Cultivating Moral Attention: a Virtue-Oriented Approach to Responsible Data Science in Healthcare

Abstract

In the past few years, the ethical ramifications of AI technologies (in particular data science) have been at the center of intense debates. Considerable attention has been devoted to understanding how a morally responsible practice of data science can be promoted and which values have to shape it. In this context, ethics and moral responsibility have been mainly conceptualized as compliance to widely shared principles. However, several scholars have highlighted the limitations of such a principled approach. Drawing from microethics and the virtue theory tradition, in this paper, we formulate a different approach to ethics in data science which is based on a different conception of “being ethical” and, ultimately, of what it means to promote a morally responsible data science. First, we develop the idea that, rather than only compliance, ethical decision-making consists in using certain moral abilities (e.g., virtues), which are cultivated by practicing and exercising them in the data science process. An aspect of virtue development that we discuss here is moral attention, which is the ability of data scientists to identify the ethical relevance of their own technical decisions in data science activities. Next, by elaborating on the capability approach, we define a technical act as ethically relevant when it impacts one or more of the basic human capabilities of data subjects. Therefore, rather than “applying ethics” (which can be mindless), data scientists should cultivate ethics as a form of reflection on how technical choices and ethical impacts shape one another. Finally, we show how this microethical framework concretely works, by dissecting the ethical dimension of the technical procedures involved in data understanding and preparation of electronic health records.

Mapping the ethic‐theoretical foundations of artificial intelligence research

“…To the researchers, the most striking feature of the literature map is its fractured composition. There are only three lines of citations: [1] Khargonekar and Sampath (2020)–Mezgár and Váncza (2022), [2] Bryson (2018)–Constantinescu et al (2021), and [3] Hagendorff (2020)–Ratti and Graves (2021). There appear to be no shared ethic‐theoretical foundations between the publications in [1], but in both [2] and [3] the papers adopt a virtue‐based ethical stance.…”

“…Rather than wrestling with attempting to develop or adopt “the perfect solution” it is perhaps more important that stakeholders are concerned with understanding and communicating the principles upon which systems are based, since, as Mezgar and Vancza (2022, p. 402) claimed “the real threat of AI is … unexplained decisions.” To this end, DeBellis (2018, p. 247) maintain “any ethical system must start with … a statement of what the individual values … because that is the way they desire the world to be.” This places the burden of responsibility, at least initially, upon those that are tasked with the construction of any system. As Ratti and Graves (2021, p. 1819) suggest, “data scientists should cultivate ethics as a form of reflection on how technical choices and ethical impacts shape one another.” The onus of responsibility also extends to those whom commission or adopt such systems since “it is key for ‘those with the power to choose’ to remember that their decisions will shape future individual and societal outcomes” (Khargonekar & Sampath, 2020, p. 17013). We posit that valuable empirical research could be undertaken within organizations that are at the vanguard of AI development and implementation.…”

“…This places the burden of responsibility, at least initially, upon those that are tasked with the construction of any system. AsRatti and Graves (2021, p. 1819) suggest, "data scientists should cultivate ethics as a form of reflection on how technical choices and ethical impacts shape one another." The onus of responsibility also extends to those whom commission or adopt such systems…”

What ethics can say on artificial intelligence: Insights from a systematic literature review

“…It involves using moral exemplars, meeting community expectations, and practical simplicity to develop computationally rational Artificial Moral Agents (AMAs). Ratti and Graves (2021) argue that instead of relying solely on adherence to widely accepted principles, they advocate for cultivating moral virtues within the practice of data science. This perspective emphasizes that ethical decision‐making involves the development of moral abilities or virtues through practice and posit the concept of “moral attention,” which consists of the capability of data scientists to discern the ethical implications of their work.…”

What Do We Teach to Engineering Students: Embedded Ethics, Morality, and Politics

“…Preventive ethics is usually formulated in terms of rules, which can be problematic for a number of reasons. First, the emphasis on rules suggests that all you need for being ‘ethical’ is to know the rules themselves, which can be mindlessly applied (Ratti & Graves, 2021 ). This ‘compliance’ paradigm of rule-based ethics forces a legalistic interpretation of rules, which in turn promotes loop-hole reasoning and a perception of ethics as externally imposed (Kelly, 2018 ).…”