Top

AI & SOCIETY

Published in:

27-01-2022 | Original Article

Against “Democratizing AI”

Author: Johannes Himmelreich

Published in: AI & SOCIETY | Issue 4/2023

Activate our intelligent search to find suitable subject content or patents.

search-config

AI-assisted search

Off

Abstract

This paper argues against the call to democratize artificial intelligence (AI). Several authors demand to reap purported benefits that rest in direct and broad participation: In the governance of AI, more people should be more involved in more decisions about AI—from development and design to deployment. This paper opposes this call. The paper presents five objections against broadening and deepening public participation in the governance of AI. The paper begins by reviewing the literature and carving out a set of claims that are associated with the call to “democratize AI”. It then argues that such a democratization of AI (1) rests on weak grounds, because it does not answer to a demand of legitimization, (2) is redundant in that it overlaps with existing governance structures, (3) is resource intensive, which leads to injustices, (4) is morally myopic and thereby creates popular oversights and moral problems of its own, and finally, (5) is neither theoretically nor practically the right kind of response to the injustices that animate the call. The paper concludes by suggesting that AI should be democratized not by broadening and deepening participation but by increasing the democratic quality of the administrative and executive elements of collective decision making. In a slogan: The question is not so much whether AI should be democratized but how.

previous article Varieties of transparency: exploring agency within AI systems

next article Legal personhood for the integration of AI systems in the social context: a study hypothesis

Dont have a licence yet? Then find out more about our products and how to get one now:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

über 102.000 Bücher
über 537 Zeitschriften

aus folgenden Fachgebieten:

Automobil + Motoren
Bauwesen + Immobilien
Business IT + Informatik
Elektrotechnik + Elektronik
Energie + Nachhaltigkeit
Finance + Banking
Management + Führung
Marketing + Vertrieb
Maschinenbau + Werkstoffe
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

inform now

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

über 67.000 Bücher
über 340 Zeitschriften

aus folgenden Fachgebieten:

Bauwesen + Immobilien
Business IT + Informatik
Finance + Banking
Management + Führung
Marketing + Vertrieb
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

inform now

Springer Professional "Technik"

Online-Abonnement

Mit Springer Professional "Technik" erhalten Sie Zugriff auf:

über 67.000 Bücher
über 390 Zeitschriften

aus folgenden Fachgebieten:

Automobil + Motoren
Bauwesen + Immobilien
Business IT + Informatik
Elektrotechnik + Elektronik
Energie + Nachhaltigkeit
Maschinenbau + Werkstoffe

Jetzt Wissensvorsprung sichern!

inform now

These three ideas—fairness, freedom and equality—mean different things. By “freedom” I understand the capacity to see one’s will carried out and, more generally, a robust congruence between one’s actions and the conditions of one’s life on the one hand and the authentic expression of one’s values on the other. By “equality” I understand, first, the tenet that each individual has the same moral worth and, second, that this tenet finds its expression in how individuals relate to each other—that they relate to each other as equals. By “fairness” I mean an impartial appraisal of the reasons that each individual could offer on matters of common concern.

Be there no doubt: Those who emphasize the participatory and populist elements of democracy are in no way to blame or even complicit in this authoritarian misappropriation.

I argue for this latter claim elsewhere: Proposals, which nominally aim to improve democracy, often hollow out its values (Himmelreich 2022).

Strictly speaking, algorithms are abstract objects, like theorems and arithmetic operations. It is not obvious how this vast class of abstract objects—and not just their implementations—are supposed to give rise to ethical problems.

A related set of claims is defended for the practice of scientific research across the board by Kitcher (2011).

Because of this condition to increase or introduce direct democratic powers, the so-called Moral Machine experiment is not a form of democratizing AI. Some proponents of such surveys—and they are usually just surveys and not experiments—argue that public attitudes about ethics and technology must be studied, identified, and articulated to be “cognizant of public morality” (Awad et al. 2018). The idea is that the hence elicited public attitudes are to limit policymaking, because, otherwise, “societal push-back will drastically slow down the adoption of intelligent machines” (Awad et al. 2020). This approach is flawed (Jaques 2019; Himmelreich 2020). The overall idea contrasts with the call to “democratize AI” because it 1) aims mainly to inform and 2) sees individuals as subjects in an investigation. By contrast, the demand to democratize AI seeks to empower individuals, to endow marginalized groups with novel ways to make their voices heard (although in the wrong way as I argue here), and to give citizens greater direct influence—collective power—over decisions.

