as preached at the First Unitarian Universalist Church of Houston, November 17, 2024
My text for this Sunday consisted of three separate segments: a pulpit editorial, a set of readings created with the assistance of ChatGPT, and a sermon.
Pulpit Editorial
Some Unitarian Universalist congregations have the tradition of pulpit editorials. These are short theological reflections by a minister on current events when it does not seem desirable or possible to shift the entire service to address the challenges of the world. The topic of this week’s sermon “Silicon Gods: Artificial Intelligence and Religion” was picked by our February auction winner. Rather than try to change the entire program for Sunday and reschedule the auction sermon, we thought it best to bring the practice to First Unitarian Universalist. Moving forward I anticipate that pulpit editorials will come to be an occasional feature.
They derive, in a sense, from the sentiment expressed by the poet Marge Piercy. “The work of the world is common as mud,” she tells us. In these times, we must continue to do the common work of the world even as we prepare for what is to come. This common work–the work of love, birth, and death; the quotidian tasks of planting gardens, baking bread, balancing checkbooks, taking the children to school, trips to the grocery store, the garage, and the workshop–is in a sense who we are in hard times and when the world is at its most beautiful.
Mary Oliver, another poet beloved by many Unitarian Universalists, offers a similar sentiment. She writes:
Tell me about despair, yours, and I will tell you mine.
Meanwhile the world goes on.
Meanwhile the sun and the clear pebbles of the rain
are moving across the landscapes,
over the prairies and the deep trees,
the mountains and the rivers.
“The work of the world is common as mud,” “meanwhile the world goes on,” in the last week the future President has clarified, if the forecast was ever murky, what his administration will be. The New York Times columnist and Russian dissident M. Gessen has argued that it will be one characterized by “morally unconstrained collective egoism.” Such a situation, they claim, will be distinguished by “permission to abandon societal inhibitions, to amplify the grievances of one’s own group and heap hate on assorted others.” It will be one where the politics of cruelty run rampant and civil institutions become the playthings of corrupt, unqualified, malignant sycophants who primarily seek private gain.
Long ago the historian Eric Hobsbawm sought to understand why many working people found living in such a society to be in some way satisfying. He studied medieval mobs, the collection of journeymen, laborers, and others who came together in what we might call riots to make demands. They clamored for bread when bread was scarce, relief from work when labor was overly harsh, and shelter when homes were hard to find. Some of these mobs directed their anger towards the powers and principalities of the day. Others focused their hostility on “foreigners,” those we might call immigrants, and lifted up traditional sources of reactionary power–the Pope, the local bishop, the monarch–as the solution to their problems.
Hobsbawm called these the “Church and King Mob.” He wanted to understand why such constellations misdirected their anger. Instead of expressing rage at the people actually responsible for their suffering–the lords of wealth–they transfered it to the most vulnerable members of society. He came to the conclusion that in these kinds of situations “the ruler … in some sense symbolizes the people and its way of life.” Additionally, for them, “the ruler … represents justice.”
Consider those dimensions of his analysis for a moment and then place them alongside Gessen’s. Taken together is a sort of psychological explanation for the basis of the politics of cruelty. Many of the future President’s supporters fear that their way of life is being destroyed. He is able to successfully cast himself as their representative. His elevation becomes their elevation. His targeting of civic institutions becomes an expression of justice. He ushers in a morally unconstrained collective egoism in which the actual sources of suffering are hidden from view and the most marginalized are blamed.
This is another description of the dynamics of totalitarianism. Totalitarian systems are organized around the personality of a leader who personifies the state’s power. They seek total subjugation of all who live within their borders. They claim that they are the carriers of social norms.
Here we turn to the subject of artificial intelligence and religion. Social media has prompted so many working people to assign the wrong groups responsibility for their difficulties. In reality, a shortage of housing is the result bad federal housing policy. In truth, rising inequality stems primarily from a failure to tax the rich at sufficient rates.
