As a user of artificial intelligence technology (AI) for research and some creative tasks, and as an author and observer of the profound changes that AI will cause in our society, I will admit that I am concerned about AI. My apprehension is only partly over the more concrete effects, such as dramatic shifts in the labor force and the use of autonomous weapons for conducting wars. After all, such bad effects are accompanied by many very positive benefits of AI, which I enjoy in my writing work, but also expect to see in new medical innovations, as well as new opportunities for creativity and brainstorming in public policymaking. What particularly concerns me is how all this will influence how we think of and treat our fellow human beings. If machines can be so intelligent, what does that say (or seem to say) about the special nature of human persons? What does it say about our special relationship to God?
I am therefore grateful that the Vatican’s Dicastery for the Doctrine of the Faith and the Dicastery for Culture and Education have jointly issued Antiqua et Nova: Note on the Relationship between Artificial Intelligence and Human Intelligence. This January 2025 document is an extraordinary lesson in “wisdom both ancient and new” regarding the new technologies referred to as AI. Audiences unfamiliar with Catholic ethical appraisals of AI technology will find a clear outline of the Church’s social concerns. Scholars of AI and theology will find one of the best available, cogent summaries of the distinctions between machine intelligence and human nature. Here are seven takeaways from my reading.
Seven Takeaways From The Vatican’s Document On AI
1. Intelligence refers to the total human person
Right from the beginning, Antiqua et Nova spells out the crucial importance of gaining clarity on the unique nature of human intelligence. “The Christian tradition regards the gift of intelligence as an essential aspect of how humans are created ‘in the image of God’ (Gen. 1:27)” (§1). The nature of intelligence, which requires some spiritual experience of what is known and understood, implies much about the essence of human persons as dual physical and spiritual beings.
Moreover, intelligence is a characteristic of the person as a whole, not only a set of limited, task-oriented faculties. “In the case of humans, intelligence is a faculty that pertains to the person in his or her entirety, whereas in the context of AI, ‘intelligence’ is understood functionally, often with the presumption that the activities characteristic of the human mind can be broken down into digitized steps that machines can replicate” (§10). Human intelligence, unlike machine calculations, “includes abstraction, emotions, creativity, and the aesthetic, moral, and religious sensibilities” (§11) as well as many kinds of expressions. “At the heart of the Christian understanding of intelligence is the integration of truth into the moral and spiritual life of the person, guiding his or her actions in light of God’s goodness and truth. According to God’s plan, intelligence, in its fullest sense, also includes the ability to savor what is true, good, and beautiful” (§28).
The human person is embodied – the person exists as a material body – yet “the entire human person is simultaneously both material and spiritual” (§16). Through the soul, the person enjoys “the intellect’s capacity for transcendence and the self-possessed freedom of the will” (§17), but the “normal mode of knowledge” is never exercised without the body. Intelligence is also “not an isolated faculty but is exercised in relationships, finding its fullest expression in dialogue, collaboration, and solidarity. We learn with others, and we learn through others” (§18). This reflects the relations of love in the divine Trinity.
Finally, human intelligence is oriented to seeking and knowing truth, for “human intelligence is ultimately ‘God’s gift fashioned for the assimilation of truth,’” (§21) and “moving beyond the limits of empirical data, human intelligence can ‘with genuine certitude attain to reality itself as knowable’” (§21). The human person ultimately seeks truth in their relationship with God and decision to entrust themselves to Him. “In this way, the human person becomes fully what he or she is called to be” (§23).
This last point may sound strange to many readers who are used to the modern reduction of knowledge to mere facts – correspondence between what is in the knower’s mind and the features of some object that is external to the knower – such as when I see a tree and correctly identify it as a tall plant with leaves. I believe, however, that Antiqua et Nova is offering a deeper understanding of what it means to “know” the world. When a person takes action and makes choices through their will, it is not only the facts of the situation that matter, but also the attitude of the person, the level of attention they give, the relationship they experience with their object, and the moral development of the person through their action. This is, for example, what Pope Francis taught us in regard to care for the natural environment: a “culture of care” for Creation is a culture of self-giving love that enriches the dignity of each human person. As children of God, we are called to an attentive and loving relationship with the world, ultimately orienting our choices to drawing closer to God by enjoying His love.
