AI, Pope Leo XIV, The Salvation Army, and the Theology of Solidarity (Essay)
Estimated reading time: 0 minutes
We are living through a moment of seismic cultural change unlike anything experienced since the Industrial Revolution. The growth of AI is even more significant than the creation of the internet or even nuclear weapons. AI, and the fundamental questions it poses to us, is a deeply theological issue. To respond to AI’s questions, we need to rediscover a theology of solidarity. This is particularly important for The Salvation Army.
The introduction, growth, and widespread integration of increasingly complex AI programmes raise serious questions for all of us.
- AI is incredibly energy hungry, with data centres requiring huge amounts of power and water to operate.
- The integration of AI into social security systems, policing structures, and military programmes poses significant questions about the necessity of human agency.
- Emphasising optimisation and efficiency raises questions about the value of labour, the dignity of human life, and where we place our value.
- Digital poverty becomes increasingly stark as AI replaces roles, controls systems, and is integrated into everyday life.
- Technology is never neutral. AI is never just a tool. It always reflects the people who make it, programme it, and put it to use.
AI isn’t something that simply exists or is a neutral tool to be used. Neither is it something that is going to go away any more than we are going to stop using the internet.
We need to understand and engage with AI in a way that is reflective, liberating, and above all, holy.
This article explores how focusing on a theology of solidarity can help us negotiate the use of AI.
Solidarity is not simply empathetically identifying with people, nor is it adopting another’s position, or even being with another. Solidarity means taking responsibility for another person’s suffering, recognising that their suffering affects me, and actively seeking to secure justice for the suffering.
Our use of AI should support us in forming solidarity.
What is AI, and What Problems does it pose for Theology?
At its heart, the questions that AI raises are less about AI and more about what it means to be human.
The forms of machine intelligence we have created are shaped by human assumptions and beliefs. So these systems and structures can teach us, and sometimes warn us, about how our society and culture see what it means to be human.
What is AI? What can AI do? What can’t AI do?
Artificial intelligence is not one single technology. It is a family of systems designed to perform tasks that would normally require human intelligence — things like recognising faces, translating languages, writing text, or making decisions.
The term has been in use since the 1950s, but what we are dealing with today is far more capable and pervasive than early researchers imagined.
Large Language Models
The most important form of AI in public life right now is the Large Language Model, or LLM.
This is the technology behind tools like ChatGPT, Google Gemini, and Claude. An LLM works by processing enormous quantities of text — billions of words drawn from books, websites, conversations, and articles — and learning to predict which word is most likely to follow the previous one.
It does not read. It does not think. It does not understand. It produces text that is statistically likely to be coherent, accurate, and useful. Sometimes it is. Sometimes it is confidently, fluently wrong. This is called a hallucination.
A Large Language Model processes patterns, not meaning.
That distinction matters enormously. An LLM has no inner life. It has no experience of the world. It cannot suffer, hope, grieve, or love. It cannot encounter God. It has been trained on human words about human experience — which means it can reflect those things back to us — but it has not lived them.
When a Large Language Model produces a sermon about loss, it has never lost anyone. When it writes a prayer about hope, it has never hoped for anything.
This is not a reason to dismiss AI. It is a reason to understand it clearly.
What is AI Good For?
AI systems are genuinely impressive at processing large amounts of information quickly, identifying patterns, producing first drafts, translating languages, and supporting people with accessibility needs. Used well, these are significant gifts.
The question is never whether AI can be useful. The question is always: useful to whom, in what way, and at whose expense?
The Problems and Harm of AI
A fairness analysis of Universal Credit claimants in February 2024 confirmed that tools intended to ensure equity and efficiency may have discriminated against marginalised communities. Bias was found according to age, disability, marital status, and nationality.
Such a bias does not come from nowhere. It is a matter of programming and ‘training.’ Further research revealed that over 200,000 people had been wrongly investigated for housing benefit fraud and benefits errors due to poor algorithmic judgment.
Without a human backstop, thousands of claimants could have been wrongly declared ineligible due to algorithmic errors. The secrecy surrounding the system leaves the people harmed by it powerless to challenge it.
