Research

I study how algorithmic systems govern — and are contested — through uncertainty. The program has two faces. The first is governance: how consequential systems whose rules cannot be read produce vigilance, striving, and self-censorship in the people who must live within them. The second is action: how people coordinate, verify, and decide under the same illegibility. The method is ethnographic, because the mechanism is a lived condition — it does not show up in an audit of outputs.

Governance Through Uncertainty (FAccT ’26)

Lau, Jason C. 2026. “Governance Through Uncertainty: What Chinese Algorithmic Systems Reveal About the Limits of Fairness, Accountability, and Transparency.” In Proceedings of the ACM Conference on Fairness, Accountability, and Transparency (FAccT ’26), 4061–4077. ACM. doi.org/10.1145/3805689.3806495

The paper argues that algorithmic systems can govern not despite their opacity but through it. Drawing on fieldwork in China from 2013 to 2020, it names the mechanism — governance through uncertainty — and the environment that sustains it, the microworld: rule-governed, consequential, epistemically opaque, impossible to opt out of. It then reframes fairness for such systems: not only whether the algorithm is biased, but who can afford illegibility — and who pays for it.

The essay states the argument in public form.

The book

Governance Through Uncertainty: Algorithmic Power and Everyday Life in China develops the argument at monograph scale, drawing on ethnographic fieldwork in China and Hong Kong. Moving from Shenzhen’s shanzhai phone workshops through China’s digital worlds to its emerging AI regulation, it asks what fairness and accountability mean when transparency is not achievable. And it makes the case for the mechanism’s portability: the conditions that produce it are specifiable — and China is not the only place they hold.

Current work

A second paper — on how people act through uncertainty during social unrest — is in preparation. The next project, Regulating Through Uncertainty, asks how states and technology firms build, operate, and regulate algorithmic systems — and what happens when the regulation is as unpredictable as the systems it governs.

Earlier work

Before turning to algorithmic governance, I studied how technology is made in China: Shenzhen’s shanzhai phone workshops, informal manufacturing and design ecologies, and the platforms that grew out of them — research I began with support from the National Science Foundation (USA) and the Wenner-Gren Foundation, and continued as an M+ Museum / Design Trust Research Fellow in Hong Kong. That work is where this program’s method was formed: to understand the people governed by technological systems, know the people who build them.

ORCID: 0000-0001-8755-2155