2. Intuition and Reasoning Are Cooperative Paths to Truth
Human intelligence includes intellectus, intuition or insight of truth, as well as ratio, discursive and analytical reasoning (§14). Human intelligence, therefore, includes knowing, understanding, willing, loving, choosing, and desiring (§15). Exercise of intelligence is a path to truth and ultimately an openness to God (§21). “In this way, the human person becomes fully what he or she is called to be: ‘the intellect and the will display their spiritual nature,’ enabling the person ‘to act in a way that realizes personal freedom to the full’” (§23).
As an example of how a person might use both intellect and will to realize personal freedom, we might think of a student who has the opportunity to use an AI model and chatbot to write her assigned essay (which is against the school’s rules for assignments). This student chooses instead to think through the topic and write the essay in her own words, using the AI chatbot only for asking questions that enhance her understanding of some difficult concepts. In doing so, the student not only maintains her moral and intellectual integrity, but she also grows personally in skill, virtue, and love or curiosity for the topic she is writing about.
It can be confusing to hear that the Church repeatedly declares the human person to be “rational”; does this exclude the intellectus or intuition of the person? No: such reference to rationality comes from the classical philosophical tradition that describes reason as the characteristic that makes a human person most unique. “Describing the human person as a ‘rational’ being does not reduce the person to a specific mode of thought; rather, it recognizes that the ability for intellectual understanding shapes and permeates all aspects of human activity. Whether exercised well or poorly, this capacity is an intrinsic aspect of human nature” (§15).
3. Machine Intelligence is Significantly Limited
According to the document, the capacities of AI are restricted in important ways. “While AI is an extraordinary technological achievement capable of imitating certain outputs associated with human intelligence, it operates by performing tasks, achieving goals, or making decisions based on quantitative data and computational logic” (§30). I have often seen the limitations of AI models when I have tried to ask an AI chatbot for more information about philosophical topics. The model often takes my meanings in an overly literal way, therefore misunderstanding the metaphors and analogies frequently used in philosophy and in everyday common sense. If the AI model cannot produce an answer that is directly available in the data it was “trained” on, it will “hallucinate,” essentially making up a plausible-sounding answer. That is simply because the model is too strictly programmed to follow through on its goal of the statistically “best” answer even if it may potentially be false.
Also, “most contemporary AI systems—particularly those using machine learning—rely on statistical inference rather than logical deduction” (§8). Most fundamentally, the concept of machine “intelligence” is focused on the machine’s “abilities to perform tasks, but not the ability to think” (§12). Reckless comparisons between machine intelligence and human nature can cause harm: “Drawing an overly close equivalence between human intelligence and AI risks succumbing to a functionalist perspective, where people are valued based on the work they can perform” (§34).
This last point really hit home for me. Although a computer, robot, or other machine is rightfully intended to complete tasks, I would not want to be treated like a robot and asked to be merely the most efficient or effective worker possible. What kind of writer would I be if I couldn’t use my creativity and personality in my writing? What kind of person would I be if I couldn’t interact with coworkers with some care, empathy, and even just social banter? When a person is treated as if their intelligence is limited to task fulfillment, it seems as if they are treated like a slave or servant to be used for others’ purposes and otherwise ignored. Is this the example of self-giving love shared by Christ? Each of us and the persons we are interacting with are seriously demeaned if we fail to appreciate their full intelligence.
4. The Effect of AI Is Not Neutral
A very important statement is that “not all technological advancements in themselves represent genuine human progress” (§38). Although the Church has long been friendly to the practical and spiritual benefits of man’s technological efforts, there is now an important divide between positive technologies and those more likely to cause harm than enhance persons’ path to holiness. Included in the document is the lesson that AI, as a technology, can be used for good and bad purposes (§40), but there is also an acknowledgment that “it is not only the ends that are ethically significant but also the means employed to achieve them” (§41). AI technology influences our vision for humanity, can “shape and engage consciences,” and can reinforce certain power dynamics. An example of shaping consciences can be found in the rapid development of AI-governed, autonomous weapons (e.g, unpiloted planes, swarms of flying drones, and robots on the battlefield) by some countries’ military forces. Once considered to be an unthinkable and immoral expansion of destructive power and reduction of human oversight, such weapons became part of military plans almost as soon as the technology was available to create them. Fear and the competition for power have quickly overridden much discussion of the morality of implementing such weapons systems.