This proposal by Mills (2019) can be distinguished into a proposal about organizational function (the data trust) and a proposal about the trust’s mode of governance (deeper participation). My argument is only about the latter.

Kitcher writes (2011, 127): “Current scientific research neglects the interests of a vast number of people, except insofar as their interests coincide with those of people in the affluent world.”.

Kitcher also writes (2011, 126): “Privatization of scientific research will probably matters worse.” Given that much research on AI is privatized and proprietary, the problems that animate Kitcher are amplified in the case of AI.

Admittedly, some of the relevant experts in cases of AI injustice are those who suffer the injustice. It is their expertise that must find its way into our deliberation and collective decision-making. But I disagree that broadening and deepening participation is the right way of doing so.

On the tension between participation and deliberation see Cohen (2009, sec. 5).

The distinction intrinsic vs. instrumental value conflates a distinction about values’ location (intrinsic vs. extrinsic) with a distinction about their relations (final vs. instrumental). See Korsgaard (1983).

Questions about the legitimacy of the state: Why should you respect what the state asks you to do? Why can some demands of the state be enforced, even coercively? Questions about justifying democracy: Why should you value, and perhaps choose, democracy over alternative systems?

The distinction between autonomy as sovereignty and autonomy as non-alienation in these terms is due to Enoch (Enoch 2017; 2020).

Many associations are governed democratically. Labor unions, recreational clubs, or church parish administrations are examples. In addition to exhibiting triggers of legitimatization requirements, these associations can also be seen as essential parts of a democratic society. In other words, they might be part of a state democracy and part of meeting legitimization requirements that are triggered by the state.

This assumes, of course, that there are feasible alternatives that have fewer of the costs outlined above.

I here argue against the second assumption of Sclove’s argument presented earlier.

Structural injustice are systematic violations of particularly important moral claims or liberties, the maintenance of which is explained by non-individual entities such as cultures, norms, or practices.

Although some argue that the distinction between procedure and substance collapses ( Cohen 1993, 1997). Thus, there might be no such thing as a purely proceduralist conception of democracy.

The question, of course, is whether a rejection of material equality is reasonable.

Some deliberative democrats want public reasons. That is, they demand that this justification should be based on reasons that everyone can accept. Roughly, the same justification should be offered to everyone. By contrast, others argue that each individual can be offered a different justification as long as each can be offered some reasons. Roughly, they contend that different reasons can be offered to different people.

Free expression is the broader category, it includes, for example, artistic expression.

Abizadeh A (2007) Cooperation, pervasive impact, and coercion: on the scope (not site) of distributive justice. Philos Public Aff 35(4):318–358CrossRef

Anderson E (2017) Private government: how employers rule our lives (and why we don’t talk about it). Princeton University Press, PrincetonCrossRef

Angwin J, Larson J, Mattu S and Kirchner L (2016) Machine bias. ProPublica. https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing. Accessed 24 July 2018

Awad E, Dsouza S, Kim R, Schulz J, Henrich J, Shariff A, Bonnefon J-F, Rahwan I (2018) The moral machine experiment. Nature. https://doi.org/10.1038/s41586-018-0637-6CrossRef

Awad E, Dsouza S, Bonnefon J-F, Shariff A, Rahwan I (2020) Crowdsourcing moral machines. Commun ACM 63(3):48–55. https://doi.org/10.1145/3339904CrossRef

Bartels LM (2002) Beyond the running tally: partisan bias in political perceptions. Polit Behav 24(2):117–150. https://doi.org/10.1023/A:1021226224601CrossRef

Bell DA (2016) The china model: political meritocracy and the limits of democracy. Princeton University Press, PrincetonCrossRef

Blake M (2008) International justice. In: Edward NZ (ed) The Stanford encyclopedia of philosophy. Winter 2008. http://plato.stanford.edu/archives/win2008/entries/international-justice/

Brennan J (2016) Against democracy. Princeton University Press, PrincetonCrossRef

Brock G (2021) Global justice. In: Edward NZ (ed) The Stanford encyclopedia of philosophy. Winter 2021. Metaphysics Research Lab, Stanford University https://plato.stanford.edu/archives/win2021/entries/justice-global/

