Autonomy has been central to moral and political philosophy for millenia, and has been positioned as a critical aspect of both justice and wellbeing. Research in psychology supports this position, providing empirical evidence that autonomy is critical to motivation, personal growth and psychological wellness. Responsible AI will require an understanding of, and ability to effectively design for, human autonomy (rather than just machine autonomy) if it is to genuinely benefit humanity. Yet the effects on human autonomy of digital experiences are (...) neither straightforward nor consistent, and are complicated by commercial interests and tensions around compulsive overuse. This multi-layered reality requires an analysis that is itself multidimensional and that takes into account human experience at various levels of resolution. We borrow from HCI and psychological research to apply a model (“METUX”) that identifies six distinct spheres of technology experience. We demonstrate the value of the model for understanding human autonomy in a technology ethics context at multiple levels by applying it to the real-world case study of an AI-enhanced video recommender system. In the process we argue for the following three claims: 1) There are autonomy-related consequences to algorithms representing the interests of third parties, and they are not impartial and rational extensions of the self, as is often perceived; 2) Designing for autonomy is an ethical imperative critical to the future design of responsible AI; and 3) Autonomy-support must be analysed from at least six spheres of experience in order to approriately capture contradictory and downstream effects. (shrink)

In 2019, the IEEE launched the P7000 standards projects intended to address ethical issues in the design of autonomous and intelligent systems. This move came amidst a growing public concern over the unintended consequences of artificial intelligence (AI), compounded by the lack of an anticipatory process for attending to ethical impact within professional practice. However, the difficulty in moving from principles to practice presents a significant challenge to the implementation of ethical guidelines. Herein, we describe two complementary frameworks for integrating (...) ethical analysis into engineering practice to help address this challenge. We then provide the outcomes of an ethical analysis informed by these frameworks, conducted within the specific context of internetdelivered therapy in digital mental health. We hope both the frameworks and analysis can provide tools and insights, not only for the context of digital healthcare, but for data-enabled and intelligent technology development more broadly. (shrink)

Concerns about threats to human autonomy feature prominently in the field of AI ethics. One aspect of this concern relates to the use of AI systems for problematically manipulative influence. In response to this, the European Union’s draft AI Act (AIA) includes a prohibition on AI systems deploying subliminal techniques that alter people’s behavior in ways that are reasonably likely to cause harm (Article 5(1)(a)). Critics have argued that the term ‘subliminal techniques’ is too narrow to capture the target cases (...) of AI-based manipulation. We propose a definition of ‘subliminal techniques’ that (a) is grounded on a plausible interpretation of the legal text; (b) addresses all or most of the underlying ethical concerns motivating the prohibition; (c) is defensible from a scientific and philosophical perspective; and (d) does not over-reach in ways that impose excessive administrative and regulatory burdens. The definition is meant to provide guidance for design teams seeking to pursue responsible and ethically aligned AI innovation. (shrink)

Promoting the use of contact tracing technology will be an important step in global recovery from the COVID-19 pandemic. Across two studies, we assessed two messaging strategies as motivators of intended contact tracing uptake. In one sample of 1117 Australian adults and one sample of 888 American adults, we examined autonomy-supportive and controlling message framing and the presence or absence of information safety as predictors of intended contact tracing application uptake, using an online randomized 2 × 2 experimental design. The (...) results suggested that the provision of data safety assurances may be key in affecting people’s intentions to use contact tracing technology, an effect we found in both samples regardless of whether messages were framed as autonomy-supportive or controlling. Those in high information safety conditions consistently reported higher intended uptake and more positive perceptions of the application than those in low information safety conditions. In Study 2, we also found that perceptions of government legitimacy related positively to intended application uptake, as did political affiliation. In sum, individuals appeared more willing to assent to authority regarding contact tracing insofar as their data safety can be assured. Yet, public messaging strategies alone may be insufficient to initiate intentions to change behavior, even in these unprecedented circumstances. (shrink)

Responsible AI (RAI) guidelines aim to ensure that AI systems respect democratic values. While a step in the right direction, they currently fail to impact practice. Our work discusses reasons for this lack of impact and clusters them into five areas: (1) the abstract nature of RAI guidelines, (2) the problem of selecting and reconciling values, (3) the difficulty of operationalising RAI success metrics, (4) the fragmentation of the AI pipeline, and (5) the lack of internal advocacy and accountability. Afterwards, (...) we introduce a number of approaches to RAI from a range of disciplines, exploring their potential as solutions to the identified challenges. We anchor these solutions in practice through concrete examples, bridging the gap between the theoretical considerations of RAI and on-the-ground processes that currently shape how AI systems are built. Our work considers the socio-technical nature of RAI limitations and the resulting necessity of producing socio-technical solutions. (shrink)

New research on affective computing aiming to develop computer systems that recognize and respond to affective states can also contribute to the issues raised by Coan. Research on how humans interact with computers, and computer models that automatically recognize affective states from features in our physiology, behaviour, and language, may provide insights on how emotions that are experienced and expressed come to be. For example, there is empirical evidence that affect recognition techniques using several modalities are more accurate than those (...) using a single modality, but it is not clear if these improvements are caused by superadditivity (i.e., emergence) or redundancy. (shrink)