Demonic or not it's definitely on everyone's radar...

If you’re referring to Pope Leo XIV’s new encyclical, Magnifica Humanitas (“On Safeguarding the Human Person in the Time of Artificial Intelligence”), my reaction is that it is probably the most ambitious moral critique of AI yet produced by a major religious institution. Whether you’re Catholic, atheist, or a recovering PowerPoint user, it’s worth reading because it tackles questions that the technical world often treats as afterthoughts.

What struck me most is that Leo is not anti-technology. He explicitly accepts that AI can contribute to science, medicine, education, and productivity. His concern is that societies tend to ask, “Can we build it?” long before asking, “Who benefits?” and “What kind of people will this turn us into?”

• Human dignity is the central issue, not machine capability.
  
  
  Much of the AI debate revolves around intelligence, consciousness, AGI, and whether a chatbot can someday write a novel or pass a bar exam. Leo shifts the focus to the human person. He argues that even if AI becomes extraordinarily capable, human beings remain morally unique because they possess responsibility, conscience, relationships, and the capacity for genuine moral judgment.
  
  AI can imitate aspects of intelligence; it does not bear responsibility for its actions in the way a human does. That distinction is easy to dismiss as philosophical fluff until you realize it underlies questions like: Who is accountable when an algorithm denies someone a job, parole, insurance coverage, or military protection?
  
  
• Concentrated AI power worries him more than AI itself.
  
  
  The encyclical repeatedly warns about the accumulation of power in a small number of corporations and institutions. Leo argues that the problem is not merely technical error but political and economic concentration. If a handful of organizations control the models, the data, the infrastructure, and increasingly the flow of information, society risks creating a new form of dependency.
  
  Frankly, this may be the strongest part of the document. Many AI discussions focus on killer robots or sentient machines while overlooking the much more mundane possibility that a few companies quietly become the operating system for modern life.
  
  
• He is deeply concerned about work and human agency.
  
  
  The encyclical consciously echoes Rerum Novarum, Pope Leo XIII’s response to the Industrial Revolution. The analogy is obvious: industrial machinery transformed physical labor; AI may transform cognitive labor
  
  Leo argues that efficiency alone is not a sufficient social goal if millions of people lose meaningful participation in economic life. Retraining, worker protections, and social responsibility therefore become moral issues, not merely economic ones.

Where I think the encyclical is especially interesting is that it pushes back against both techno-utopianism and simplistic doomism.

• AI is evil.
• Progress should stop.
• Machines will inevitably destroy humanity.

Instead, it says that technologies are shaped by values, incentives, and institutions. In other words, the future of AI is a political and moral question as much as an engineering one.

My main criticism is that the encyclical is much stronger on diagnosis than on implementation. Calling for human dignity, democratic oversight, and ethical governance is easier than designing institutions that can actually enforce those goals. Humanity has spent centuries being extremely creative at agreeing with moral principles and then immediately finding loopholes. That talent remains undefeated.

Still, as a contribution to the AI debate, I think it’s significant. Most AI policy discussions are dominated by engineers, executives, investors, and governments. Leo is trying to reintroduce older questions: What is a human being? What makes work meaningful? What should never be automated? Those are not technical questions, but they may end up being the most important ones.

That’s a line a surprising number of AI researchers would agree with. The fear that keeps many people awake is not that a machine will suddenly wake up and declare itself emperor. It’s that humans will use increasingly powerful tools in exactly the ways humans have always used powerful tools: carelessly, greedily, shortsightedly, or under the banner of some noble-sounding justification.

A hammer doesn’t decide to commit fraud. A spreadsheet doesn’t decide to create a financial crisis. A social media algorithm didn’t wake up one morning and conclude that outrage should dominate public discourse. Humans built systems that rewarded those outcomes, then acted surprised when the incentives worked. AI may amplify this pattern. The technology itself has no ambition, ego, tribalism, lust for power, fear of losing status, or desire to win elections. Humans, meanwhile, have spent several thousand years demonstrating an almost artistic commitment to those activities.

In that sense, the Pope’s concern is less “the machine becomes human” and more “humans become machine-like.” When decisions about hiring, policing, warfare, education, healthcare, or public discourse are delegated to systems nobody fully understands, people can start treating moral choices as technical ones. The algorithm says so. The dashboard says so. The model predicts so. Responsibility quietly evaporates.

The deepest danger may not be artificial intelligence surpassing human intelligence. It may be human beings voluntarily surrendering judgment because thinking is hard, accountability is uncomfortable, and automation is convenient. Civilization has always been vulnerable to the temptation of replacing wisdom with procedure.

The irony is that AI often exposes human weaknesses more than machine strengths. Every new capability becomes a mirror. We look into it hoping to see the future and instead find ourselves staring back. Not always our best selves, either. More often the version that clicks “I have read and agree to the terms and conditions” without reading them and then acts shocked when consequences arrive. Machines are getting smarter. The open question is whether humans will use the extra leverage to become wiser, or merely more efficient at being foolish.

History, that relentlessly sarcastic teacher, has not yet issued a final grade.