Brooks T (2020) The Oxford handbook of global justice. Oxford University Press, OxfordCrossRef

Broome J (2012) Climate matters: ethics in a warming world. W. W. Norton & Company, New York

Cammaerts B, Mansell R (2020) Digital platform policy and regulation: toward a radical democratic turn. Int J Commun 14(22):135–154

Carugati F (2020) A council of citizens should regulate algorithms. Wired, December 6, 2020. https://www.wired.com/story/opinion-a-council-of-citizens-should-regulate-algorithms/. Accessed 17 Aug 2020

Checkoway B (1981) The politics of public hearings. J Appl Behav Sci 17(4):566–582. https://doi.org/10.1177/002188638101700411CrossRef

Cohen J (1989) Deliberation and democratic legitimacy. In: Hamlin AP, Pettit P (eds) The good polity: normative analysis of the state. Blackwell, Oxford, pp 17–34

Cohen J (1993) Pluralism and proceduralism. Chicago-Kent Law Rev 3(1994):589–618

Cohen J (1997) Procedure and substance in deliberative democracy. In: James B, William R (eds) Deliberative democracy: essays on reason and politics. The MIT Press, Cambridge Mass, pp 407–437

Cohen GL (2003) Party over policy: the dominating impact of group influence on political beliefs. J Pers Soc Psychol 85(5):808–822. https://doi.org/10.1037/0022-3514.85.5.808CrossRef

Cohen J (2009) Reflections on deliberative democracy. In: Thomas C, John C (eds) Contemporary debates in political philosophy. Wiley, Chichester, pp 247–263. https://doi.org/10.1002/9781444310399CrossRef

Crawford K (2021) The Atlas of AI: power, politics, and the planetary costs of artificial intelligence. Yale University Press, New HavenCrossRef

Curato N, Dryzek JS, Ercan SA, Hendriks CM, Niemeyer S (2017) Twelve key findings in deliberative democracy research. Daedalus 146(3):28–38. https://doi.org/10.1162/DAED_a_00444CrossRef

Dahl RA (1989) Democracy and its critics. Yale University Press, New Haven

Ditto PH, Pizarro DA, Tannenbaum D (2009) Motivated moral reasoning. In: Bartels DM, Bauman CW, Skitka LJ, Medin DL (eds) Psychology of learning and motivation: moral judgment and decision making. Academic Press, San Diego, pp 307–338CrossRef

Dryzek JS, Bächtiger A, Chambers S, Cohen J, Druckman JN, Felicetti A, Fishkin JS et al (2019) The crisis of democracy and the science of deliberation. Science 363(6432):1144–1146. https://doi.org/10.1126/science.aaw2694CrossRef

Enoch D (2017) Hypothetical consent and the value(s) of autonomy. Ethics 128(1):6–36. https://doi.org/10.1086/692939CrossRef

Enoch D (2020) False consciousness for liberals, part I: consent, autonomy, and adaptive preferences. Philos Rev 129(2):159–210. https://doi.org/10.1215/00318108-8012836CrossRef

Gabriel I (2018) The problem with yuppie ethics. Utilitas 30(1):32–53. https://doi.org/10.1017/S0953820817000024CrossRef

Gilens M, Page BI (2014) Testing theories of American politics: elites, interest groups, and average citizens. Perspect Polit 12(3):564–581. https://doi.org/10.1017/S1537592714001595CrossRef

Goldfarb A, Tucker C (2019) Digital economics. J Econ Lit 57(1):3–43. https://doi.org/10.1257/jel.20171452CrossRef

Gould CC (2019) How democracy can inform consent: cases of the internet and bioethics. J Appl Philos 36(2):173–191. https://doi.org/10.1111/japp.12360CrossRef

Haidt J (2012) The righteous mind: why good people are divided by politics and religion. Penguin, New York

Heap SH, Hollis M, Lyons B, Sugden R, Weale A (1992) The theory of choice: a critical guide. Blackwell, Oxford

Heath J (2020) The machinery of government: public administration and the liberal state. Oxford University Press, OxfordCrossRef

Himmelreich J (2020) Ethics of technology needs more political philosophy. Commun ACM 63(1):33–35. https://doi.org/10.1145/3339905CrossRef

Himmelreich J (2022) Should we automate democracy? In: Carissa V (ed) The Oxford handbook of digital ethics. Oxford University Press, Oxford

