Pope Leo urges UN to disarm AI: 'Machines are tools, not persons'

While Silicon Valley and legal scholars keep daydreaming about whether superintelligent algorithms might one day deserve constitutional rights, the Vatican just slammed the door on the idea.

On Friday, Pope Leo XIV used his Papal Address to call on United Nations member states to take concrete steps toward banning lethal autonomous weapons during their meeting in Geneva next month. AI systems, he argued, need to be disarmed, insisting that humanity carries the ultimate responsibility to guide, reshape, and when necessary, flat-out reject dangerous deployments.

Keep humans on the trigger

With UN delegates preparing to convene in Geneva next month, Pope Leo placed battlefield automation squarely in the crosshairs.

Decisions affecting human life, personal freedom, and the use of force must always remain subject to effective, informed, and responsible human oversight, he warned. In other words, delegating the choice to kill to software is fundamentally unacceptable.

This is not the Holy See's first warning on algorithmic warfare. Back in 2024, Pope Francis pushed a similar message at international summits, declaring that no machine should ever choose to take the life of a human being.

Tools, not persons

Pope Leo went a step further by taking aim at the legal theories creeping around autonomous systems. He explicitly cautioned against granting AI systems rights or legal personhood, laying down a strict boundary: machines are tools, not persons.

The address directly mirrors his first major papal document released in May, which centered on safeguarding the human person in the age of artificial intelligence.

Pope Leo even shared why he chose his papal name in the first place. Just as Pope Leo XIII led the church through the societal upheaval of the Industrial Revolution, he believes his papacy must confront the seismic shifts unleashed by modern frontier technology.

To enforce this boundary, he called on international institutions to set up shared technical standards, independent oversight, and clear mechanisms of accountability. Any decision around developing or deploying AI must be open to scrutiny, challenged whenever necessary, and backed by proper remedies if harm is done.

Why it matters

Defense contracts are quietly turning into one of the most lucrative revenue streams for frontier AI labs and autonomous drone manufacturers.

We think Pope Leo's move to tie lethal autonomous weapons directly to the debate over legal personhood is sharp. Anthropomorphizing AI might sound like fascinating philosophy at dinner parties, but on the battlefield, it creates a dangerous liability shield. If a drone misidentifies a civilian target, blaming an algorithmic mind lets the commanding officer and the contractor walk away clean.

By insisting that machines are strictly tools, the Vatican strips away that fog. A tool cannot bear guilt, meaning every single casualty caused by software points right back to human hands.

Getting UN member states to agree on an outright ban in Geneva next month will be an uphill battle, especially as global powers race to automate their air, sea, and land forces. But morally, the line has been drawn loud and clear: algorithms cannot serve as humanity's moral scapegoat.