But AI is harmful to more than humans. It has an ecological impact. Global data-centre electricity use is estimated at 448 terawatt-hours in 2025 and could rise to 945 terawatt-hours by 2030. That energy has to come from somewhere. The associated water footprint of AI data centres is projected at 9.3 trillion litres.
As of 2025, only 32 countries (16% of nations) host AI-specialised data centres. 90% of that capacity is concentrated in China and the USA. The environmental costs are distributed globally, while the economic benefits are concentrated in two of the world’s richest and most confrontational countries.
AI Isn’t Neutral – The Theological Problems
AI systems also carry the values of the people who build them.
Every decision about what data to train on, what outputs to reward, and what problems to solve reflects human choices — and human biases. An AI trained predominantly on English-language text from wealthy Western countries will reflect the assumptions of that culture.
When those systems are then used to make decisions about welfare benefits, parole hearings, or medical diagnoses, those assumptions become policy. They affect real people. They affect the people the Salvation Army exists to serve.
This is why AI is a theological issue, not just a technological one. The questions it raises are not primarily about processors and code. They are about power, dignity, and whose humanity gets counted.
Babylon or Jerusalem? Pope Leo XIV on AI and Humanity
In 2026, Pope Leo XIV published his first encyclical titled Magnifica Humanitas.
What is a Papal Encyclical and Why Does it Matter?
A papal encyclical is a pastoral letter addressed primarily to Roman Catholic bishops and faithful laity, though it is also written for a wider audience.
In their internal role, encyclicals are an extension of the Pope’s formal teaching authority, and Roman Catholics are obliged to give these letters religious assent, which means presuming the letter is true and aligning their beliefs and actions with what it contains.
For the wider audience, an encyclical often has ethical authority, particularly for Catholic governments and politicians. The unique position held by the Pope, both in practice and in the public imagination, gives these public and pastoral letters huge influence beyond the Roman Catholic Church.
What Will We Build? Babylon or Jerusalem.
Pope Leo XIV said that AI is not going anywhere and is not simply a neutral tool; therefore, how we use it and what we use it for matter. He draws out this distinction through a metaphor that pits two cities, Babylon and Jerusalem, against each other.
People are going to build a city. But what kind of city will they build? One built on pride and arrogance, or one where God and humanity can dwell together.
The Tower of Babel was conceived without any reference to God. It relied on domination and uniformity. The proto-Babylonian Empire enforced the use of Akkadian throughout its dominions. It eliminated diversity and enforced homogenisation over any sense of communion.
When a city is built on that kind of pride and founded on a claim to self-sufficiency, communication breaks down, and people can no longer understand each other. The ultimate result is disruption, division, and harm.
Jerusalem stands in contrast to the city of pride. Pope Leo points to the story of Nehemiah, who brought the whole community of Jerusalem together to rebuild the city walls. Jerusalem’s walls aren’t built through the initiative and work of one man, but through everyone from every walk of life coming together. It is an undertaking with God at the centre where relationships are rebuilt through shared mutual labour for the common good.
Jerusalem is not built on domination, but on interdependence.
Pope Leo’s Warning
Pope Leo reminds us that technology is never neutral.
We cannot think of AI as just a tool like a hammer. Technology takes on the characteristics of those who create, finance, regulate, and use it. This is particularly true of things like AI, especially LLMs, which must be trained and taught how to respond and to ‘think’.
The uncritical use of AI by the general population, governments’ dependence on its systems, and its introduction into warfare pose a potentially existential threat. Not only to our society as a whole, but also to each of us as individuals. The locus of AI in neoliberal, late-capitalist societies poses a risk to the most vulnerable.
The choice is not between yes and no to AI. It is about whether we will use AI to build Babylon or Jerusalem.
Optimisation, Labour, and a Theology of Human Dignity
AI and the systems that depend on it are based on the assumption of optimisation. The faster a job can be done at the lowest cost with the fewest resources, the better.
AI, as it currently operates, functions both as an engagement of individual capacity and as an autonomous agent capable of performing repetitive or analytical tasks. The inherent capabilities of AI pose a threat to the workforce in the same way that the introduction of the steam loom or the word processor did in the past.
The theological issue is how we perceive the value of human labour.
