In Short:

Algorithmic governing power should not simply be eliminated but can be properly exercised—provided it aims at substantively justified ends and is used according to legitimate procedures by those with authority to do so.

Abstract

Political philosophy’s central task is to help us determine how to live together. It analyses our social relations, diagnoses their failings, and articulates ideals to guide their revision. A century ago, John Dewey argued that “[s]team and electricity have done more to alter the conditions under which [people] associate together than all the agencies which affected human relationships before our time,”2 and called for political philosophy to be revised in light of these changes. In recent decades, computing technologies have had no less significant an impact on how we associate together than did steam and electricity. And political philosophy is overdue another update.3 Our social relations are now infused with, sometimes even constituted by, computational systems. These may not enable, as Lessig thought, entirely new modalities of power. But the prevailing means of governing power have undoubtedly shifted.4 And political philosophers must determine whether and how that power can be properly exercised.

This essay begins that task. I begin by clarifying how social relations have changed in the information age. I introduce “the Algorithmic City”—the network of algorithmically mediated social relations of which we are now all part. I then characterize how power operates through these algorithmic intermediaries, highlighting a shift in both the means and prevailing modalities of power compared to our pre-algorithmic social relations. I argue that algorithmic governing power should not simply be eliminated but can be properly exercised—provided it aims at substantively justified ends and is used according to legitimate procedures by those with authority to do so. In Sections VII-X, I show how algorithmic governance raises presumptive challenges for familiar political philosophical approaches to these three normative standards, and consider a series of objections to my thesis.

About the author

Seth Lazar

Professor, SChool of Government and Policy

Seth Lazar is a leading voice in the moral and political philosophy of artificial intelligence, bridging traditional philosophical ethics with contemporary technological challenges.