Agency Without Wisdom
A Call for AI Caution
“Humanity, created by God in all its grandeur,” writes Pope Leo in his 40,000-word encyclical on artificial intelligence, “is today facing a pivotal choice: either to construct a new Tower of Babel or to build the city in which God and humanity dwell together.”
It’s a stark contrast, the Tower of Babel or the City of God. As Pope Leo writes, Babel was “supported by a uniformity that eliminated diversity and that chose homogenization over communion” while the City of God, by contrast, “rebuilds relationships before rebuilding with stones” and “rediscovers a common language—not one of uniformity, but one of communion.”
In short, the City of God celebrates multiplicity founded in humility. It’s the Body of Christ, where “the body is not one member but many.” In this spirit, Pope Leo doesn’t limit his message to Catholics, but expands it to include all people, believers and nonbelievers alike. He writes, “The Church regards all who sincerely seek ‘truth, goodness and beauty’ as companions on the journey, and considers them as ‘precious allies’ in defending the dignity of every person.”
I personally feel called as a companion on the journey to focus on my cultural setting—namely, the Wasatch Front in Utah. It was here that I recently attended a presentation by Jon Cheney, founder of the General AI Proficiency Institute, which helps employees use AI to handle “drudgery and repetitive work” so they can “adapt, create, and take responsibility for meaningful work.” Cheney created this business in a weekend and booked his first client five days later, leading him to be featured in the news and mentioned on The Joe Rogan Experience, the world’s most-listened-to podcast.
For Cheney, the core concern around AI has to do with maintaining our individual agency, which, alluding to a Latter-day Saint scripture, he defines as “the choice to act rather than be acted upon.” This is the central sentiment of his company’s manifesto (or “doctrine,” as Cheney calls it). It’s also a sentiment I encounter frequently in different tech circles on the Wasatch Front—a sentiment I find simultaneously inspiring and troubling.
Namely, I’m inspired by the call to avoid mindlessly giving my eyeballs (and by extension my life) over to companies that don’t have my best interest in mind. I’m also inspired by Cheney’s call for a new way of living. He writes that he longs for a life centered on “deep human connection, creative problem solving, ethical and philosophical guidance, and hands-on craftsmanship.” This call appeals to me deeply, and I long for this same thing—a world where people do work that matters with people they love.
Yet I question Cheney’s underlying assumptions about AI and agency, and, given how widespread these assumptions are today, I feel a sense of urgency to spell out a few counterarguments—not to dismiss Cheney personally (again, I share his call for deep human connection) but to add a measure of caution in the spirit of Pope Leo.
Cheney writes that he is “unapologetically pro-AI” and urges people to use the technology or get left behind.1 “History shows,” he writes, “that when individuals or organizations fail to adapt to technological change, the outcome is usually a gradual decline—or even worse—catastrophic collapse. Entire companies, careers, and communities disappear, not because they were immoral or lazy, but because they failed to learn, adapt, and act in time.”
What goes unsaid here, however, is why slow-adapting communities collapse. It’s not because humans need advanced technology to survive, since we existed (albeit uncomfortably) for hundreds of millennia with rudimentary technology.2 Rather, it’s often because powerful people with advanced technology use their agency in short-sighted and destructive ways.
In addition, the notion that agency is strictly defined as “the choice to act rather than be acted upon” fails to recognize the way in which agency also includes a passive element—a willingness to surrender to that which is beyond our control, including certain aspects of the physical world such as finite natural resources.3 This passive element is also part of agency. Again, a willingness to let go.
Cheney doesn’t acknowledge any of this. Instead, he simply says that “the future does not belong to those who wait. It belongs to those who choose to act.”
But we might ask: What if the primary problem we’re facing is that “those who choose to act” aren’t as introspective, generous, or wise as they should be? What if the problem isn’t so much a lack of bold action, but a lack of wisdom, which weaves together the active and passive elements of agency?
Such questions make me reflect on how early Latter-day Saint settlers damaged Utah Lake, a lake I’ve lived near since childhood. When Latter-day Saints immigrated to this area and built Fort Utah in what became Provo, they fought the native tribes,4 seized control of the lake, and immediately got to work fishing it. Within a decade, they’d overfished it.
Then, committing to “act rather than be acted upon,” they (in conjunction with the US government) re-filled the lake with carp, an invasive species that sludged up the waters and decimated the ecosystem.
If that weren’t enough, as recently as 1967, local inhabitants routed sewage into the lake, further polluting an already compromised body of water. This combination of carp and crap—coupled with algae blooms—gave the lake a reputation of being unsanitary and unusable, a reputation that still haunts the place today. Efforts to restore the lake are costing millions.
