Ethical Reflections on Artificial Intelligence

Public and academic discourse about the ethics of artificial intelligence, machine learning, and data science has largely focused on the algorithms and the companies deploying them. Little attention has been paid to the ethical agency of the developers. This study is the first of its kind that centers developers in the ethical environment. Semi-structured interviews with 40 developers about the ethics of being a developer revealed more than 20 themes, 3 of which are the subject of this paper: ethics in the occupational ecosystem, developer ethical agency, and the characteristics of an ethical developer. These themes reveal significant gaps between how developers perceive themselves and the reality of their work experiences. Their ethical agency is likewise variable. They have some authority to intervene for ethical reasons in systems they work on, but they often do not realize just how many ethical decisions they make. Nonetheless, this study reveals a growing ethical wisdom in this community, one that needs to be surfaced and nurtured by engaging with developers.

This paper explores ethical wisdom in the artificial intelligence (AI) developer community. Despite robust literature about the need for virtue ethics approaches in AI development, little research has directly engaged with the developer community about their progress in this regard. We have thus conducted semi-structured interviews with a worldwide cohort of 40 developers, which focused on their awareness of ethics issues, how they navigate ethical challenges, and the barriers they encounter in developing ethical wisdom. We find developers are largely aware of the ethical territories they must navigate and the moral dilemmas they personally encounter, but they face limited and inconsistent resources for ethical guidance or training. Furthermore, there are significant barriers inhibiting the development of ethical wisdom in the AI developer community, including the industry’s fixation on innovation, the narrow scope of technical practice, limited provisions for reflection and dialogue, and incentive structures that prioritize profits and prestige. The paper concludes by emphasizing the need to address the gap in domain-specific ethical skill and provides recommendations for organizations, educators, and the AI developer community.

There are two types of artificial general intelligence (AGI) safety solutions: global and local. Most previously suggested solutions are local: they explain how to align or “box” a specific AI (Artificial Intelligence), but do not explain how to prevent the creation of dangerous AI in other places. Global solutions are those that ensure any AI on Earth is not dangerous. The number of suggested global solutions is much smaller than the number of proposed local solutions. Global solutions can be divided into four groups: 1. No AI: AGI technology is banned or its use is otherwise prevented; 2. One AI: the first superintelligent AI is used to prevent the creation of any others; 3. Net of AIs as AI police: a balance is created between many AIs, so they evolve as a net and can prevent any rogue AI from taking over the world; 4. Humans inside AI: humans are augmented or part of AI. We explore many ideas, both old and new, regarding global solutions for AI safety. They include changing the number of AI teams, different forms of “AI Nanny” (non-self-improving global control AI system able to prevent creation of dangerous AIs), selling AI safety solutions, and sending messages to future AI. Not every local solution scales to a global solution or does it ethically and safely. The choice of the best local solution should include understanding of the ways in which it will be scaled up. Human-AI teams or a superintelligent AI Service as suggested by Drexler may be examples of such ethically scalable local solutions, but the final choice depends on some unknown variables such as the speed of AI progress.

Over 2000 years the Catholic Church has slowly developed a posture towards technology which is predominantly techno-optimist and techno-progressive, and yet the Church does not have this reputation today. Concomitantly, Church institutions and individuals have made crucial contributions to the advance of science and technology, yet despite this practical effort to better human development, Christian theology has been remarkably uninterested in the subject of technology. This lack of interest is no longer tenable; scholars of religion and theologians should seriously engage technology because it is empowering humanity in ways that were previously reserved only for gods. This blind spot has not only hampered the Church’s ability to understand itself and our world, but also impeded the ability of the Church to fulfill its mission. Pope Francis’s 2015 encyclical Laudato Si has begun to address this neglect, but is best understood in the context of Christian history, not only as written, but more so as practiced.