Hulme D (2016) Should rich nations help the poor? Polity, Cambridge

Irvin RA, Stansbury J (2004) Citizen participation in decision making: Is it worth the effort? Public Adm Rev 64(1):55–65. https://doi.org/10.1111/j.1540-6210.2004.00346.xCrossRef

Jaques AE (2019) Why the moral machine is a monster. In: 10. Miami School of Law. https://robots.law.miami.edu/2019/wp-content/uploads/2019/03/MoralMachineMonster.pdf. Accessed 22 Oct 2021

Jurowetzki R, Hain D, Mateos-Garcia J, Stathoulopoulos K (2021) The privatization of AI research(-Ers): causes and potential consequences—From University-Industry Interaction to Public Research Brain-Drain?. http://arxiv.org/abs/2102.01648

Kahan DM (2012) Ideology, motivated reasoning, and cognitive reflection: an experimental study. SSRN Scholarly Paper ID 2182588. Rochester: Social Science Research Network. https://doi.org/10.2139/ssrn.2182588

Kahneman D (2011) Thinking, fast and slow. Farrar, Straus and Giroux, New York

Kitcher P (2011) Science in a democratic society. Prometheus Books, Amherst

Korsgaard CM (1983) Two distinctions in goodness. Philos Rev 92(2):169–195. https://doi.org/10.2307/2184924CrossRef

Lakoff G (2008) The political mind: a cognitive scientist’s guide to your brain and its politics. Penguin, New York

Lenz GS (2013) Follow the leader?: How voters respond to politicians’ policies and performance. University of Chicago Press, Chicago

Lodge M, Taber CS (2013) The rationalizing voter. Cambridge University Press, CambridgeCrossRef

Lord CG, Ross L, Lepper MR (1979) Biased assimilation and attitude polarization: the effects of prior theories on subsequently considered evidence. J Pers Soc Psychol 37(11):2098–2109. https://doi.org/10.1037/0022-3514.37.11.2098CrossRef

Mayer R (2001) Strategies of justification in authoritarian ideology. J Political Ideol 6(2):147–168. https://doi.org/10.1080/13569310120053830CrossRef

McQuillan D (2018) People’s councils for ethical machine learning. Soc Media Soc 4(2):2056305118768303. https://doi.org/10.1177/2056305118768303CrossRef

Miller D (2009) Democracy’s domain. Philos Public Aff 37(3):201–228. https://doi.org/10.1111/j.1088-4963.2009.01158.xCrossRef

Mills CW (2013) Retrieving Rawls for racial justice?: A critique of Tommie Shelby. Crit Philos Race 1(1):1–27. https://doi.org/10.5325/critphilrace.1.1.0001MathSciNetCrossRef

Mills CW (2015) Racial equality. In: George H (ed) The equal society: essays on equality in theory and practice. Lexington Books, Lanham, pp 43–72

Mills CW (2018) I—Racial justice. Aristot Soc Supplement 92(1):69–89. https://doi.org/10.1093/arisup/aky002CrossRef

Mills S (2019) Who owns the future? Data trusts, data commons, and the future of data ownership. SSRN Electron J. https://doi.org/10.2139/ssrn.3437936CrossRef

Moellendorf D (2015) Climate change justice: climate change justice. Philos Compass 10(3):173–186. https://doi.org/10.1111/phc3.12201CrossRef

Nabatchi T, Leighninger M (2015) Public participation for 21st century democracy. Wiley, HobokenCrossRef

O’Neill O (2016) Justice across boundaries: Whose obligations? Cambridge University Press, CambridgeCrossRef

Plato (2008) Republic. In: Robin Waterfield (ed) Translated. Oxford University Press, Oxford

Pogge T (2005) World poverty and human rights. Ethics Int Aff 19(1):1–7. https://doi.org/10.1111/j.1747-7093.2005.tb00484.xCrossRef

Pogge T (2008) World poverty and human rights: cosmopolitan responsibilities and reforms. Polity, Cambridge

Rahwan I (2018) Society-in-the-loop: programming the algorithmic social contract. Ethics Inf Technol 20(1):5–14. https://doi.org/10.1007/s10676-017-9430-8CrossRef

Rawls J (1971) A theory of justice. 1999, Revised. Harvard University Press, CambridgeCrossRef

