Abstract
In Aldous Huxley's 1932 novel Brave New World every person is conditioned through technologies to fit their social role, which is the source of the society's alleged stability and rightness. This utopia of 'precision justice' is alive today in projects that deploy data and algorithmic models in decision-making in diverse branches of life to realize computing's promise to create more just societies. I identify the sociotechnical imaginary of 'precision justice' by analyzing the promise and contestation of the use of an algorithmic model to calculate exam grades in the United Kingdom during the Covid-19 pandemic. 'Precision justice' is the product of the coupling of a normative concept of just distribution with data practices of identification and risk assessment, and is characterized by interventionist action, optimal distribution, and system management. It crystallized in the contexts of the emerging 'information society' in the 1970s United States, when visions of the risks and opportunities of information in digital form converged with the popular theory and practices of distributive justice. At stake in this imaginary is the model of the human with which it operates and that it reproduces. Instead of keying people to a substantive and expansive concept of justice, the union of distributive justice and data practices bind people to indicators and allocate them to specific places in society. To move beyond precision justice, this article calls for the need to look at justice and data symmetrically, as a simultaneously epistemic and normative set of concerns that must be addressed together in terms of what worlds we want to build.