Alexandra Prégent

The prospect of AI systems that I call “ideal” emotion recognition technologies (ERTs) is often defended on the assumption that social life would benefit from increased affective transparency. This paper challenges that assumption by examining the technosocial risks posed by ideal ERTs, understood as multimodal systems capable of reliably inferring inner affective states in real time. Drawing on philosophical accounts of emotional expression and social practice, as well as empirical work in affective science and social psychology, I argue that the appeal of such systems rests on a misunderstanding of the social functions of emotional expression. Emotional expressions function not only as read-outs of inner states, but also as tools for coordinating action, enabling moral repair, sustaining interpersonal trust, and supporting collective norms. These functions depend on a background of partial opacity and epistemic friction. When deployed in socially authoritative or evaluative contexts, ideal ERTs threaten this expressive space by collapsing epistemic friction, displacing relational meaning with technology-mediated affective profiles, and narrowing the space for aspirational and role-sensitive expressions. The result is a drift towards affective determinism and ambient forms of affective auditing, which undermine both social cohesion and individual agency. I argue that, although it is intuitive to think that increasing accuracy would legitimise such systems, in the case of ERTs accuracy does not straightforwardly justify their deployment, and may, in some contexts, provide a reason for regulatory restraint. I conclude by defending a function-first regulatory approach that treats expressive discretion and intentional emotional expression as constitutive of certain social goods, and that accordingly seeks to protect these goods from excessive affective legibility.

Faisant suite aux critiques de plusieurs femmes de la société qui désapprouvaient et délégitimisaient les théories féministes prenant comme modèle de la femme la conception ethnocentrique blanche et hétérosexiste, Iris Marion Young tente de penser le groupe « femme » en tant que groupe, sans exclure aucune de ses parties. Cette tâche la conduit vers la notion sartrienne de structure sérielle pour penser les femmes comme un groupe social dont les membres n’ont pas nécessairement à partager les mêmes attributs pour être reconnu-e-s en tant que tel. Cette approche lui permet d’envisager la possibilité d’un féminisme qui ne repose pas sur la catégorie « femmes » basée sur des objets communs, mais comme une catégorie qui ferait une unité avec la pluralité. Embrassant les différentes pratiques et les transformant en des enjeux politiques uniques à la condition de femme, Young rend ainsi possible de penser le féminisme comme un groupe dans lequel les femmes, à la fois en tant qu’êtres singuliers et en tant qu’êtres appartenant à plusieurs catégories, peuvent se retrouver, se réunir, mais surtout s’identifier. Le féminisme considéré comme théorie et pratique politique devient un groupe auquel toutes les femmes s’identifient.

Contextual Integrity (CI) is built to assess potential privacy violations of new sociotechnical systems and practices. It does so by evaluating their respect for the context-relative informational norms at play in a given context. But can CI evaluate new sociotechnical systems that severely disrupt established social practices? In this paper, I argue that, while CI claims to be able to assess privacy violations of all sociotechnical systems and practices, it cannot assess the ones that cause severe changes and disruptions in the norms and values of a given context. These types of technology are known as socially disruptive technologies (SDTs) and this paper argues that they are beyond CI’s scope. It follows that at best, a privacy assessment of those technologies by CI would be useless and, at worst, lead to potential harm, including failure to identify privacy violations and unwarranted legitimisation of privacy-threatening technology. Government actors, policymakers, and academics should refrain from relying on CI to assess this type of technology.

The promise and ambition of emotion recognition software is to provide a technological tool to access the inner self of others; their feelings, apprehensions, joys and passions. For several years now, the demand for emotion recognition software (ERS) has been increasing drastically in both countries and private companies, especially in the areas of security and consumer profiling. Exploring the relationships between emotion recognition software (ERS) and humans, and its impact on privacy is thus imperative. In doing so, we are forced to question the prominent frameworks of privacy in terms of their current ability to protect our inner affective states. The main challenge is to use the tools and knowledge that philosophy of technology, mind and morality offer me to test and specify the limits, if there are any, to the use of this technology so that it remains within the scope of what we consider morally acceptable in our society. The idea is to develop a comprehensive solution, within the boundaries of a normative framework, so it can be a useful guide for political legislators to draw upon when legal decisions on this matter will be required.

Dans cet article, nous proposons une nouvelle avenue pour répondre en partie au casse-tête du deuil posé par Marušić dans Do Reasons Expire ? An essay on grief. Selon Marušić, le deuil ne peut être expliqué rationnellement, entre autres, par ce qu’il appelle la conception processuelle. Notre proposition est fondée sur une conception de l’identité généralement absente dans les approches occidentales de l’identité : une conception dividuelle. Les conceptions dividuelles de l’identité présentent l’identité d’un individu comme étant un composite de ses relations interpersonnelles, c’est-à-dire qu’autrui est en partie moi, et je suis en partie autrui. À partir de cette conception, nous réintroduisons une nouvelle version de la conception processuelle ayant pour objet le manque de rationalité cognitive entre la partie notre identité constituée par le proche faisant l’objet de notre deuil et son référent dans le monde (le proche maintenant décédé). Selon notre proposition, la diminution du deuil serait ainsi explicable rationnellement par le réajustement de l’identité à son référent. Les particularités de la conception de l’identité et de la conception processuelle que nous proposons permettent d’éviter les critiques que Marušić adresse à la conception processuelle. Cette conception nous permet ainsi d’expliquer rationnellement la diminution du deuil, du moins en partie.

Emotion recognition systems (ERS) offer the ability to identify the emotions of others, based on an analysis of their facial expressions and regardless of culture, ethnicity, context, gender or social class. By claiming universalism in the expression as well as in the recognition of emotions, we believe that ERS present significant risks of causing great harm to some individuals, in addition to targeting, in some contexts, specific social groups. Drawing on a wide range of multidisciplinary knowledge - including philosophy, psychology, computer science and anthropology - this research project aims to identify the current limitations of ERS and the main risks that their use brings, with the goal of providing a clear and robust analysis of the use of ERS and their contribution to greater social justice. Pointing to technical limitations, we refute, on the one hand, the idea that ERS are able to prove the causal link between specific emotions and specific facial expressions. We support our argument with evidence of the inability of ERS to distinguish facial expressions of emotions from facial expressions as communication signals. On the other hand, due to the contextual and cultural limitations of current ERS, we refute the idea that ERS are able to recognise, with equal performance, the emotions of individuals, regardless of their culture, ethnicity, gender and social class. Our ethical analysis shows that the risks are considerably more numerous and important than the benefits that could be derived from using ERS. However, we have separated out a specific type of ERS, whose use is limited to the field of care, and which shows a remarkable potential to actively participate in social justice, not only by complying with the minimum requirements, but also by meeting the criterion of beneficence. While ERS currently pose significant risks, it is possible to consider the potential for specific types to participate in social justice and provide emotional and psychological support and assistance to certain members of society.