Rawls J (1993) Political liberalism. Columbia University Press, New York

Rawls J (2001) Justice as fairness: a restatement. Harvard University Press, CambridgeCrossRef

Rejali D (2009) Torture and democracy. Princeton University Press, PrincetonCrossRef

Ronzoni M (2009) The global order: a case of background injustice? A practice-dependent account. Philos Public Aff 37(3):229–256. https://doi.org/10.1111/j.1088-4963.2009.01159.xCrossRef

Sclove R (1995) Democracy and technology. Guilford Press, New York

Shelby T (2003) Race and social justice: Rawlsian considerations. Fordham l Rev 72:1697

Shue H (1980) Basic rights: subsistence, affluence, and U.S. foreign policy. Princeton University Press, Princeton

Shue H (2020) Basic rights: subsistence, affluence, and U.S. foreign policy: 40th, Anniversary. Princeton University Press, PrincetonCrossRef

Shughart WF, Thomas DW (2019) Interest groups and regulatory capture. In: Congleton RD, Grofman B, Voigt S (eds) The Oxford handbook of public choice, vol 1. Oxford University Press, Oxford, pp 584–603. https://doi.org/10.1093/oxfordhb/9780190469733.013.29CrossRef

Singer P (1972) Famine, affluence, and morality. Philos Public Aff 1(3):229–243

Sloane M, Moss E, Awomolo O and Forlano L (2020) Participation is not a design fix for machine learning. http://arxiv.org/abs/2007.02423

Stanley ML, Henne P, Yang BW, De Brigard F (2020) Resistance to position change, motivated reasoning, and polarization. Polit Behav 42(3):891–913. https://doi.org/10.1007/s11109-019-09526-zCrossRef

Sunstein CR (2002) The law of group polarization. J Polit Philos 10(2):175–195CrossRef

Sunstein CR (2006) Infotopia: how many minds produce knowledge. Oxford University Press, Oxford

Taylor RS (2009) Rawlsian affirmative action. Ethics 119(3):476–506. https://doi.org/10.1086/598170CrossRef

Tutt A (2017) An FDA for algorithms. Admin Law Rev 69(1):83–124

Tversky A, Kahneman D (1974) Judgment under uncertainty: heuristics and biases. Sci New Ser 185(4157):1124–1131

Unger P (1996) Living high and letting die: our illusion of innocence. Oxford University Press, OxfordCrossRef

Weaver VM, Prowse G (2020) Racial authoritarianism in U.S. democracy. Science 369(6508):1176–1178. https://doi.org/10.1126/science.abd7669CrossRef

Westen D (2008) The political brain: the role of emotion in deciding the fate of the nation. PublicAffairs, New York

Wong P-H (2020) Democratizing algorithmic fairness. Philos Technol 33:225–244. https://doi.org/10.1007/s13347-019-00355-wCrossRef

Young IM (1990) Justice and the politics of difference. Princeton University Press, Princeton

Zimmermann A, Di Rosa E and Kim H (2020) Technology can’t fix algorithmic injustice. Boston Rev. http://bostonreview.net/science-nature-politics/annette-zimmermann-elena-di-rosa-hochan-kim-technology-cant-fix-algorithmic. Accessed 09 Jan 2020

Title: Against “Democratizing AI”
Author: Johannes Himmelreich
Publication date: 27-01-2022
Publisher: Springer London
Published in: AI & SOCIETY / Issue 4/2023
Print ISSN: 0951-5666
Electronic ISSN: 1435-5655
DOI: https://doi.org/10.1007/s00146-021-01357-z

Springer Professional

Abstract

Please log in to get access to your license.

Dont have a licence yet? Then find out more about our products and how to get one now:

Springer Professional "Wirtschaft+Technik"

Springer Professional "Wirtschaft"

Springer Professional "Technik"

Other articles of this Issue 4/2023

Maximizing team synergy in AI-related interdisciplinary groups: an interdisciplinary-by-design iterative methodology

A democratic way of controlling artificial general intelligence

Many hands make many fingers to point: challenges in creating accountable AI

The application of chatbot on Vietnamese misgrant workers’ right protection in the implementation of new generation free trade agreements (FTAS)

Psychoanalyzing artificial intelligence: the case of Replika

Caring in the in-between: a proposal to introduce responsible AI and robotics to healthcare

Premium Partner