Whereas, that this the Third Convention of the Episcopal Diocese of the Great Lakes affirms our commitment through our baptismal covenant to seek and serve Christ in all persons, loving our neighbor as ourselves; and
Whereas, that this Diocese reaffirms its commitment to the importance of pastoral care provided through the Church to all those in need by bishops, priests, deacons, chaplains, and those who serve in other ministries of presence, counseling, and support; and
Whereas, that this Diocese expresses its belief that human beings make up the body of Christ and all believers and are called to, among many other things, the care of their fellow humans, that we are called to pastoral care by Christ and in the writings of Peter, Paul, and Luke among others in the Holy Scripture, and that while technological tools can help enable that care those tools can never replace the human role and responsibility in that care and often further distance us as members of the body; so be it
Resolved, that this Diocese commits to immediate action through prayer and discernment with regards to the ethics and morality of AI use both broadly and specifically with regards to its use in pastoral care, endeavoring to seek God’s timeless direction in a fast-changing world, with the goal of creating formal policy within one year to shepherd and protect the people and ministries of the diocese and stewardship of the resources entrusted to it; and be it further
Resolved, that congregations in this diocese are encouraged to create their own policies regarding the use of AI tools, including but not limited to large language models (LLMs) and generative pre-trained transformers (GPTs)1 , in pastoral care and in the church more broadly within the guidelines, directions and policies provided by this diocese and the Episcopal Church; and be it further
Resolved, that this Convention directs that the Secretary of Convention send copies of this resolution to the Presiding Bishop, the President of the House of Deputies, the Executive Council of the Church, Providence V Synod, and the Task Force on Artificial Intelligence and Intellectual Property so that they may know our commitments with regard to these issues and our desire to see action taken by our church in times like these; and be it further
Resolved, that all communities within the Episcopal Diocese of the Great Lakes entirely refrain from the use of AI in church matters until they officially adopt the above described AI use policy, either locally or from the diocese, which is consistent with expectations with all church business, that actions have policy governing them, both as outlined in the Church Manual of Business Practices and the Canons.
Prayerfully Submitted By
Mr. Jacob Russell, St. Andrew’s, Big Rapids
The Very Rev. Radhajyoti Michel-Trapaga, Central Michigan Episcopal Covenant (St. Andrew’s, Big Rapids & St. Mary’s, Cadillac)
Mx. Rayna Hauter, St. Paul’s, Flint
The Rev. Alex Quick BCC, St. Andrew’s, Grand Rapids
The Rev. Beckett J. Leclaire, Grace, Port Huron
The Rev. Charles A. Russell, St. Philip’s, Beulah
The Rev. Ryan T. Lockwood, Grace, Traverse City
Mr. Ryan Talaski, St. Alban’s, Bay City
Ms. Kathryn Borowicz, Holy Trinity, Manistee
Dr. Isabel J. Kempton, PH.D., St. Andrew’s, Big Rapids
1 Examples include ChatGPT, Claude, Grok, and EpiscoBot, among many others.
Explanation
We live in rapidly changing times; this is a truth that only becomes more sure as the years pass. Technology has changed the way that the Church carries out its work in incredible ways, for both better and worse, as we have entered the digital age. At its best, new technology allows us to evangelize and connect with others more effectively. At worst, it has sapped the humanity from the Church, leaving us without relationship despite having so many more ways to connect. The rise of ‘artificial intelligence’ tools, including general large language models (LLMs) and specific tools designed for use by the Church is the latest crest in this wave of change.
