Welcome to the Nexus of Ethics, Psychology, Morality, Philosophy and Health Care

Welcome to the nexus of ethics, psychology, morality, technology, health care, and philosophy

Sunday, January 2, 2022

Towards a Theory of Justice for Artificial Intelligence

Iason Gabriel
Forthcoming in Daedelus vol. 151, 
no. 2, Spring 2022


This paper explores the relationship between artificial intelligence and principles of distributive justice. Drawing upon the political philosophy of John Rawls, it holds that the basic structure of society should be understood as a composite of socio-technical systems, and that the operation of these systems is increasingly shaped and influenced by AI. As a consequence, egalitarian norms of justice apply to the technology when it is deployed in these contexts. These norms entail that the relevant AI systems must meet a certain standard of public justification, support citizens rights, and promote substantively fair outcomes -- something that requires specific attention be paid to the impact they have on the worst-off members of society.

Here is the conclusion:

Second, the demand for public justification in the context of AI deployment may well extend beyond the basic structure. As Langdon Winner argues, when the impact of a technology is sufficiently great, this fact is, by itself, sufficient to generate a free-standing requirement that citizens be consulted and given an opportunity to influence decisions.  Absent such a right, citizens would cede too much control over the future to private actors – something that sits in tension with the idea that they are free and equal. Against this claim, it might be objected that it extends the domain of political justification too far – in a way that risks crowding out room for private experimentation, exploration, and the development of projects by citizens and organizations. However, the objection rests upon the mistaken view that autonomy is promoted by restricting the scope of justificatory practices to as narrow a subject matter as possible. In reality this is not the case: what matters for individual liberty is that practices that have the potential to interfere with this freedom are appropriately regulated so that infractions do not come about. Understood in this way, the demand for public justification stands in opposition not to personal freedom but to forms of unjust imposition.

The call for justice in the context of AI is well-founded. Looked at through the lens of distributive justice, key principles that govern the fair organization of our social, political and economic institutions, also apply to AI systems that are embedded in these practices. One major consequence of this is that liberal and egalitarian norms of justice apply to AI tools and services across a range of contexts. When they are integrated into society’s basic structure, these technologies should support citizens’ basic liberties, promote fair equality of opportunity, and provide the greatest benefit to those who are worst-off. Moreover, deployments of AI outside of the basic structure must still be compatible with the institutions and values that justice requires. There will always be valid reasons, therefore, to consider the relationship of technology to justice when it comes to the deployment of AI systems.