Our economic and cultural systems tend to view labour through a cost-benefit lens. Labour has value because it is inherently productive. Its value is for the company’s owner, the business as a whole, or the shareholders.
Labour possesses no value for the labourer except for the wage that they are paid for their work. AI has emerged within that understanding of labour, and so, naturally, it is positioned to replace human work, since the goal is productivity.
Christian theology at its best must push back against this. For Christianity, labour has a broader and deeper purpose.
A New Industrial Revolution
Pope Leo XIV chose his name for a reason. Leo XIV chose his name to position himself within the legacy of Pope Leo XIII, who wrote the encyclical Rerum Novarum in 1891. Leo XIV signed his encyclical on the same day that Leo XIII signed his.
This matters because Rerum Novarum was written in response to the Industrial Revolution. Rerum Novarum emphasises the essential dignity of human labour, the danger of objectifying it, and the demand for fair wages and protections for it in the face of industrialised economies. It formed the basis for what became Catholic Social Teaching.
Leo XIV is grounding his teaching in the tradition of Leo XIII, in which labour is essential to human dignity. Labour cannot be reduced to purely productive or economic terms.
The Threat to Labour
A study by King’s College London into the early impact of AI on the workforce found that firms whose workforces were highly exposed to AI capabilities reduced their total employment by an average of 4.5%, with the effect concentrated almost entirely in junior positions, which fell by 5.8%.
Entry-level and early graduate jobs are most affected by the increasing integration of AI technology, which has a broader impact on the economy and living standards.
Retraining for a new role doesn’t always go well. Reflect on the miners’ experiences after the mines closed in the 1980s. The end of typing pools with the introduction of word processors, and the reduction in personal assistants through AI systems.
The Nobel Prize-winning computer scientist Geoffrey Hinton helped to build the AI technology that is reshaping our world. In September 2025, he warned that AI would create massive unemployment and that any economic gains from AI would be concentrated in a small number of people rather than the majority of society.
Fake News, Deep Fakes, and The End of Critical Thinking
Perhaps the most dangerous aspect of AI is its ability to distort and disrupt our ability to know what is true. We increasingly live in an age of doubt, where videos and images must be tested to determine whether they are real or AI-generated.
Any healthy society requires at least a basic ability to trust what is real. Fake news, especially when crafted to look real, is a tool of oppression and destruction. This is especially true for a democracy, where one of the most important elements of the good functioning of the democratic process is a well-informed electorate.
Not only does this doubt in what is real and true have a direct impact on our ability to know what is true. It also has a much darker register. AI can be used to commit abuse and harm against people. This is particularly the case with the creation of deepfake videos.
A deepfake creates a video in which a real-life person is made to act in a way they never did. This has been used to exploit women, create illegal content, and generally abuse and oppress the vulnerable. In this sense, AI poses a direct threat to people in our society.
This is a demonstration of the Babylon-or-Jerusalem argument. You could use this technology to create a training video or a cartoon (although both uses have implications for labour). Or you can use it to turn someone into an object, harming them and humiliating them in the process.
A more hidden threat is the end of critical thinking.
Young people are growing up using AI systems to create for them, yet lacking the skills and knowledge to use them effectively or to engage with them critically, because they sacrifice those basic skills to an artificial knowledge system.
We can see this in the increasing use of AI to write essays, conduct research, answer questions, and provide shortcuts in the learning process.
This fails to grasp the point of learning. It is not to create a passing essay. The purpose of a learning task is to increase understanding. How do we actually make and understand an argument? What do you do to engage with multiple competing sources?
Writing is part of that process. By choosing optimisation and expediency over effort and labour, something essential is lost.
Nehemiah, Human Flourishing, and Holiness
God wants us to be happy, to find joy, and to flourish in this life. Salvation is not just about going to heaven when we die. It is about how we live our lives here and now. God desires our good and calls us to support each other in finding that flourishing. Because we cannot flourish on our own. We need other people.
God has inscribed in our hearts a desire for happiness that encompasses all dimensions of our lives. Every part of our life, including our work, should fulfil us and contribute towards our flourishing. We are called to life “in all its fullness” (John 10:10). It is a sin that people have been forced to work in awful conditions simply to survive.
The story of Nehemiah can be helpful in understanding the fullness of life in every part of life.