Put simply, the drive to “act rather than be acted upon” (again, defined strictly as the active form of agency) caused tremendous harm to my local environment.
It’s especially tragic because the damage wasn’t inevitable. Latter-day Satins who trekked here from Nauvoo could have instead fully integrated into the area by working with the landscape and native people rather than colonizing them through subjugation.
Admittedly, this level of integration takes tremendous patience, a painful measure of patience. It’s not limp passivity. It isn’t non-action. It’s wise action, “the art of sailing rather than the art of rowing” as one thinker puts it. It’s a call to work, aligned with nature.
In this way, wise action rejects subconscious subservience to cultural programming and instead sees again and again with beginner’s eyes. Consider, to give one more local example, how Utahns (including myself) adopted standard cultural practices, planting water-intensive lawns and crops in a desert state. If we had instead ignored this cultural programming, which was handed to us by European aristocrats, and listened to our desert landscape, we would have developed sustainable practices decades ago and therefore wouldn’t be in a state-wide water emergency today. As things stand, however, it looks like we’ll have to start the arduous process of collectively replacing our crops and lawns with less water-intensive options, resulting in far more work than if we’d just aligned with our environment from the outset.
And yet, tragically, we’re heading in the opposite direction with the call to build water-sucking,5 electricity-sapping,6 heat-generating7 data centers here. (The footnotes here capture some of the complexity surrounding these issues.) We have to build data centers, the argument goes, or other states will get these contracts! We have to build them or China will leave us behind!
In response to such concerns, we might ask whether we’ll even have the option to live here if we destroy our lakes. It’s a dilemma that’s not strictly hypothetical, as Lake Urmia, once the largest saltwater lake in the Middle East, has “almost completely dried up” over the past few decades as a result of human action and mismanaged water use. Now the Iranian government is facing the difficult decision of whether to move their nation’s capital from arid Tehran to wetter regions, at an estimated cost of 100 billion dollars. Given these facts, we might cautiously consider whether we can destroy our lakes without suffering similar catastrophic consequences.8
Of course, environmental destruction is not the only reason to be cautious of AI.
We also must consider how these models are trained, who reaps the financial benefits they bring, and how they affect the intellectual and spiritual formation of young people.
I’m haunted by the story of Mophat Okinyi, a Kenyan who was hired as part of a team of moderators paid between one to three dollars an hour to help ChatGPT categorize the most horrific digital content imaginable. Okinyi read and categorized roughly 700 highly explicit passages a day, causing him to slip into a state of paranoia. Months later he returned home from work one day to find a note from his wife saying that she was leaving him. “You’ve changed,” it read. “You’re not the man I married. I don’t understand you anymore.” It seems that the cost of training AI is the uncountable suffering of people like Okinyi.
In addition, AI models have been fed writing, art, and music without the consent of creators. This wouldn’t be such a travesty if everyone collectively shared the financial upside of AI instead of concentrating it in the hands of the few. Unfortunately, it now looks like those who have created content online—recipe bloggers, political commentators, etc.—won’t be fully compensated for their work as users bypass those sources for the AI summaries that pull from those sources. It’s no wonder that when Open AI CEO Sam Altman said, “We see a future where intelligence is a utility, like electricity or water, and people buy it from us on a meter,” one commenter responded, “They stole the internet’s knowledge and all the work that went into it and they want to sell it back to us for a fee. What a business model.”
And then there’s the matter of how AI affects intellectual and spiritual formation. At schools across the nation, kids receive Chromebooks fully integrated with Gemini, nudging kids to offload effortful yet valuable mental labor like writing, and allowing them to explore content that’s not appropriate for kids (including semi-explicit romantic roleplay with the chatbot) when they should be doing their homework. None of this is the stuff of healthy formation. As sci-fi author Ted Chiang has said, using AI to write papers is “like bringing a forklift into the weight room.” There’s no muscle gained. No development. We shouldn’t be surprised, then, that an MIT study found that “over four months, LLM users consistently underperformed at neural, linguistic, and behavioral levels” and that 83 percent of those who used LLMs couldn’t later recall the content of a single sentence they “wrote” compared to 11 percent of those who did the work themselves. We also shouldn’t be surprised that the creators of several leading AI models, driven by a hunger for financial gain, have loosened the restrictions around sexually explicit material. For example, even after public outcry, for months Grok allowed users to undress people in photos (including underage people) in its so-called “spicy mode.” The hope there, it seems, is that this titillating content will make the product more addictive, leading to market dominance.
Where does all this leave us?