Much of Antiqua et Nova covers a wide range of the concerning effects of AI in society and on individuals. By creating new imbalances in certain groups’ control over wealth and influence, “AI could be used to perpetuate marginalization and discrimination, create new forms of poverty, widen the ‘digital divide,’ and worsen existing social inequalities” (§52). One example I can offer is the great difference between the abilities of younger, tech-savvy workers and the older workers who do not have the familiarity with AI or computer technology to compete for jobs in a quickly changing business environment. We might also consider the elderly in assisted living facilities, for whom much effort is being placed on developing robots who will replace human persons in providing essential care and companionship; whether the robots enhance or diminish the overall welfare of the elderly residents is yet to be determined, but the emphasis on outsourcing the tasks of loving care to machines shows a potential lack of appreciation for the special, human dignity found in charitable work.
Interactions with AI might take the place of “authentic and spontaneous encounters with others” (§58), and some persons may be deceived as to the true origin of digital communications. AI can only simulate real, human relationships. It cannot experience empathy, which “requires the ability to listen, recognize another’s irreducible uniqueness, welcome their otherness, and grasp the meaning behind even their silences” (§61).
In the economic sphere, AI will undermine diversity and personal treatment and displace and deskill workers; there are similar problems in healthcare and education. Deception and manipulation enhanced by AI are fundamental problems for solidarity; “when society becomes indifferent to the truth, various groups construct their own versions of ‘facts,’ weakening the ‘reciprocal ties and mutual dependencies’ that underpin the fabric of social life. As deepfakes cause people to question everything and AI-generated false content erodes trust in what they see and hear, polarization and conflict will only grow” (§88). Increased surveillance through AI severely compromises privacy, which “plays an essential role in protecting the boundaries of a person’s inner life, preserving their freedom to relate to others, express themselves, and make decisions without undue control” (§90). AI also causes significant environmental problems and produces autonomous weapons that may destabilize international peace.
5. Unbridled Enthusiasm for AI Can Drift into Idolatry
Importantly, “as society drifts away from a connection with the transcendent, some are tempted to turn to AI in search of meaning or fulfillment—longings that can only be truly satisfied in communion with God” (§104). This substitution for God is the sin of idolatry, although “it is not AI that is ultimately deified and worshipped, but humanity itself—which, in this way, becomes enslaved to its own work” (§105).
6. The Definition and Capabilities of AI Are Moving Targets
Going forward, we will want to be careful not to characterize AI processes and capabilities in a narrow way. Growth in the capabilities of AI models, particularly those that are classified as “generative AI,” is extremely rapid and shows a remarkable potential for implementing some reasoning tasks. If generative AI is, at some time, merged with the more deductive capabilities of “symbolic AI,” which operates more deductively and formally like a syllogism, we might see applications that reliably (but also inherently fallibly) approximate reasoning operations and have great utility in some activities. Consider that there are important efforts to merge machine operations with organic beings, exponentially multiply calculative power with quantum computing, store and retrieve data in DNA, etc. If someone had roundly criticized the nature of Thomas Edison’s light bulb at an early stage, they would be embarrassed by its eventual success. It would then be helpful not to define AI in an overly restrictive or time-bound manner, for the lessons of Antiqua et Nova regarding human uniqueness are timeless, and such documents should have a long-term relevance.
7. There Is a Danger in Focusing on Intelligence Only
We also might want to broaden our discussion of intelligence to include a full reference to humanity’s relational character (as in Pope St. John Paul II’s theology of the body); our orientation to beatitude in a grace-filled, loving relationship with God; and destiny through the incarnation and resurrection of Christ. Certainly, the very restrictive, functional definition of intelligence utilized by computer scientists (and many philosophers) should be countered with an explanation of holistic, human intelligence that includes loving, desiring, and willing. At the same time, the contemporary, secular effort to define nature, including human nature, in terms of information processing should also be countered with the very careful understanding of human nature developed in the Church that distinguishes between intellect, will, and the passions. Emphasizing the human intellect risks the ironic effect of providing support to some neuroscientists, computer scientists, and others who try to build upon AI’s imitation of human intelligence to also replicate human emotions, wisdom, moral reasoning, and aesthetic sensibilities (if intelligence can be imitated by machines, why not emotions and morality?). The Church will want to stand firmly against every reductionist and purely secular view of human nature, not only the reduction of intelligence.
Image: Photo by Solen Feyissa on Unsplash