In the morass of social media different explanations are put forth. False messages are amplified by algorithms. Click on something appealing that is untrue and more lies will pop-up in your social media feed. The algorithms are designed to show “more of what you will like,” whatever that means, rather than direct you towards reliable sources. Pretty quickly it is easy to find yourself living in world of illusion. It can be an isolating one. The quick hits of dopamine that come from engaging with social media easily inspire many of us to spend more time looking at our screens and less time connecting with those around us. It becomes harder to get to know your neighbors and easier to imagine those immediately around you as some kind of other rather than as fellow human beings.
The isolating world of illusion, in such a situation the tasks of our religious community become clear. Two are worth lifting up. The first is perhaps the easiest. Collective egoism, totalitarianism, call it what you will, it is built on lies. We have a free pulpit. We are a free religious tradition. We can, we must name the lies and counter them with truth.
We must especially name one lie. We might call it the biggest lie. The Universalist part of our tradition is rooted in its opposition. The biggest lie: there are such things as races. The truth: there is only one race, the human race. The biggest lie: nations and borders are natural things. The truth: they are human creations. The biggest lie: there is this group and that group and they have nothing in common. The truth: we are all bound up together, all part of the great family of all souls.
We are all part of the great family of all souls, this brings me to the second task before us: building communities of care. A politics of care is based on connection to those around us. It requires us to reconnect with our neighbors, to strengthen social ties, to remind each other that no matter how different we are we all have something in common.
We are all children of the Earth. We are all children of the same long dead star. We are star children, connected by the very air we breathe, the water that flows through our veins, and the stones and soil that forms our bones.
To help each other remember this, we need to build community. We need to shine the clear light of connection strongly enough that it casts away the isolating world of illusion. Last week, I asked you to invite someone to come to church with you. I encouraged you to share your Unitarian Universalist story with them by answering the question: Why is your life better for being part of First Unitarian Universalist?
As we face what is to come, I am going to continue to encourage you to invite people to join us. Together our community offers an antidote to viral algorithms that spread misdirection and separation and encourage the politics of cruelty.
It is not just about inviting people to church. We can, and must, widen love’s circle by creating as many circles of care as possible. This means getting to know your neighbors. This means building stronger relationships with your coworkers. It is through these acts of community creation that we create the resources necessary to counter the politics of cruelty. Such politics are based in abstraction. They come from things you think you know about others rather than the real of their lives and yours. It is through this work that we can create new social norms that are focused on compassion and commonality rather fear and disconnection.
Hannah Arendt, the great scholar of totalitarianism, believed that the principal task of enduring times like these was maintaining “the possibility of friendship between … human beings.” And truly, that is the task for us in the approaching years, to be a place where we can come to find and strengthen friendships. This is how we widen love’s circle. This is how we survive. This is how we help others survive.
Over the next months, I will be inviting you into conversations about how we can best be a place for building such relationships and nurturing circles of care. But for now, I am going to turn things over to Chelsea for a musical meditation. Then we are going to pursue our announced topic.
Rev. Scott vs. ChatGPT
Colin: In this segment of the service, titled “Rev. Scott vs. ChatGPT,” we have asked Rev. Scott and ChatGPT to answer the same question. ChatGPT is perhaps the most famous generative artificial chatbot. It is a large language model. It has been trained on a massive amount of text with the intention of enabling it to mimic human creativity as much as possible. The engine we used was GPT-4.
The question that we asked both Rev. Scott and ChatGPT to answer was the same. It came from my sermon last week. It is: “Why is my life better for being part of First Unitarian Universalist?” Here is what Rev. Scott has to say:
Scott: Last week, Rev. Colin told of his conviction that building the Beloved Community is the way forward during these hard times. He asked for each of us to consider the question, “why is my life better for being part of this church?” — one that is committed to building a multicultural, multigenerational, and multiracial community.
Well, I was drawn to Unitarian Universalism because I believe spirituality and theology are too important in one’s life to accept, without question, someone else’s vision and interpretation of them. Not that I grew up questioning authority, quite the opposite. But once I realized that I was gay and saw how most churches talked about the subject, it made me wonder what else they were getting wrong. Reason should also inform our spirituality and theology. So, Unitarian Universalism’s acceptance of divergent spiritual frameworks, reason, and my sexual orientation has certainly made my life better.
