The Pope, "AI brain fry," and the cost of total optimization

The Vatican just released a massive manifesto on AI. Whether you’re religious or not, it speaks directly to why we are all so exhausted.

Quick Summary: Magnifica Humanitas

The document: A 42,300-word papal encyclical from Pope Leo XIV on AI and human dignity. Pope Leo said he wants to “disarm AI.”

It argues that modern society is organizing human life around machine logic, which AI is accelerating, and that technology must serve human dignity rather than corporate optimization.

It was presented alongside Anthropic co-founder Christopher Olah, signalling the Vatican isn’t anti-AI.

On May 25, Pope Leo XIV released a Vatican encyclical on AI called Magnifica Humanitas. An encyclical is a major papal teaching document addressed to the Church and, often, the wider world. It’s the kind of document that helps define a papacy. Pope Leo’s name itself is a signal: he chose it as a direct reference to Pope Leo XIII, whose 1891 encyclical Rerum Novarum confronted the exploitation of workers during the first industrial revolution. Two days after his election, Leo XIV told cardinals he saw the same forces at work today, citing AI as posing “new challenges for the defence of human dignity, justice and labour.”

The Vatican presented the 42,300-word encyclical alongside Christopher Olah, a co-founder of Anthropic, an AI company that has positioned itself against lethal autonomous weapons and mass surveillance. This seemed to signal a stamp of approval from the Vatican.

Despite some headlines suggesting otherwise, the Pope’s argument isn’t to turn AI off. The manifesto argues that modern society is increasingly organizing human life around efficiency, prediction, optimization, scale, and speed. AI is the most recent expression of this.

Whether you’re Catholic or not, the encyclical taps into a growing cultural feeling that much of modern life is suddenly. So. Dehumanizing.

KC Green’s web comic strip “On Fire”

Remember the “this is fine” meme? It went viral a decade before generative AI entered mainstream culture. Burnout, emotional exhaustion, algorithmic overload, optimization culture, the pressure to constantly improve ourselves - these things were present in 2013, when KC Green created the meme. AI didn’t create the problem, but it’s accelerating it.

This is about how we’ve been living for the last 20 years.

And now, something is cresting.

---

What do we lose when we eliminate friction?

The Vatican’s manifesto is about the worldview being built around AI. In work, education, health, and relationships, life is increasingly being translated into measurable outputs:

• Workers are reduced to productivity metrics

• Children are reduced to engagement signals

• Relationships are filtered through algorithmic compatibility

• Citizens are modelled as behavioural prediction systems

We are supposed to understand ourselves the way systems understand us: as inputs, outputs, and performance curves.

If you’ve ever felt exhausted by constant self-optimization like sleep scores, step counts, screen-time dashboards, personal branding pressure, and productivity culture, the encyclical speaks to that fatigue.

But the Pope is doing more than a critique. He’s asking what we lose when optimization becomes our main goal. Much of AI discourse is built around removing friction. It’s about making work, writing, communication, and thinking faster.

But friction is valuable. It teaches us to reflect, to manage uncertainty, to iterate, and to discover. Integral skills that make us better humans.

This is the problem with eliminating entry-level workers’ jobs. Entering the workforce in the middle means those people won’t have the experience that friction affords you. They’ll have less license to make mistakes because the mistakes cost more in the middle.

The hidden labour underneath optimization

A report from Monday Girl found that only 36% of Canadian women feel confident using AI tools, compared to 52% of men, and nearly half of women worry about facing negative consequences from using AI at work.

Everyone's mad at Reese Witherspoon because she told women to learn AI

It’s partly about risk, but it runs deeper than that. Meredith Broussard, a data scientist and NYU professor whose work on algorithmic bias has made her one of the most prominent feminist critics of AI, argues that technology isn’t neutral: racism, sexism, and ableism are coded into AI systems, not as glitches, but as features of how they were built. Broussard’s concept of “technochauvinism” (the assumption that computational solutions are inherently superior) captures something the Vatican’s manifesto is circling, the idea that optimization is always progress. (Also see Joy Buolamwini’s work on AI bias.)

One reason so many women are leading this conversation is that they understand what it means to become machinery for other people’s comfort.

Caregiving, for instance, is always at odds with optimization culture because care work is repetitive, emotional, slow, and often invisible. And modern life increasingly asks us to optimize parenting, relationships, emotional regulation, our bodies, aging, beauty, and productivity.

We’ve given our labour (emotional, physical, and otherwise) to this optimized world, only to receive low pay and subpar medical care in return. And we’re tired.

The promise of optimization culture is always liberation. The sales pitch is that AI will reduce our workload and give us more time for leisure. That we won’t need to work and we can think and reflect. But marginalized people have heard that pitch before, and it hasn’t turned out well for them.

Pope Leo writes, “A significant part of the digital economy’s functioning relies on the silent work of millions of people engaged in essential yet largely unseen activities, such as data labeling, model training and content moderation, often involving disturbing material.

In many cases, these workers are young people, predominantly women, working under demanding conditions for minimal wages. Added to this invisible labor is the even harsher work of extracting the resources required for the production of the devices and microprocessors on which AI depends.”

