Thomas Ferretti

This article explores the cooperation of government and the private sector to tackle the ethical dimension of artificial intelligence (AI). The argument draws on the institutionalist approach in philosophy and business ethics defending a 'division of moral labor' between governments and the private sector (Rawls 2001; Scheffler and Munoz-Dardé 2005). The goal and main contribution of this article is to explain how this approach can provide ethical guidelines to the AI industry and to highlight the limits of self-regulation. In what follows, I discuss three institutionalist claims. First, principles of AI ethics should be validated through legitimate democratic processes. Second, compliance with these principles should be secured in a stable way. Third, their implementation in practice should be as efficient as possible. If we accept these claims, there are good reasons to conclude that, in many cases, governments implementing hard regulation are in principle (if not yet in practice) the best instruments to secure an ethical development of AI systems. Where adequate regulation exists, firms should respect the law. But when regulation does not yet exist, helping governments build adequate regulation should be businesses' ethical priority, not self-regulation.

Suppose a principle of distributive justice states that social institutions should maximize the freedom of the least well-off. Understanding how to do so would be easier if freedom only depended on one good, like income. If it depends instead on a composite index of social primary goods, a question arises: Which combination of social primary goods can maximize the freedom of the least well-off? This is John Rawls’ indexing problem. Solving it requires addressing two related problems. The first consists in evaluating, in theory, under which conditions it is acceptable to substitute goods, that is, their substitution rates. The second consists in evaluating which acceptable substitutions are feasible in practice. This article proposes a framework to think clearly about this indexing problem within a Rawlsian, resourcist conception of distributive justice. I conclude by discussing a path towards solving the indexing problem. While further empirical exploration is needed, plausible assumptions about social regimes suggest that maximizing the freedom of the least well-off is likely to require giving them access to a social position with a balanced combination of social primary goods.

This article explores the cooperation of government and the private sector to tackle the ethical dimension of artificial intelligence (AI). The argument draws on the institutionalist approach in philosophy and business ethics defending a ‘division of moral labor’ between governments and the private sector (Rawls 2001; Scheffler and Munoz-Dardé 2005). The goal and main contribution of this article is to explain how this approach can provide ethical guidelines to the AI industry and to highlight the limits of self-regulation. In what follows, I discuss three institutionalist claims. First, principles of AI ethics should be validated through legitimate democratic processes. Second, compliance with these principles should be secured in a stable way. Third, their implementation in practice should be as efficient as possible. If we accept these claims, there are good reasons to conclude that, in many cases, governments implementing hard regulation are in principle (if not yet in practice) the best instruments to secure an ethical development of AI systems. Where adequate regulation exists, firms should respect the law. But when regulation does not yet exist, helping governments build adequate regulation should be businesses’ ethical priority, not self-regulation.

Suppose a principle of distributive justice states that social institutions should maximize the freedom of the least well-off. Understanding how to do so would be easier if freedom only depended on one good, like income. If it depends instead on a composite index of social primary goods, a question arises: Which combination of social primary goods can maximize the freedom of the least well-off? This is John Rawls’ indexing problem. Solving it requires addressing two related problems. The first consists in evaluating, in theory, under which conditions it is acceptable to substitute goods, that is, their substitution rates. The second consists in evaluating which acceptable substitutions are feasible in practice. This article proposes a framework to think clearly about this indexing problem within a Rawlsian, resourcist conception of distributive justice. I conclude by discussing a path towards solving the indexing problem. While further empirical exploration is needed, plausible assumptions about social regimes suggest that maximizing the freedom of the least well-off is likely to require giving them access to a social position with a balanced combination of social primary goods.

Le modèle déductivo-nomologique domine depuis longtemps la réflexion philosophique concernant l’explication en science économique. Or, pour plusieurs, comme James Woodward et Tony Lawson, ce modèle ne considère pas suffisamment la causalité dans l’explication. L’objectif de cet article est double : 1- renforcer la critique que Tony Lawson adresse à l’économie contemporaine et 2- évaluer la théorie alternative qu’il propose, le « réalisme critique », qui tente de réintroduire la causalité dans l’explication en science économique. Nous conclurons que les considérations causales sont très importantes pour l’explication en science économique mais aussi pour bien comprendre les limites de cette science