Pope Leo released his encyclical, “Magnifica Humanitas,” on “safeguarding the human person in the time of artificial intelligence” today, May 25.
He presented it at the Vatican along with a panel of experts, including Christopher Olah, a co-founder of one of the world’s leading AI developers, Anthropic.
In a speech at the presentation, Leo stressed that the encyclical had been born out of a process of listening—to scientists and engineers developing A.I.; political leaders, parents, and teachers; and people concerned about autonomous weapons systems and various forms of exclusion happening on the basis of mass data gathering.
The document is about 42,000 words, slightly longer than “Laudato Si’” but shorter than “Amoris Laetitia.” In it, the pope compares the current course of A.I. development with the Biblical image of the Tower of Babel, saying in the introduction: “I ask everyone to abandon the construction of yet another Tower of Babel and to join forces in building up the common good.”
Leo recaps the evolution of Catholic Social Teaching through its major documents and by explaining its key principles, in many places anticipating and rebutting criticisms that the church should either stay out of politics or completely rule it. In a later paragraph, he discusses the evolution of church teaching on slavery, apologizing for the legitimization of slavery by popes in the 15th century, and the church’s failure to universally denounce slavery until the 19th century under Pope Leo XIII. “The memory of past complicity and blindness,” he says, becomes a call to vigilance against contemporary forms of trafficking.
He then moves on to the explicit examination of A.I., its development and its effects on humanity in the spheres of education, labor, technology addiction, democracy and many other areas, giving concrete recommendations on how A.I. development could be better supervised by various configurations of local and international civil, religious and educational authorities. He denounces repeatedly the ways in which A.I. will exacerbate global inequalities, and how it is already concentrating power and decision-making in the hands of a minority of powerful individuals.
Leo stresses that, for all of us, the ethics of A.I. cannot be simplified into “good use of A.I. is good, evil use of A.I. is evil”—A.I. is not, he says, a morally neutral tool, but “embodies choices and priorities through what it measures, ignores and optimizes, and how it classifies people and situations.” In other words, the moral discernment of A.I. cannot merely look at the uses to which it is put, but also how it is designed and what vision of the human person and society” is embedded into it. Leo raises tough questions about who should be held morally responsible for decisions seemingly made by A.I. and who is responsible for correcting those mistakes. He also takes aim at “transhumanism” and “posthumanism,” the idea that humanity should try to transcend itself via technological means.
As expected, Leo then turns to the use of A.I. in war. In a particularly interesting paragraph, he dismisses just war theory as “outdated,” saying it has “all too often been used to justify any kind of war” and that “[h]umanity possesses far more effective and capable tools for promoting human life and resolving conflicts, such as dialogue, diplomacy and forgiveness.” He warns against simplistic “us vs. them” thinking, which proves deadly when it is used for humans or machines to make quick decisions during war times. Later in the document, he stresses that dialogue happens best in person.
The document ends with a long meditation on peace, in which Leo outlines “five paths toward daily and public responsibility.” These include: Disarming our words; building peace through a love for justice; listening to victims of war and exclusion; cultivating a healthy realism; and reviving dialogue and multilateralism.
Politics
Faith
Culture
Vatican
Scripture
Podcasts