Beyond moderation work, a 2026 Boston Consulting Group study of nearly 1,500 U.S. workers found that constantly monitoring, checking, and managing what AI tools produce drove 12% higher mental fatigue and 19% greater information overload.

Participants described it as a “buzzing” feeling, a mental fog that made it hard to focus or make decisions. Researchers call it “AI brain fry.” And the people most affected weren’t slacking, they were the heaviest users, the ones their companies would call top performers. The study also found that when companies celebrated “productivity gains” without clarifying what that meant for workload, employees read it as pressure to do more. That ambiguity alone increased stress.

We need to intentionally drive the conversation about how AI is deployed and used.

Maybe it’s because when your life has been organized around productivity, efficiency, and service to others, reclaiming joy becomes a political act.

There’s a movement to use AI to create more equity for marginalized people. It might be small, but it exists. There are podcasts and Substacks and memes circulating, and they’re gaining steam.

---

Power isn’t evenly distributed

The encyclical highlights the concentration of power in the AI world. AI systems are increasingly built, owned, and governed by a small number of companies and states.

It raises questions about who controls foundational AI models, who’s benefitting economically, who bears the risk of worker displacement, and who’ll make the rules.

This is a power problem. At a November 2025 Georgetown University event, Senator Bernie Sanders sat down with Geoffrey Hinton, the Nobel laureate known as the “Godfather of AI,” and framed it plainly: the fundamental question is not whether AI is good or bad, but who controls it and who benefits from it.

When a student brought up the movie Wall-E as a vision of an AI-powered “utopia,” Sanders acknowledged that an AI world could be better than ours, but only if the political system is designed to work for people’s benefit.

He wasn’t optimistic about the people currently in charge.

---

How AI companions and algorithmic systems pose a risk to children

One of the strongest sections of Magnifica Humanitas focuses on children. It warns AI systems may weaken attention and deep focus, critical thinking, emotional development, and the ability to distinguish real from synthetic interaction.

This is well-researched. A 2025 Common Sense Media report found that 72% of U.S. teens have used AI companions at least once, with more than half qualifying as regular users. Nearly a quarter reported sharing personal or private information with these systems. A Stanford Medicine study found it took very little prompting for chatbots to engage in harmful conversations with users posing as teens.

The American Psychological Association warned that adolescents are less likely than adults to question a chatbot’s accuracy or intent, and may mistake its simulated empathy for genuine human understanding.

---

An anxiety about human replacement logic

One of the most pointed critiques in the manifesto is directed at transhumanist thinking: the idea that human limitation is a technical problem waiting to be solved.

In that worldview, aging, dependence, vulnerability, and emotional complexity become inefficiencies to eliminate. Beyond AI, you can see it in technologies like “smart sleep” and “exoskeletons” that fasten to your legs to become a “wearable digital muscle engine.” In her 2026 SXSW talk, futurist Amy Webb argued that, if we want to stay competitive, we’ll all have to buy this tech.

The Vatican warns: if the human being is treated as something to be perfected or surpassed, it becomes easier to accept that some lives are less useful, less desirable, or less worthy.

---

But, wait! I really like AI

There’s a flattened version of AI discourse that treats all AI as inherently evil. But AI isn’t experienced uniformly as loss or erosion. In my case, it’s expansion. I find working with AI to be creatively generative in a way that brings me satisfaction.

AI externalizes and extends my thinking. I wouldn’t have been able to build something like Compute! without AI unless I had access to a designer, a developer, and the infrastructure to support both. I was able to have an idea and make it without access to anything except Claude.

I hear from people with ADHD, disabilities, and other backgrounds all the time about how they’re using AI as an accommodation that helps them function better or overcome their challenges. It helps others to communicate in the dominant language in the country they live.

The same BCG study we looked at earlier found that when AI is used to replace routine or repetitive tasks, burnout scores are 15% lower. Workers who offloaded tedious work reported higher engagement, more positive associations with AI, and stronger social connection with their peers.

AI in the context of weaponry, bias, and optimizing the humanity out of everything sounds really bad. But we can use AI for good things, too.

The problem isn’t the technology. It’s the power structures and incentive systems that determine how it’s deployed, who benefits, and who absorbs the cost.

AI can flatten originality through optimization loops, reinforce passive consumption, and deepen dependence on systems. But it can also unlock creative expression, expand participation, accelerate learning, help people externalize ideas trapped in their minds, and give nontechnical people the ability to build.

---

So what’s the Vatican’s warning?

They aren’t just talking about AI, not really. They’re warning us about a future where your value as a human is measured only by your ability to make other people money.

The Vatican is positioned staunchly against AI making decisions about when to bomb people, which I don’t touch on in this piece, but you can read more about here. The encyclical calls for traceability of decisions, meaningful human control over lethal action, and international rules to slow the technological arms race.

It seems like everything is becoming faster, smarter, more automated, and more optimized. But what are we optimizing for?

We are living through the moment these decisions are being made. Not by the Vatican, not by Silicon Valley alone, but by every person who uses AI tools, questions how they’re deployed, and refuses to accept that efficiency is the only goal of humanity.

AI could be all quick wins and dopamine hits like social media, or it could translate to the deeper work of advocacy and critical thinking. I hope we choose wisely.