Although church was important to me as a youth, because I sensed I was gay, it didn’t occur to me that being a minister would even be a possibility. This particular church has made my life better because this is the place where I have been able to do what I was called to do, minister to loving and thoughtful people.
Helping to build the Beloved Community by having this place to congregate, a place for those who want to discuss their questions, not accept ready-made answers; a place for those who need a safe space because we diverge from the norm; a place where we stand up for the persecuted, is exactly why my life is better for being a part of this church.
Colin: ChatGPT provides a more accurate response when it is given more context. We thought it would be helpful to expand the prompt to include some of the things that Rev. Scott would know when he was composing his response. I provided ChatGPT the following prompt: “Write a 250-word response to the question: ‘Why is my life better for being part of First Unitarian Universalist?’ The response will be read at a worship service at the First Unitarian Universalist Church of Houston titled “Silicon Gods: Artificial Intelligence and Religion.” Please provide the reader with the clearest possible reading instructions. These should include where in the text she should pause for emphasis, any hand gestures she should make, and any general directions for blocking. The ChatGPT text will be proceeded by a text prepared by the congregation’s Associate Minister that answers the same question.”
Here is the full text and instructions that ChatGPT generated.
Sheryl Abrahams: Reading Instructions for “Why is my life better for being part of First Unitarian Universalist?”
General Directions:
- Speak slowly and warmly, as if sharing a personal story with close friends.
- Make eye contact with the congregation as often as possible.
- Use natural gestures to emphasize gratitude and connection, such as placing a hand on your chest • when discussing personal feelings.
- Stand comfortably, feet hip-width apart, to convey confidence and openness.
Pauses and Gestures:
[Pause] indicates a moment to breathe and let the words resonate. [Smile] reminds the reader to use a warm and inviting expression.Text:
My life is better for being part of First Unitarian Universalist because this community reminds me what it means to truly belong. [Pause]
Here, I find a sanctuary—not just from the busyness of life, but from its loneliness. [Pause, look around the congregation] Each week, I am reminded that I am not alone in my questions, my doubts, or my hopes. [Hand on chest, soften voice]
In a world driven by algorithms and artificial connections, First UU offers something profoundly human: a space to connect with real people, with shared values, and with the deeper mysteries of life. [Pause, glance upward briefly]
Being part of this congregation has challenged me to grow—spiritually, ethically, and personally. [Smile] The sermons stretch my mind; the conversations nourish my soul; the acts of service remind me that love is a verb. [Slight forward gesture with one hand]
And perhaps most importantly, First UU gives me hope. [Pause] Hope that, in an age of Silicon Gods, we can still gather to celebrate the divine spark in each of us. [Smile, brief glance upward]
For all of this, and for all of you, I am deeply grateful. [Look around, hand on heart] [Pause briefly, then step back slightly to signal the end.]
Colin: Here ends the readings. In moment, we will have some commentary. But first, Dr. Rocke and the Chelsea are going to bring us together in song.
Sermon
So, Rev. Scott, I have got a joke for you. It is a bit dated but I think it might still be funny. Ready? Here goes.
Rev. Scott, I recently got a new VCR–see I told you it was dated. It is really great.
Scott: What makes it so great?
Well, it is fully programmable. It can set it up to watch all of my favorite shows. That way I do not have to watch them. It saves a lot of time!
It saves a lot of time, I am not sure if the joke is funny. Though, I do appreciate Rev. Scott being a good sport about it. It touches on a central point about artificial intelligence that we will be considering throughout this sermon: What work do we need to do to be human?
What work do we need to do to be human? I will return to that question momentarily. But first, a brief explanation about this Sunday’s service. As I mentioned earlier, every year an auction winner gets to pick a topic for a sermon. This year’s winner selected something that is on many of our hearts: artificial intelligence. Particularly, they wanted us to reflect on artificial intelligence and religion.