Our role as the Church, and more directly as leaders in the Episcopal Diocese of the Great Lakes, is to continue to shepherd God’s flock through these changing times while also stewarding creation. Through prayer, discernment, careful thought and discussion and action it is incumbent on us to follow God’s will and adapt to this changing world in a way that meets people where they are but also doesn’t give up the core of our faith. We know that times change but God is eternal. As the psalmist declared, “Before the mountains were brought forth or ever you had formed the earth and the world, from everlasting to everlasting you are God.”2
Pastoral care is a core part of the role of the Church, and is deeply important to the Christian Life. An Episcopal Dictionary of the Church speaks of this importance, and the cost of neglect, as follows: The work of pastoral care has always been deeply important in Anglicanism. It has been seen as the first work of the parish priest and of the bishop in a diocese. There have been times when pastoral care has been neglected on a wide scale. Those times have in turn provoked a demand among Christian people for a revival of pastoral care, as in the Wesleyan and evangelical revivals of the eighteenth century and the catholic revival of the nineteenth century. Each of those revivals has resulted from a sense that the church and the ordained ministry of the church have the care of God’s people as their first responsibility, even while the form of that care may vary in time and place.3
As we confront changing times, it’s important to consider the responsibility we carry as leaders to continue to grow as the Church for the better. Are we bringing Christ’s love to the people more effectively, or are we further disconnecting the parts from the body? Do the choices we make bring us together or further divide an already fractured church? When it comes to AI tools, the answers to these questions are not trivial. No such ability to manipulate and distort our society has ever existed. There are no previously tested answers for this new situation. We must consider questions of morality when it comes to the use of tools that can only exist on the back of human work taken and consumed with neither permission nor compensation. More urgently, as some have proposed using this emerging technology to either provide direct pastoral care or to advise pastoral care professionals, we must decide whether such tools are useful and good, or if they represent an abdication of our basic roles and responsibilities in the Church. We the undersigned posit that it is the latter, that no predictive large language model can provide or support pastoral care as described by the Episcopal Church and as prescribed by Holy Scripture.
Large language models, which make up most of what we know as ‘AI’ today and the specific tools of concern here are defined by Stanford University as “a type of artificial intelligence designed to understand and generate human-like text based on the input they receive. These models are built using deep learning techniques, particularly neural networks with many layers, which allow them to process vast amounts of text data and learn complex patterns in language.”4 The first limitation explained in that article is that “[d]espite their ability to generate human-like text, LLMs do not truly understand the content they process. They rely on patterns in the data rather than genuine comprehension, which can lead to plausible but incorrect or nonsensical outputs.”4 Put more simply, these are tools that predict what a response should be based on available data. They do not and cannot understand the content they provide, nor the context of the human to which they are responding, and cannot provide a human emotional connection. Further, by their very nature they cannot guarantee the privacy and confidentiality required in pastoral care, as their interactions are used to further train models.
The use of AI tools in pastoral care, both in support of and as a replacement to humans, is antithetical to the purpose of the work. This resolution proposes that we affirm the real and important role of humans in pastoral care. It supports the further engagement of humans with humans in pastoral care, rejecting the idea that AI models can or should replace us in those roles. Further it encourages both the parishes within our diocese and the broader Episcopal Church to consider deeply the questions presented here.
2 Psalm 90:2 (NRSVUE)
3 “Pastor, Pastoral Ministry”, in An Episcopal Dictionary of the Church, https://www.episcopalchurch.org/glossary/pastor-pastoral-ministry/, Retrieved 8/23/2026
4 “AI Demystified: Introduction to large language models”, Stanford University IT, https://uit.stanford.edu/service/techtraining/ai-demystified/llm, Retrieved 8/23/2026
Biblical Citations
28Keep watch over yourselves and over all the flock, of which the Holy Spirit has made you overseers, to shepherd the church of God that he obtained with the blood of his own Son.
29I know that after I have gone, savage wolves will come in among you, not sparing the flock. 30 Some even from your own group will come distorting the truth in order to entice the disciples to follow them. 31Therefore be alert, remembering that for three years I did not cease night or day to warn everyone with tears.
Acts 20:28-31 (NRSVUE)
1Now as an elder myself and a witness of the sufferings of Christ, as well as one who shares in the glory to be revealed, I exhort the elders among you 2to tend the flock of God that is in your charge, exercising the oversight, not under compulsion but willingly, as God would have you do it, not for sordid gain but eagerly. 3Do not lord it over those in your charge, but be examples to the flock. 4And when the chief shepherd appears, you will win the crown of glory that never fades away.
1 Peter 5:1-5 (NRSVUE)
11He himself granted that some are apostles, prophets, evangelists, pastors and teachers 12to equip the saints for the work of ministry, for building up the body of Christ, 13until all of us come to the unity of the faith and of the knowledge of the Son of God, to maturity, to the measure of the full stature of Christ. 14We must no longer be children, tossed to and fro and blown about by every wind of doctrine by people’s trickery, by their craftiness in deceitful scheming; 15but speaking the truth in love, we must grow up in every way into him who is the head, into Christ, 16from whom the whole body, joined and knit together by every ligament with which it is equipped, as each part is working properly, promotes the body’s growth in building itself up in love.
