressing privacy concerns, we argue that this technology does not fix all underlying data ethics concerns. Benefiting from federated learning in digital health requires acknowledgement of its limitations.",

author = "Marieke Bak and Madai, \{Vince I.\} and Celi, \{Leo Anthony\} and Kaissis, \{Georgios A.\} and Ronald Cornet and Menno Maris and Daniel Rueckert and Alena Buyx and Stuart McLennan",

note = "Publisher Copyright: {\textcopyright} Springer Nature Limited 2024.",

year = "2024",

month = apr,

doi = "10.1038/s42256-024-00813-x",

language = "English",

volume = "6",

pages = "370--372",

journal = "Nature Machine Intelligence",

issn = "2522-5839",

publisher = "Springer International Publishing",

number = "4",

}

TY - JOUR

T1 - Federated learning is not a cure-all for data ethics

AU - Bak, Marieke

AU - Madai, Vince I.

AU - Celi, Leo Anthony

AU - Kaissis, Georgios A.

AU - Cornet, Ronald

AU - Maris, Menno

AU - Rueckert, Daniel

AU - Buyx, Alena

AU - McLennan, Stuart

PY - 2024/4

Y1 - 2024/4

N2 - Although federated learning is often seen as a promising solution to allow AI innovation while addressing privacy concerns, we argue that this technology does not fix all underlying data ethics concerns. Benefiting from federated learning in digital health requires acknowledgement of its limitations.

AB - Although federated learning is often seen as a promising solution to allow AI innovation while addressing privacy concerns, we argue that this technology does not fix all underlying data ethics concerns. Benefiting from federated learning in digital health requires acknowledgement of its limitations.

UR - https://www.scopus.com/pages/publications/85188073856

U2 - 10.1038/s42256-024-00813-x

DO - 10.1038/s42256-024-00813-x

M3 - Comment/debate

AN - SCOPUS:85188073856

SN - 2522-5839

VL - 6

SP - 370

EP - 372

JO - Nature Machine Intelligence

JF - Nature Machine Intelligence

IS - 4

ER -

Federated learning is not a cure-all for data ethics

Abstract

Access to Document

Other files and links

Fingerprint

Cite this