The Dangers of Dependency: The Creation of Digital Bystanders
The problematic use of AI breeds dependency. We end up using AI for tasks that we find laborious or boring, but in doing so, we give ourselves something that matters. We give up our responsibility.
This can be as simple as believing the answer AI gives you without fact-checking it. Or giving your AI a task without first setting guardrails. However, it can also look like trusting the results of an AI system that is biased against people because of a particular characteristic.
Dependence and uncritical reliance on AI turn us into digital bystanders. We let things go by without questioning, without checking, and without asking where the answers have come from.
Human Weakness and Human Flourishing in Holiness
Human flourishing takes time and effort. We cannot take a shortcut to who we are meant to become.
Human limits such as illness, ageing, suffering and vulnerability are not simply defects to be corrected. They are what make us human. The need to be cared for, the vulnerability inherent in humanity, opens us up to our neighbours and creates the basis for a loving relationship.
Human flourishing does not depend upon removing our weaknesses and failings. Flourishing lies through them and in them. Because it is in our weaknesses that Christ’s power is made known.
Our flourishing emerges from community, not only from our individuality. To know myself, I need to know you as well. Our mutual interdependence creates the possibility for the kind of love and attention which forms us into flourishing human beings.
Life in All Its Fullness
The Bible says that Christ came so that we can have life in all its fullness (John 10:10). There are two words for life in the Bible: bios, which means everyday life that is shared with animals and plants and bacteria, and zoe, which means the kind of life that brings meaning and completeness to our existence. John uses the word zoë in this verse.
Flourishing and life in all its fullness go together.
We see this in the story of Nehemiah. The community comes together to rebuild the wall around Jerusalem. This task is not just about defence or infrastructure. It was the community reforming itself through shared labour. Gutierrez (America Magazine, interview with Daniel Hartnett SJ (2003)) wrote that:
The preferential option for the poor is ultimately a question of friendship. Without friendship, an option for the poor can easily become commitment to abstraction – to a social class, a race, a culture, an idea.
Pope Leo reminds us of what Jesus said, that the stone the builders rejected will become the cornerstone. While Christ was referring to himself, the same truth applies to human weakness and fragility.
The poor, the sick, the migrant, the refugee, the homeless, the disabled, the abused and oppressed, will become the cornerstone. It is not simply that we make space for them in our city. Rather, without them, our city cannot be built.
AI and The Salvation Army
The Salvation Army needs to work out how it will use AI at the organisational level as well as how its individual employees, ministers, and members should use it.
To continue to use the Pope’s imagery, the Army needs to decide whether it will contribute to Babylon or Jerusalem. The Salvation Army has already taken steps towards that decision. It has created its own guardrails for AI use. You can read them here.
The Salvation Army and Technology
The Salvation Army has never shied away from using technology to benefit the most vulnerable in society. From its earliest days, the Army understood that meaningful labour was necessary for dignity, and that education was necessary for people to enter into work.
The Army set up factories, recycling centres, laundries, book-binding workshops, and its training farm. People were taught how to use the technology, and then supported either to find work or to work for the Army in its own workshops. In In Darkest England, William Booth wrote:
I see the folly of hoping to accomplish anything abiding, either in the circumstances or the morals of these hopeless classes, except there be a change effected in the whole man as well as in his surroundings.
The modern inheritance of that process has been the UK Army’s Employment Plus scheme, which supports adults into education, training, or employment.
This is one of the ways the Army can draw on its heritage and missional resources to address the digital poverty that AI can exacerbate. The Army can support people in learning to utilise AI effectively and ethically, both at a personal level and in a work context.
The Risk of AI for The Salvation Army
AI poses a series of risks to The Salvation Army if it isn’t used in ways that build Jerusalem.
At the organisation level, the risk becomes using AI to operate systems that make decisions about people without effective guardrails, transparent programming, and human backstops being in place.
At the individual level, there is a danger that AI will be used to write sermons, pastoral notes, and missional strategies. There is a difference between writing a sermon and writing a work email. The sermon is fundamentally a response to God’s revelation.
An AI can help you do the research for a sermon (although letting the AI replace your own research or work is a form of digital bystanding), or spell check it for you, but it cannot write the sermon. It could write a speech about God, but not a sermon. Because a sermon is an act of faith, and AI cannot have faith.