It is a subject to which we will return in the spring as part of our Future Visions, Future Selves series. But for now, I want to begin our reflections by inviting you to read responsively #567 “To Be of Use” by Marge Piercy in our grey hymnal. I will read the plain text. You can read the italics.
I want to be with people who submerge in the task,
Who go into the fields to harvest and work in a row and pass the bags along.
Who stand in line and haul in their places,
Who are not parlor generals and field deserters but move in a common rhythm when the food must come in or the fire be put out.
The work of the world is common as mud. Botched it smears the hands, crumbles as dust.
But the thing worth doing well done has shape that satisfies, clean and evident.
Greek amphoras for wine or oil, Hopi vases that held corn, are put in museums but you know they were made to be used.
The pitcher cries for water to carry and a person for work that is real.
The pitcher cries for water to carry and a person for work that is real.
The last line describes some of the stakes in artificial intelligence. What kind of human work will it replace? Will it replace the sort of work that gives life meaning? Will it merely make life easier?
In posing the questions, a confession: I am not certain that they are right questions to ask about artificial intelligence. It is subject that truthfully is far from my expertise. It is not something to be approached lightly. It is vast and complicated. It has profound implications for the human future. Some who study it, like the computer scientist Stuart Russell, call the creation of artificial intelligence “the most profound change in human history.”
The most profound change in human history, quite a statement and one that deserves deep theological reflection for certain. Whether it is justified remains to be seen. The core question within it is about something called “general purpose or super intelligent” artificial intelligence.
This is basically the crafting of God. Such an intelligence would have access to the full scope of human knowledge. It would be able to perform any human skill. It would be self-motivating and pursue its own objectives. Presumably, it would be able to harvest the sum of human accomplishment and direct it towards some larger goal.
There are both dystopian and utopian visions of what such a general purpose artificial intelligence might resemble. The Matrix series offers one. It it humans function as little more than fodder for a society of super intelligent machines.
Others tell more pleasant stories. Stuart Russell describes their dream as one in which general purpose artificial intelligence “would be everything as a service.” Under its reign, “[t]here would be no need for armies of specialists in different disciplines, organized into hierarchies of contractors and subcontractors, to carry out a project.” Whatever your desire–a home remodeling project, a better relationship, a cure for cancer, a plan to visit Venice–all you will have to do is let the general-purpose artificial intelligence know and, if you had the resources, it will you dream into being.
Such a world is still some ways off. How far, no one knows. Some think it will emerge in our lifetimes. They believe that it is perhaps in the incubation stage somewhere amid the servers of Silicon Valley. Others think that it is distant or that it might never emerge–that anything we might create will only be the facsimile of intelligence, not human brilliance itself.
In the meantime, artificial intelligence is already profoundly shaping our lives. Algorithms are pushing us further and further into our political silos. The idea of a shared media landscape–and with it a shared national narrative–that was so prevalent throughout much of the twentieth century is gone.
Artificial intelligence increasingly plays a role in warfare. The Israeli military uses it to select targets in Gaza. Images generated by artificial intelligence have appeared online in efforts to make Israeli war crimes appear even worse than they are. At the opening of the war in Ukraine, artificial intelligence was used to create a video of President Volodymyr Zelenskyy ordering his fellow Ukrainians to surrender. Lethal autonomous weapon systems, things we euphemistically call killer robots, already exist. According to the BBC, it is possible to buy them on the internet–a fact I have chosen not to verify.
Grim as this comment might be, artificial intelligence is also present in more pleasant and productive spheres. It is used in entertainment. None of the recent Marvel movies would be possible without it. It is used in scientific research. This year’s Nobel Prize in chemistry went to scientists who deployed it to predict three-dimensional protein structure, something that has long been beyond human keen. It is used in logistics and plays a central role in shipping.
Truthfully, it is just about everywhere. It helps us to livestream our services. You almost certainly carry a device–a smart phone–in your pocket that contains all kind of AI programs. This morning I used an app to call a car to bring me to our campus. The trip was made possible because a form of artificial intelligence matched me with the nearest driver.
