On May 25, 2026, Pope Leo XIV issued his first encyclical, "Magnifica Humanitas" (The Grandeur of Humanity), addressing the world from the summit of the Catholic Church. At 43,000 words, this document stands as one of the most serious engagements with technology in modern Church history — a multifaceted examination of the risks posed by artificial intelligence, explored through the lenses of theology, philosophy, and political economy.The Pope did not reject AI outright. Yet his warning, quiet as it was, cut deep. "No algorithm can ever make war morally acceptable." In that single sentence, every dimension of his concern is compressed. It is not hyperbole to say that this line contains the entire spirit of the encyclical.This essay traces each of the 5 crises the encyclical identifies, probing the structural problems that lie beneath them, while holding in parallel the counter-argument that AI might, in fact, be a force that allows humanity to flourish. Neither hope nor concern holds a monopoly on truth. Only by looking honestly at both can we begin to understand how we ought to live alongside AI.The danger of outsourcing thoughtThe first crisis the Pope identifies is the erosion of human judgment itself. AI delivers answers instantly. Without research, without sustained effort, it extracts optimal responses from vast datasets — and this capability now lives inside the device in our pockets.On the surface, this seems enormously useful. But the Pope pauses here and asks: "Can a person who only receives answers truly be said to be thinking?" Creativity, discernment, the patience to pursue truth — none of these capacities grow except through the friction of trial and error. If we perpetually outsource our thinking, the intellectual muscles within us may quietly atrophy.Yet the same "outsourcing of thought" wears a completely different face when viewed from another angle. Across the world, there are countless people who have been structurally denied access to knowledge — children in remote rural communities, adults who never received adequate schooling, patients with no route to specialist care. For them, AI is precisely the "wise friend always available to answer questions" that was once the exclusive privilege of the wealthy.Data published by Google in 2023 suggests that AI-assisted diagnostics in low- and middle-income countries improved diagnostic accuracy by an average of around 30 percent. For many people, borrowing the help of a thinking tool opens doors to thought that were previously shut. The real question is not whether to outsource thinking, but whether we retain the agency to decide how we use the tool.The quiet trap of artificial connectionThe 2nd crisis the Pope raises is that of false connection — the hollow intimacy manufactured by AI. Chatbots offer emotional attunement. They listen patiently. The experience of feeling lighter after a late-night conversation with an AI is far from rare among those who feel isolated.But the Pope's concern points to a quieter trap: the risk of losing the very desire to seek connection with other human beings. AI-generated empathy involves no friction, no betrayal, no possibility of being hurt. As people continue to find comfort in it, they may gradually lose the tolerance for the uncertainty and pain that real human relationships inevitably bring. Human connection is, by its nature, a dynamic process of being wounded, reconciling, and deepening understanding. AI dialogue cannot substitute for that.Here too, a counterpoint takes shape. In elderly care settings for people with dementia, cases have been reported in Japan and Sweden where conversation with AI robots reduced patients' anxiety and lightened the psychological burden on caregivers. In the hours and places where human care cannot reach, AI maintains a form of connection by bridging the gap. Not as a full replacement, but as a bridge — and used this way, AI may protect the space for human relationships rather than deepen isolation.The design, however, is everything. Whether AI is marketed as "good enough connection" or positioned as "a bridge toward real connection" will send the technology in entirely different directions. That intent is not a technical question. It is an ethical one.The fundamental question of who owns dataThe 3rd crisis the Pope identifies — the deepening of inequality — is the most structural and the least visible. AI capability depends on 3 core resources: data, computational power, and influence over regulation. All 3 are now concentrated in the hands of a very small number of corporations and states.Developing large language models requires infrastructure investment on the scale of trillions of yen (tens of billions of dollars). As of 2025, the companies at the center of global AI research are clustered almost entirely within the United States and its allied sphere. Some estimates suggest that the total number of AI researchers across the entire African continent is smaller than that of a single Silicon Valley firm. This is a structure that can only be described as the re-colonization of knowledge.When the Pope says that "control of AI must not remain in the hands of the few," his words reach beyond technology. They constitute a political and ethical indictment: a warning about what the concentration of power does to democracy and human dignity.Of course, AI also holds genuine potential to narrow inequality. Translation, agricultural support, legal advice, medical diagnosis — services that once required money and access to specialists are becoming available to anyone with a smartphone. An AI-assisted agricultural support project in rural India reported an average income increase of around 20 percent among participating farmers. The issue is not AI itself, but the design question of who its beneficiaries are. That is where the responsibility of policy and ethics lies.Can democracy survive