The Theological Risk of AI
We need to understand that AI cannot experience or respond to God, and so there are limits to what AI can do.
We can use AI in ministry. There are plenty of tasks that are helpful and beneficial. But AI cannot know God, cannot have faith, and cannot experience revelation. We should not make the mistake of thinking that since AI seeks to talk to us, it is a person.
I think the root theological issue in our use of AI is the commercialisation of time. When we think of our time as having inherent value based on optimisation and efficiency, AI is a good choice for reducing the time a task takes.
Why not ask AI what this Bible verse means? It can search the same commentaries you can, and do so far faster.
However, the process of reading, researching, studying, and thinking are all moments when the Holy Spirit can speak to you. Your labour in studying the Bible is its own reward. Ministry is not an exercise in efficiency. It is the opportunity to be more fully human.
I use AI in my own work. It carries out spelling and grammar checks on my work. I also use it to help break down complex tasks into smaller step-by-step jobs, which was helpful when I had to try and fix an issue with an Excel sheet. In that role, it serves as an accessibility tool. I also use AI to help search through my Google Drive and format large documents.
I don’t use AI for writing, for pastoral care, for Bible studies or sermons, or anything generative. There is a difference between using AI as an enhancement and using AI as a replacement.
The key questions for using AI should be:
- Will using AI for this task improve my work, or will it replace my work?
- Is this use of AI enhancing my capacity, or is it creating dependence?
- Do I need to use AI for this task, or is it because I don’t want to do the work?
- Am I going to be honest about my use of AI and how it impacts the work?
- How am I going to fact-check the results the AI gives me?
- Is this a task that is dependent on faith or revelation? If yes, then AI cannot do it.
AI and a Theology of Solidarity
As I said earlier, Solidarity is not simply empathetically identifying with people, nor is it adopting another’s position, or even being with another. Solidarity means taking responsibility for another person’s suffering, recognising that their suffering affects me, and actively seeking to secure justice for the suffering.
We are called to solidarity as disciples with people in need. But we are also called to solidarity as a community, and as the Church. Bonhoeffer is helpful for us in understanding solidarity as a community. In the Letters and Papers from Prison, Bonhoeffer wrote that:
The Church is the Church only when it exists for others…not dominating, but helping and serving. It must tell men of every calling what it means to live for Christ, to exist for others.
For Bonhoeffer, the fundamental nature of the Church is to exist for the sake of others. The church is directed towards others. It does not exist for its own sake. In the same way as individual disciples are directed towards others. If we are going to use AI, we need to use it in a way which helps us build solidarity. This means being aware that many people do not have access to AI technology in the first place.
Genuine solidarity is visible in responsible action, and responsible action is motivated by genuine love, correctly directed at our neighbour in their need. This is the difference between sympathy and solidarity. Pope John Paul II wrote:
Solidarity is not a feeling of vague compassion or shallow distress at the misfortunes of so many people, both near and far. On the contrary, it is a firm and persevering determination to commit oneself to the common good.
– Sollicitudo Res Socialis (1987)
The Salvation Army needs to find ways to use AI to support its mission in a way that helps to build Jerusalem rather than Babylon. This includes how we can help people learn how to use AI, particularly to help them enter the job market. We need to ensure our staff are equipped to use AI in an ethical, loving, and solidarity-oriented way.
But most importantly, The Salvation Army needs to emphasise the inherent value of human labour. Not only as a source of essential dignity, but as the method by which formation and flourishing take place. This labour does not need to be purely physical, nor does it need to be intellectual. Labour looks different for everyone, depending on their capacity. The theological point is to support the active engagement of responsible disciples rather than the creation of digital bystanders.
AI can aid with that process, or it can get in the way. However we use AI, it must always be held accountable to justice for the oppressed, freedom for the suffering, and love for the unloved.
Sign Up To the MyTheologyCorner Newsletter for More
If you sign up for the newsletter, each week you will get a short summary of the main blog article, a bible reflection, book recommendations, updates on The Salvation Army in the news, a Salvation Army quote of the week, and a section of Salvation Army doctrine teaching.
