The Promise and Peril of AI and IA. New Technology Meets Religion, Theology, and Ethics

The arrival of Artificial Intelligence (AI) and its derivatives at the center of our civilization comes at a time of great uncertainty, making a sober and non-judgmental discussion indispensable. A highly valuable contribution to such a discussion is this essay collection. With 27 contributions, the 2023 book offers a wealth of theoretical insights into one of the most vibrant fields of the so-called Fourth Industrial Revolution. The volume covers not only AI but also AI-based applications that promise Intelligence Amplification (IA) for humans—a topic seldom explored in theological or religious research. Part One (“AI, IA, and Frankenfear”) contains the kind of meta-reflection that is a typical feature of any worthwhile theology or philosophy of culture: What does it mean that our civilization is currently implementing a number of new technologies around the AI paradigm, knowing that they will all have serious side effects, but being unable to define and describe the future impact exactly? Part Two (“AI and IA in Culture, Religion, and Ethics”) broadens the focus, as texts now argue from very different angles, for example, from the perspective of Islamic or Confucian thought, or reflect on different application contexts. Part Three (“Is There Really an “I” in “AI” or “A”?”) takes us back to a level of principal reflection, by focusing on the notion of intelligence. There is a tacit agreement not only in the first part but throughout the whole book: From a religious perspective, there is little cause for exaggerated euphoria. Rather, as Ted Peters puts it, “a healthy dose of Frankenfear, that is, caution” is warranted (17). He poses the right question in this context: Can we seriously believe that a sinful humanity will create something completely sinless, without any negative aspects (43)? Tracy Trothen and Calvin Mercer demonstrate in their contribution that both anthropomorphizing and deifying AI would be erroneous—and we can add with Peters, so would any temptation to “personify” it (15). Even in its most elaborated forms, AI does not transcend the principal limits set to the machine world. Let us take robots run by advanced forms of AI as an example: Morally, they will always depend on their programmers and the data input by means of which they train themselves. It would be false to expect them to act spontaneously or to behave compassionately when they carry out what their software tells them to do. However, even if the most advanced “robots cannot look us in the eye” (156), the encounter with them can involve several spiritual aspects, as Trothen and Mercer admit. Thus, under certain very positive circumstances, AI can even become a “source of healing and grace” (162). One of the most commendable features of this extraordinarily rich volume which I only can present selectively here is its openness to non-Christian perspectives. This is still a rarity in the discourse on AI and faith. Since AI in its concrete forms always depends on its underlying data reference, it is a highly biased technology undoubtedly tending to iterate prevailing mechanisms of social exclusion. It is an important take-home message of the volume that we had better not repeat such mechanisms. In this context, Ali-Reza Bhojani’s contribution reminds us of the up to now clearly underestimated “significance of Islam with its over two billion adherents worldwide to developers, regulators, and users of AI technologies” (246), while Lawrence A. Whitney explains the reasons for which a civilization based on Confucianist values like China so often shows greater flexibility and tolerance toward the latest trends in the AI world than, for example, large parts of the European population. The decisive point is that Confucianism is not a religion of abstract values and abstract virtue concepts. Rather, it emphasizes the alignment with given social roles, rules, and circumstances—a factor favorable for smoothly integrating AI-based tools in humans’ everyday life. This, however, will not necessarily foster their spiritual well-being, Hermina Nedelescu is convinced. Drawing on the Orthodox theosis tradition, she argues that even the most intentional implementations of AI “should not be expected to strengthen or weaken our soul's capacity to center the self, let alone recenter the self in God” (363). A true reshaping of our soul cannot come from an external stimulus, which AI will always remain. It must come both from inside and outside us—from God's Holy Spirit, to whom we owe our theosis. Which topics remain underrepresented or unaddressed in this book, despite its wealth of fascinating themes? First, the contributors pay relatively little attention to possible or already existing implementations of AI in medical, clinical, and laboratory contexts, although this is the field where the potential opportunities and risks of the new technological paradigm are most closely intertwined. Second, the massive economic expectations based on AI have been left aside. Today, almost every economy in the world places desperate hopes on the new technology to raise productivity rates. AI is expected to help us reduce the harm demographic change causes to our labor markets. But who can guarantee that these hopes will be fulfilled? And if they prove to be empty, who will help us cope with the disappointment that will then arise?