For my part, I don’t think the answer is to create a hard and fast universal stance against AI or to suggest that if we use it we should do so while feeling sadder and guiltier. AI models are here to stay in some form, and they admittedly have their uses in stem-related fields, where, among many things, researchers are now using them to more effectively pinpoint early onset cancer.9
My position, then, is not a knee-jerk, fear-based reaction against AI but instead a call for wisdom. I’m saying that if we blindly barrel ahead we risk repeating humanity’s needlessly destructive tendencies. We’ll figuratively overfish the lake, strapping our descendants with an enormous burden that could have been avoided if we’d just acted more wisely from the outset.
The truth is that if we’re going to enjoy “deep human connection,” as Cheney calls for, we must admit that agency is active and passive and that both of these elements are essential for building the common good. We must see that machine learning, as impressive and useful as it sometimes is, often carries terrible costs, particularly for the least among us. We must be honest about the way that tech companies with billion-dollar budgets pour their time, money, and effort into a siren’s song that, as the poet T.S. Eliot wrote, aims to make us “distracted from distraction by distraction . . . in this twittering world.”
If we are going to build the City of God we must stop laying the blame primarily on communities that are slow to adapt and start calling out those who use their agency to expand their dominion against the will of others.
We must, above all, recognize that without wisdom, what we call “progress” is, in reality, its opposite—an abandoned tower to nowhere.10
Jon Ogden is a cofounder at UpliftKids.org, which helps families explore wisdom and timeless values together. Find more of his writing at One Step Enough.
Art by Endre Rozsda (1913–1999).
KEEP READING
WAYFARE COLUMNS
WAYFARE POETRY
WAYFARE THEOLOGY
Check out our growing collection of thoughtful religious commentary.
.jpg){kind=link}
It’s worth questioning whether those who don’t use AI will truly be left behind in all fields, particularly when it comes to using large-language models for writing. As the ecologist Ben Lockwood suggests, “The myth of ‘AI’ is that you can use it to offload lower function mental activity in order to free up higher function. But this assumes that that higher function of thought can operate without the lower to support it, which is in pretty much direct contradiction of all philosophy of mind.”
Many people around the world still survive just fine without advanced technology!
The notion of surrender is found throughout the world’s wisdom traditions, exemplified in the Buddhist notion of acceptance, the Stoic notion of letting go of what’s beyond our control, the Taoist notion of flowing with the Way, the Islamic notion of surrendering (which is what the word Islam literally means), and the Christian notion of “not my will but thine be done.”
According to one historical source, the Fort Utah conflict started when a native man was killed for stealing a shirt. At first Brigham Young urged caution, inviting the Mormons in the area to consider, “Why should men have the disposition to kill a destitute naked Indian who may steal a shirt or a horse … when they never think of meting out a like retribution to a white man who steals?” (Young also unfortunately told them, “If you would have dominion over them for their good, which is the duty of the elders, you must not treat them as your equals.”) As skirmishes continued over the ensuing months, Young conceded to work with the US government to take the land by force, resulting in the deaths of more than 100 native people, assuring Mormon dominance in the area.
While other human activities, such as beef production, use far more water than data centers and while some commenters claim, with good evidence, that mainstream media has overblown the issue, it’s still an issue worth tracking, especially because the data on the topic hasn’t been fully disclosed. As one study on the topic reports, “Company-wide metrics from the environmental disclosure of data center operators suggest that AI systems may have a carbon footprint equivalent to that of New York City in 2025, while their water footprint could be in the range of the global annual consumption of bottled water. Further disclosures from data center operators are urgently required to improve the accuracy of these estimates and to responsibly manage the growing environmental impact of AI systems.”
Pew Reseach: “U.S. data centers consumed 183 terawatt-hours (TWh) of electricity in 2024, according to IEA estimates. That works out to more than 4% of the country’s total electricity consumption last year – and is roughly equivalent to the annual electricity demand of the entire nation of Pakistan. By 2030, this figure is projected to grow by 133% to 426 TWh.”
Specifically: “Operational data centers produce measurable warming in adjacent residential neighborhoods, with the downwind warming effect as large as 2.2 °C”
To help save our lakes, consider joining Grow the Flow, which is taking focused action to persuade local legislators to care about our landscape.
It’s my sense that many debates about AI would come to a resolution if we could see that what is useful in stem-related fields (which call for the analysis and synthesis of pre-existing parts) is often destructive in the humanities (which call for seeing and questioning whole systems, requiring the nurture of fundamental critical and creative thinking skills that are cut off at the roots when we use AI as a crutch).
Major thanks to Rachael Givens Johnson and Rachel Jardine for developmental help with this essay. It’s in a far better spot because of their guidance, and I’m grateful for it.
 | A guest post by
|
Beautifully written , insightful and frightening all at once!