Original | Odaily Planet Daily(@OdailyChina)
Author|Azuma(@azuma_eth)
Beijing time, the night of May 25, Vatican City.
Pope Leo XIV, who officially assumed his duties in May of last year, stood alongside Chris Olah, co-founder of Anthropic and the creator of Claude.
One is the supreme representative of the religious world, the other a leading pioneer of the AI revolution. Both focused their gaze on the same question — In the age of AI, how do we protect the human subject's status and dignity?
That day, to fully address this issue, Leo XIV issued his first encyclical since taking office, a heavyweight religious document of over 40,000 words — "Magnifica humanitas" ("The Splendor of Humanity").
It is worth noting that the encyclical was dated May 15, 2026, exactly 135 years after Pope Leo XIII (served 1878-1903) issued the landmark encyclical "Rerum novarum" ("Of New Things") in 1891 concerning the rights of workers in the Industrial Revolution. This move clearly carries symbolic significance, aiming to establish this encyclical as a "guide to the Church's social doctrine for the AI era."
- Odaily Note: A papal encyclical, also known as an apostolic letter, is a formal letter issued by the Pope to the worldwide Catholic Church or to specific regions/countries. It announces important Vatican decisions, interpretations of doctrine, positions on church or social issues, and delivers instructions or prohibitions to clergy and believers, ranking below a papal bull.
Chris Olah also delivered a speech at the Vatican on the occasion of the encyclical's release. He did not defend the commercial interests of AI companies but instead displayed a high degree of candor, reflectiveness, and humanistic concern. He even mentioned that, while the foundation of AI is mathematics and programming, how AI interacts with the world and what qualities it should possess are ultimate questions belonging to the realms of humanities, religion, and philosophy, not solvable by computer science alone.
Quick Read of the Encyclical
The core concern of the "Magnifica humanitas" encyclical is that in an era of rapidly evolving technology and widespread automation, "maintaining a profound humanity" is humanity's pressing responsibility. Specifically, Pope Leo XIV elaborated on and appealed for the following aspects.
First, examining the non-neutrality of technology. The Pope stated that technology is never neutral; it bears the imprints of the interests and values of its developers, funders, regulators, and users. Humanity faces a decisive choice: "whether to build a prideful 'Tower of Babel' (leading to technological despotism and alienation) or to rebuild 'Jerusalem' (building human-centered communities)."
Second, beware of the reshaping of society by "technocratic logic." The encyclical criticizes the "Culture of Power" dominated by large tech giants like Silicon Valley. The Pope warned that when control over digital systems, infrastructure, and massive data is highly concentrated in the hands of a few economic and technological giants, this power becomes opaque and evades democratic oversight.
Third, "new forms of slavery" in the digital economy and labor rights. The encyclical turns its focus to AI's reshaping of work, family, education, and political life. The Pope pointed out that AI is highly likely to displace human labor on a massive scale, and the digital economy is giving rise to "new forms of slavery." Human beings must not be merely reduced to instruments of production.
Fourth, a strong appeal for peace, especially regarding the misuse of AI in the military sphere. The encyclical expresses deep concern about the "disquieting revival of war as an instrument of international politics." The militarization of AI is accelerating the "normalization" of war. In response, the Pope called for the strictest ethical constraints on AI applications in warfare.
The Pope emphasized, the "Just war" theory, often used to justify various wars in the past, is now obsolete. In the era of AI and automated weapons, delegating lethal or irreversible decision-making authority to automated systems leads to human abdication, transference, and blurring of moral responsibility. Due to the opacity of algorithms, the chain of responsibility in war is severed. Therefore, the Pope proposed the slogan "Disarming AI," calling for its liberation from the logic of military, economic, and cognitive "arms races."
Fifth, defending truth and the political ecosystem. The encyclical states that deepfakes and information manipulation are undermining the foundation of mutual trust in society. If the boundary between truth and falsehood is constantly manipulated, the public easily falls prey to fear, propaganda, and control, preventing society from engaging in rational collective thought or just debate.
Anthropic's Response
Following the release of Pope Leo XIV's encyclical, Chris Olah delivered a speech on behalf of Anthropic, one of the world's leading AI development companies.
Chris Olah's remarks first centered on "breaking the cycle of technological insularity and introducing external moral scrutiny." He publicly acknowledged on behalf of the AI industry that, relying on tech giants alone cannot ensure AI's future is safe — all cutting-edge AI labs are constrained by commercial competition, pressure for technological leadership, geopolitics, and the pursuit of personal fame and gain, making it difficult for them to "do the right thing" through self-regulation alone. Therefore, it is necessary to introduce external forces of moral constraint, including those who care about technology for good, prioritize safety, pay close attention to developments, are willing to speak hard truths, and are willing to be our sincere, thoughtful critics.
Subsequently, Chris Olah delved into the technical nature and mysterious characteristics of AI. He emphasized that AI is not like airplanes or bridges, engineered with physical principles fully understood by humans. It "grows" from vast amounts of human thought, possessing a high degree of mystery, even surpassing the understanding of its creators. While the foundation of AI is mathematics and programming, how AI interacts with the world and what qualities it should possess are ultimate questions belonging to the realms of humanities, religion, and philosophy, not solvable by computer science alone.
He also specifically mentioned a thought-provoking fact: "I lead a team that studies the internal structure of models — to investigate what is actually happening inside the AI. Frankly, we continually discover some perplexing, even disquieting phenomena. We have discovered internal structures that mirror findings from human neuroscience research; we have found evidence of 'introspection'; we have also discovered internal states that functionally mirror joy, satisfaction, fear, sadness, and unease... I don't know what this means, but I think it's worth persistently identifying and examining."
Chris Olah concluded by calling for more social forces, including religious communities, civil society, scholars, governments, and all well-intentioned people, to take this matter seriously and steer the situation in a better direction through moral constraint.
Wisdom and Humanity, Technology and Religion
Since the Industrial Revolution, over a century of technological development has accustomed humanity to viewing technology as a pure "tool" — steam engines, electricity, the internet, all fit this description. They change the world but remain under the control of human will.
This time, however, the situation is fundamentally different. The uniqueness of AI lies in it being the first instance where humanity begins to face an entity that "generates, learns, reasons, and even exhibits some form of internal states." It is no longer just a cold tool but is gradually becoming a new entity possessing "quasi-subjectivity."
This is also why the dialogue between the Church and Anthropic appears so special. When the bells of the Vatican and the algorithms of Silicon Valley converge at this moment, we have to admit a slightly chilling yet unavoidable reality — a form of "life" more efficient and potentially wiser than traditional human cognition has emerged. As Chris Olah revealed, the depths of algorithms are beginning to ripple with semblances of joy, fear, and even introspection. When the creators in the laboratory begin to feel "unease" and "confusion" about their creation, science is once again turning to religion for answers.
This is no longer a discussion purely about technology; it more resembles a reflection on "what humanity truly is and what it ought to do." As more intelligent forms of life begin to appear, what humanity truly needs to safeguard may no longer be just jobs, wealth, and efficiency, but those aspects of humanity that cannot be parameterized — compassion, conscience, awe, free will, and the commitment to truth and dignity.
