Paper Type
ERF
Abstract
Growing evidence of the limitations of algorithm control in gig platforms has accelerated the adoption of hybrid human–algorithm control worldwide. Yet platforms combine algorithms and managers in different ways, from algorithm-dominant control to human-dominant ones, raising unresolved questions about the legitimacy of these configurations across cultures. We develop a culturally contingent legitimacy model that explains how control configurations shape gig workers’ perceived legitimacy and how Hofstede’s cultural dimensions orientations moderate these effects. We will test the model with a cross-cultural survey of gig workers (e.g., in Australia, China, and Nigeria) using established scales and new measures of perceived human–algorithm control configuration. The study advances human–algorithmic interaction and control research by introducing a culturally grounded framework for platform governance and culturally responsive human–algorithm control design. This study helps platforms design fairer human–algorithm control configurations across cultures, lowering turnover and improving workforce stability.
Paper Number
1693
Recommended Citation
Huang, Xuemei; Ehimen-Ebitibituwa, Ekhomen; Khomenko, Mariia; and owobu, wilfred, "The Legitimacy of Human–Algorithm Control: A Cross-Cultural Study in Gig Work" (2026). AMCIS 2026 Proceedings. 5.
https://aisel.aisnet.org/amcis2026/ccris/sig_ccris/5
The Legitimacy of Human–Algorithm Control: A Cross-Cultural Study in Gig Work
Growing evidence of the limitations of algorithm control in gig platforms has accelerated the adoption of hybrid human–algorithm control worldwide. Yet platforms combine algorithms and managers in different ways, from algorithm-dominant control to human-dominant ones, raising unresolved questions about the legitimacy of these configurations across cultures. We develop a culturally contingent legitimacy model that explains how control configurations shape gig workers’ perceived legitimacy and how Hofstede’s cultural dimensions orientations moderate these effects. We will test the model with a cross-cultural survey of gig workers (e.g., in Australia, China, and Nigeria) using established scales and new measures of perceived human–algorithm control configuration. The study advances human–algorithmic interaction and control research by introducing a culturally grounded framework for platform governance and culturally responsive human–algorithm control design. This study helps platforms design fairer human–algorithm control configurations across cultures, lowering turnover and improving workforce stability.
Comments
SIG CCRIS