When Exceptions Are the Norm

Abstract

HRI researchers have made major strides in developing robotic architectures that are capable of reading a limited set of social cues and producing behaviors that enhance their likeability and feeling of comfort amongst humans. However, the cues in these models are fairly direct and the interactions largely dyadic. To capture the normative qualities of interaction more robustly, we propose "consent" as a distinct, critical area for HRI research. Convening important insights in existing HRI work around topics like touch, proxemics, gaze, and moral norms, the notion of consent reveals key expectations that can shape how a robot acts in social spaces. Consent need not be limited to just an explicit permission given in ethically charged or normatively risky scenarios. Instead, it is a richer notion, one that covers even implicit acquiescence in scenarios that otherwise seem normatively neutral. By sorting various kinds of consent through social and legal doctrine, we delineate empirical and technical questions to meet consent challenges faced in major application domains and robotic roles. Attention to consent could show, for example, how extraordinary, norm-violating actions can be justified by agents and accepted by those around them. We argue that operationalizing ideas from legal scholarship can better guide how robotic systems might cultivate and sustain proper forms of consent. CCS Concepts: • Human-centered computing → Interaction design;

Sexuality in the era of transhumanism: A point of view

“…The main job lies in understanding the context in which the robot is exposed. Besides, the social environment's impact includes the possible trust that the robot generates in some humans (Cominelli et al, 2021; Jung & Hinds, 2018; Sarathy et al, 2019). Likewise, a modulation in phycological responses to stress has been described after a haptic interaction initiated by a robot over human and positive reactions in humans with the exposure of facial and verbal expressions of happiness and surprise by social robots (Chuah & Yu, 2021; Willemse & van Erp, 2019).…”

Understanding the spirit of a norm: Challenges for norm‐learning agents

“…In previous work, we have argued that exceptions to certain ways of fulfilling norms can actually constitute part of those norms, and indeed that competence with norms often means knowing common exceptions (a waiter clearing a plate from the table seeing that a diner is trying to eat more off it) (Sarathy, Arnold, and Scheutz 2019). While in some cases, there might be repairs or adjustments to navigate a norm‐informed interaction (“I'm sorry, I thought you were done with your meal”), the “spirit” of a norm represents an implicit set of presuppositions that norms entail in uncertain or unexpected circumstances.…”

“…This seems especially critical to tackle given the different roles of sanctioning across social and/or moral norms, including how artificial agents relate to sanctioning practices (Sarathy, Arnold, and Scheutz 2019; Jackson and Williams 2019). There are a host of social attributions that could lead to robots eliciting confusion or incurring blame as agents in shared space, but for norm learning, the primary question might be how observed sanctioning contributes to building a properly scoped norm representation?…”

Mitigating Adversarial Norm Training with Moral Axioms

“…There has been a recent surge of interest in machine learning of norms. Like the framework we build on, Sarathy et al (2017) use DS theory for norm learning. Forbes et al (2020) instead take a neural approach.…”