The pitcher cries for water to carry and a person for work that is real, in this context there are so many questions we could be asking about artificial intelligence. And here I turn to the question of work. Queries about a silicon God–the creation of a general purpose artificial intelligence to whom we might lift up hymns of praise and say, “All hail computer,” is some ways off. The impact on other forms of artificial intelligence on our lives is already here.
Can you think of the ways in which your life has benefited from artificial intelligence? They are likely to be myriad. Can you think of the ways in which it has create problems or represent challenges? They, too, are possibly manifold.
The question of work, and by work I mean human activity, has always been fraught by automation. As far back as the Greek philosopher Aristotle, there have been fears about the potential of machines to replace humans. He observed, “If every instrument could accomplish its own work, obeying or anticipating the will of others … chief workmen would not want servants.”
Would not want servants, automation has long been sold as way to release us from unpleasant tasks. Laundry machines, dishwashers, labor saving devices meant to preserve us from undesirable activities and free us for more congenial pursuits. Artificial intelligence offers much the same vision.
Here we should be cautious. I am from Michigan. Automation, the arrival of various kinds of robots to the factory floor, decimated the industrial workforce throughout the latter half of the twentieth-century. In some cases, people found other forms of work. In others, they were simply cast aside. Inequality has risen. The kind of labor you can do without a college education has become less lucrative. Communities have been severely damaged and the lives of working-class men have become demonstrably worse. A major source of the appeal of the future President is that he has, disingenuously, promised to fix all of this.
But what has been done cannot be undone. Those jobs–those forms of work that is real–is gone. It might be replaced by something else. Economists generally believe that when technology eliminates one kind of work it replaces it with new forms of labor. This might be true. It might not be. No one truly knows the answer.
What I do know is this, the pitcher cries for water to carry and a person for work that is real. We make meaning, in part, by doing. I find crafting a sermon satisfying. I know that Rev. Scott does as well. Work binds us together. It connects us to community and to the terrestrial sphere which we share. It shapes the ways in which we make meaning. It forms a crucial part of our life together.
I also suspect that the kind of artificial intelligence that can replace our life together has not yet arrived. ChatGPT is a ubiquitous chatbot that is able to generate complicated texts. But it I am not sure that it can replicate real relationships.
Still, it can provide useful outlines and even essays that make entertaining reading. Do not use it to generate bibliographies. It sometimes hallucinates—makes up information that does not exist—and references texts that do not exist.
Overall, the text it generates tends to be largely generic. Often, they need human help. They only come into being through the construction of a prompt. The better the prompt, the more useful the result. More often they necessitate a bit of human intervention to realize their potential. I have been told that our ChatGPT generated choral anthem would have been a disaster if Chelsea had not intervened to score it.
And let us face it. Rev. Scott’s answer to our prompt was much better to the algorithm. This is gets to the core distinction between artificial intelligence and us, at least for the moment. Rev. Scott has a life story. He has experiences. He can craft something that comes from deep within. He can have a relationship with you. But ChatGPT, it is just a computer program. Any experiences, any stories, it might come up with will just be material from its vast databases.
The replacement of the human with the robot is still a little while off. Will it ever come; I cannot know. But a final comment before we close. To me, the most distressing dynamic in all of the conversations about artificial intelligence is that they largely take place outside of the context what truly binds us together. They are not connected to the one, genuine, universal religion–our shared residency on Earth.
Artificial intelligence consumes vast amounts of electricity. It–and its associated technologies–currently account for 2% of global energy use. The number is projected to double by 2026. If such increases continue, energy use for artificial intelligence has the potential to undermine the scant progress being made to address the climate crisis. Amazon and Google are already so worried about this that they are hoping to build their own nuclear reactors to power their machines.
This reminder prompts me to end almost exactly where I did earlier. Artificial intelligence, the replacement of meaningful work, the rise of a silicon God, whatever course things take, the most important truth before us remains this: We are children of the Earth. We are all bound up together, all members of the great family of all souls. The human future depends upon our ability to recall this. Vast computers, artificial intelligence or not, our central task is to take care of each other and common treasury that we call home.
Amen
