Making Sense of Algorithms: Relational Perception of Contact Tracing and Risk Assessment during COVID-19

Abstract

Governments, institutions, and citizens of nearly every nation have been compelled to respond to COVID-19. Many measures have been adopted, including contact tracing and risk assessment, whereby citizen whereabouts are constantly monitored to trace contact with other infectious individuals and isolate contagious parties via algorithmic evaluation of their risk status. This paper investigates how citizens make sense of Health Code (jiankangma), the contact tracing and risk assessment algorithm in China. We probe how people accept or resist the algorithm by examining their ongoing, dynamic, and relational interactions with it over time. By seeking a deeper, iterative understanding of how individuals accept or resist the algorithm, our data unearths three key sites of concern. First, how understandings of algorithmic surveillance shape and are shaped by notions of privacy, including fatalism towards the possibility of true privacy in China and a trade-off narrative between privacy and twin imperatives of public and economic health. Second, how trust in the algorithm is mediated by the perceived competency of the technology, the veracity of input data, and well-publicized failures in both data collection and analysis. Third, how the implementation of Health Code in social life alters beliefs about the algorithm, such as its further role after COVID-19 passes, or contradictory and disorganized enforcement measures upon risk assessment. Chinese citizens make sense of Health Code in a relational fashion, whereby users respond very differently to the same sociotechnical assemblage based upon social and individual factors.

Engineering care in pandemic technogovernance: The politics of care in China and South Korea’s COVID-19 tracking apps

“…"Today, are you green?" became a widely used catchphrase as the green code was actively mobilized to represent not only mobility but also "trustworthiness" in daily interactions, as similarly identified in Liu and Graham's (2021) interview study with the Health Code users. Businesses in the service industry publicly showed their employees' green codes to customers as evidence of their health to present the company as honest and credible.…”

Chinese Public’s Support for COVID-19 Surveillance in Relation to the West

“…I also used keywords "surveillance (jiankong)" and "Health Code" to collect social media posts from Weibo, the Chinese Twitter. 2 In general, Chinese citizens are well aware of the lack of privacy and invasive surveillance in society (Liu and Graham 2021). One avid supporter of Health Code told me during the interview that "In China no one has privacy, at least now they can use the information for something good."…”

Seeing Like a State, Enacting Like an Algorithm: (Re)assembling Contact Tracing and Risk Assessment during the COVID-19 Pandemic

“…In this case, elaborate surveillance citizenship has also emerged: People must be surveilled and datafied by specific platforms to access certain public spaces or services. Even when we acknowledge the public’s perception of the surveillance as proof of caring ( Liu 2021 ; Liu and Graham 2021 ) and recognize algorithmic assemblages are multiple, elastic, and porous, this caring and flexibility do not apply to everyone equally. As we have observed, some people game Health Code to save five minutes waiting in a line, while others are stuck at home for weeks due to confusion about the technology.…”

“…These enrollments and assemblings are not uninterrupted projections of an algorithmic design but, instead, a process that always involves disassembling and reassembling. Planned infrastructures and devices that support the algorithm may break down ( Houston, Gabrys, and Pritchard 2019 ; Vertesi 2014 ); actors perceive and behave around the algorithm differently ( Amelang and Bauer 2019 ; Liu and Graham 2021 ); resistance and struggles occur ( Chen and Sun 2020 ); and, in the process, new sociotechnical actors become involved ( Lee et al 2019 ). These messy, diverse, and interconnected engagements co-constitute the algorithm’s sociotechnical assemblage.…”

“…My reliance on online recruitment also restrained my ability to reach out to low SES populations. A more detailed methodological reflection about their exclusion from digital studies was published in another paper ( Liu and Graham 2021 )…”