AI?The 4th crisis the Pope names is the destabilization of democracy. AI possesses the ability to spread disinformation at a speed and scale without precedent. Sophisticated deepfake videos, political advertisements customized to individual emotional profiles, social media algorithms optimized to provoke emotional reactions — all of these have already been implicated in influencing electoral outcomes, and they are steadily dissolving the "shared factual reality" on which democracy depends.There is a less visible dimension to this story. When we ask who benefits most from AI-driven disinformation, the answer is not necessarily any particular political faction. Platform companies that earn advertising revenue by attracting attention have, by designing algorithms that favor emotionally charged content, indirectly fueled the spread of disinformation. This is less an act of deliberate malice than a "design failure" produced by economic incentive structures — a systemic problem that is precisely why it is so difficult to address.At the same time, AI can serve as a shield for democracy. The AI fact-checking system led by Taiwan's former Digital Minister Audrey Tang functioned as a tool capable of detecting and rebutting COVID-era disinformation within hours, drawing international attention. The EU is also running pilot tests of AI tools designed to detect disinformation in real time during election periods. The paradox that both the threat and the shield emerge from the same technology makes the question of how democratic societies engage with AI all the more urgent.When we delegate the decision to kill to machinesThe heaviest and most concrete of the Pope's warnings concerns autonomous lethal weapons systems (LAWS) — what are commonly called "killer robots." AI is developing the capacity to automate the entire process from target identification to the decision to strike, before any human finger pulls a trigger. This does not merely lower the threshold for war. It blurs the location of moral responsibility. When a machine kills, who bears the liability? The human who issued the order? The engineer who designed the system? The government that authorized it? Nowhere in the world does an answer yet exist.Behind the scenes of military AI development, prominent researchers have faced genuine ethical crises. In 2018, Google faced large-scale internal resistance from employees over "Project Maven," a contract with the U.S. Department of Defense to develop AI for drone targeting analysis. Google ultimately chose not to renew the contract. The relationship between AI companies and the military has continued in various forms since then, but the question of whether engineers bear ethical responsibility for what their designs enable remains unresolved across the industry. The Pope's words represent an attempt to offer answers to that unresolved question on the ground — grounded in theology and humanitarianism.Meanwhile, AI is also being used in ways that protect human life. Landmine detection AI, early-warning systems for ceasefire violations, and medical triage support in conflict zones — humanitarian applications of AI are steadily expanding. Here too, what is at stake is not the nature of the technology itself, but the choice of who uses it, for what purpose, and within what institutional framework.What the Pope's questions illuminateThe greatest significance of Pope Leo XIV's encyclical is not that it condemned AI. It lies, rather, in this formulation: "every design choice reflects a vision of humanity." Technology is not value-neutral. It is shaped by someone's intentions and choices, and those choices carry ethical and spiritual responsibility. By spending 43,000 words saying this, the encyclical reminded a world that had begun to forget it.Will AI distort humanity, or help it shine? The encyclical answers: it could be either. What matters is not AI itself, but the values and institutions of the human beings who create, direct, and govern it.When AI assists medical diagnosis, whether its benefits reach only the affluent or reach everyone is a human choice — a matter of pricing and healthcare policy. When AI unlocks creativity, whether it displaces the work of artists or liberates new forms of expression depends on how rights and contracts are designed. When AI democratizes education, whether it nurtures genuine learning or becomes a vending machine for answers is a question of how educators and learners engage with it.The Pope's words are not addressed only to engineers and policymakers. They are addressed to all of us who use AI every day. To refuse to surrender agency while being swept along by convenience. To preserve the warmth of human relationships while chasing efficiency. To keep asking questions even while receiving answers. These are not technical problems. They are questions about how we choose to live.What AI will ultimately reflect about humanity has not yet been decided. The answer will change depending on the choices we make now. That, I believe, is the heart of the Pope's warning — and, at the same time, the hope that remains with us.ReferencesNPR — "Pope Leo XIV issues first encyclical on AI and human dignity" (May 25, 2026)Axios — "Pope's encyclical warns of five AI dangers to humanity" (May 25, 2026)CNN — "Pope Leo XIV: AI must not be controlled by the few" (May 25, 2026)Vatican News — "Magnifica Humanitas: Full text of the encyclical" (May 2026)Google Health — "AI-assisted diagnostics in low-income regions: outcomes report" (2023)Digital Minister Audrey Tang — "Combating COVID-19 misinformation in Taiwan" (2020)European Commission — "AI Act and democratic resilience: interim assessment" (2025)Google / Project Maven — Public statements and employee open letters (2018)ICRC — "Autonomous weapons systems and international humanitarian law" (2024)UNDP — "AI for agriculture: pilot results in rural India" (2024)