Today's technology companies seem to move so fast and break so many things that it is hard for academic scholarship to keep up. Mathias Risse's Political Theory of the Digital Age: Where Artificial Intelligence Might Take Us examines the ethics and politics of various digital technologies, such as deepfake videos, artificial intelligence (AI), and data collection business practices now known as surveillance capitalism. Risse argues that our technologically transformative era necessitates a fourth generation of human rights defined in terms of epistemic rights and epistemic justice. To grasp the rise of big data and related technologies, Risse implores philosophers and theorists to reject the divisions traditionally set between various schools of thought, himself pursuing a unique methodological blend of public reason liberalism and Marxism. The result is a unique liberal theory that recognizes the need for formal rights to be supported by material conditions, including democracy and distributive justice. Risse's work is an excellent contribution to a burgeoning literature in political theory (e.g., Jennifer Forestal, Designing for Democracy: How to Build Community in Digital Environments [Oxford Univesity Press, 2022]; Eileen Hunt, Artificial Life after Frankenstein [University of Pennsylvania Press, 2020]), which illustrates the need for inclusive and open conversation among scholars to better incorporate digital technologies into our understandings of politics.
Risse expands the purview of liberal-egalitarian thought to the unique political problems posed by big data and artificial intelligence (AI), as well as “the roles of citizens as knowers and knowns—that is, as both holders and providers of ‘data’ and ‘information’” (xi). Risse toggles between our current stage of digital technological development and a speculative vision of a future defined by the rise of artificial general intelligence (AGI), machines with intellectual capacities equivalent to those of human beings across many different fields. Once we develop AGI, it could design and produce machines smarter than itself, a moment that has become known as the singularity. On what grounds would such beings respect human life? The threat of the singularity raises the stakes of Risse's work, as he argues that properly grappling with the implications of today's technological problems such as online disinformation and surveillance capitalism will point us toward the kinds of epistemic justice that will be necessary to encourage peaceful coexistence among people and superintelligence machines in the future. While Risse bases his arguments on public reason liberalism, he criticizes Rawls's neglect of the politics of technology and turns to Marxist theories of the politics of technology (30–36).
Risse shows how our fantasies about technology can reflect and perpetuate potentially harmful assumptions about politics. He outlines an optimistic hypothetical scenario whereby a massive AI system successfully provides accurate information to voters, “instantly identifies fraud and corruption, flags biased reporting and misleading arguments,” and improves the quality of governance while increasing public participation (60). As enticing as this sounds, this utopia treats politics as a technical problem or logical conundrum that can be worked out by optimizing for a singular mode of superior enough intelligence. By focusing on the creation of a successful technical system, we risk failing to develop our own capacities for ruling together (61). Additionally, the “black-box” nature of machine-learning AI means that neither the machine nor its creators can explain the processes by which it reaches a decision, putting such systems at odds with the ideal that political decisions be able to be justified by reason giving (67).
While reaffirming the importance of public reason, Risse offers a nuanced vision of epistemic rights, arguing for legal protections for truth in the public sphere while warning that a comprehensive right to truth would be at odds with the human demand for narrative and meaning, including the comprehensive doctrines that public reason seeks to leave intact (94–95). Risse emphasizes a suitably defined right to be forgotten and a right to exercise human intelligence (152–59).
Risse complements his theory of epistemic rights by attending to the material reality through which we are known and know each other: big data, a resource primarily enjoyed by private technology companies today. Risse aptly criticizes prevalent characterizations of data as the new “oil” or as “labor” for neglecting the relational nature of data. Each individual's set of data reveals information about themselves but also the many other people to which they are connected. Further, an individual's distinct set of data is not particularly valuable alone but only as part of a larger aggregate of networked information. Risse characterizes data as a Durkheimian social fact or a collectively generated pattern (194–95). The relational nature of data means that it would be technically challenging to solve debates over control of online data by dividing it into private property for each user to keep private, share, or sell as they prefer. Risse uses Hugo Grotius's treatment of the seas as a common to argue for collective ownership of data.
Risse's arguments falter when he makes more ambiguous claims regarding the materiality of technology. In a chapter based on an article Risse coauthored with Catherine Kerner (“Beyond Porn and Discreditation: Epistemic Promises and Perils of Deepfake Technology in Digital Lifeworlds,” Moral Philosophy and Politics 8, no. 1 [2021]: 81–108), Risse conscientiously attends to the many dangers of deepfake technologies, such as the reputational harms of nonconsensual deepfake pornography experienced by Indian journalist Rana Ayyub. However, Risse argues deepfake technology's pornographic uses are only the beginning of unpredictable creative possibilities, just as the internet has transcended its earlier years, when Pamela Anderson was the most-searched person in 1995–2005 (123, 130). Yet Risse does not acknowledge that Pamela Anderson was a victim of nonconsensual pornography (Amanda Chicago Lewis, “Pam and Tommy: The Untold Story of the World's Most Infamous Sex Tape,” Rolling Stone, December 22, 2014). Particularly confusing is Risse's argument that, as long as deepfake videos are not distributed beyond the creator's personal use, the videos “are not very different from how we often think about fantasies in someone's mind.” Because “fantasizing should not be punished,” Risse concludes, legal responses should target distribution (134). This claim leads to various practical questions as to under what conditions a video should be considered fantasy or material property, legally or ethically. Addressing such questions would benefit from engaging with other approaches to nonconsensual pornography, such as the civil rights framework used by Danielle Citron (Brian Feldman, “MacArthur Genius Danielle Citron on Deepfakes and the Representative Katie Hill Scandal,” New York Magazine, October 31, 2019; Danielle Citron, The Fight for Privacy: Protecting Dignity, Identity, and Love in the Digital Age [Norton, 2022]).
Nevertheless, Political Theory of a Digital Age exemplifies how grappling with the digital political economy requires conversation across scholarly traditions and disciplines, including in-depth philosophical investigation of rights and institutions. In a compelling and nuanced conclusion, Risse enables readers to envision a future where public reason liberalism could support the conditions under which AGI and human beings would agree to respect and value each other's different modes of intelligence and life. Between now and the potential singularity, there remain crucial questions about the development and implementation of AI, including enduring questions about economic relations and political institutions, with which political theorists working in all traditions must reckon.