Ephesians 4:11-16 (NRSVUE)
Example Parish AI Policy 1: Grace, Traverse City, MI, AI Policy
Grace Artificial Intelligence Usage Policy – Approved and Adopted by Vestry 8/18/2026
Purpose:
Provide clarity and guidance on how we will and will not use Artificial Intelligence as a Christian community seeking to respond ethically and with respect to God’s creation. This policy specifically addresses Large Language Models, Generative AI, and technologies that emulate the agency and operational appearance of a human being.
Ethical and Religious Considerations:
Intellectual Property – All AI models are “trained” using massive data sets, which in many cases are created by “crawling” the internet and building models that include a tremendous amount of data created by human beings who are not compensated for their work. This creates a double injustice if AI is trained using the same type of work generated by those human workers without their authorization or compensation.
Any AI that can write creatively, create images or other audio-visual products, or research and create reports or presentations will likely have been trained on content not owned by the model creators, or trained to incorporate intellectual property of others without their informed consent or compensation.
Privacy & Algorithms – In many cases, especially social media, the “product” upon which firms make their money is not an application enjoyed by users, but the information about those users and the ways in which firms can influence their behavior, with data and services then sold to third parties.
AI makes it much easier for firms to create a composite of personal information, both real and based on predictive analytics, that many people would find violating their sense of privacy. Tech companies such as Flock and Palantir are utilizing AI for purposes of real-world surveillance, granting these companies unprecedented abilities to peer into the lives of individuals, match them to their online activity, and sell that data to the highest bidder.
Additionally, the use of artificial intelligence to supercharge the algorithms that dictate what content or information a user is presented with is known to have profound impacts on human behavior, beliefs, and points of view. Many consider the influence and suggestion used by such applications to be unethical or at least amoral.
Spiritual Formation – Christians have long pointed to the importance of practices and habits in shaping our hearts and minds. Consistently practicing patience makes one more patient, practicing love makes one more loving, etc. We are also shaped by the effort it requires to do something. For example, the effort required to wrestle with a passage of scripture, to interpret the world around us, or to empathize with and understand a person/viewpoint different from one’s own shapes and molds our souls. This is not unlike what happens in physical exercise, where strain put on muscles and tendons is precisely what causes them to become stronger.
When we utilize AI to generate text, summarize content, or create images, we circumvent this process and allow our intellectual muscles–our logic, capacity for empathy, and creativity–to become atrophied. This leaves us less able to put our own ideas into words, less able to understand those different from ourselves, and less able to think creatively about ourselves and our community.
Truth and Objectivity – There is a long and documented history of AI models returning information that is objectively untrue. Large Language Models, which make up most current generation AI products including Chat GPT, Grok, and Claude, have cited academic papers that do not exist, have caused harm to people’s reputations by conflating unrelated individuals, and helped to spread misleading information.
Additionally, the rise of AI generated content across the internet has seriously impacted our relationship to truth and fact. When video/photo evidence of an event was once solid grounds for believing that event actually happened, it is now possible to generate images/footage of anyone doing anything. This opens the door to unprecedented levels of disinformation, slander, and abuse. When reality can no longer be discerned from fiction, our personal and public life suffers.
Creation Care – AI uses a tremendous amount of electricity. Whether renewable or non-renewable, electrical power is limited. When giant corporations build data centers that use existing electrical infrastructure, that raises costs for people who need electricity for their daily lives. When data centers create their own power, they create excess environmental stress, which makes the surrounding environment less livable.
Many of the materials used to create the physical infrastructure for AI are relatively rare in nature and are often extracted or produced under circumstances that include unjust labor practices. In many cases, this has already driven up costs for those wishing to use similar technology for other purposes, which in turn adds to the incentive for raw materials vendors to participate in economic usury and slavery.
God has given us all that we need in creation to live and thrive in harmony with each other and with the natural world. The gift of “enough” to thrive with good stewardship is not the same thing as infinite resources. The process of developing AI infrastructure has not been executed with an eye toward harmony with nature and a just distribution of resources.
Respect for Life – This is a broader challenge having more to do with the near future of AI modeling. Existing AI models can already convincingly emulate an authentic personality. If we cannot know for certain based on a machine’s behavior, that it is in fact not sentient, then how will we know if a man-made machine is sentient? That is to have personal desires, preferences, original ideas, independent agency; things that indicate “life” worthy of spiritual consideration.
Hierarchy of Ethical Considerations
- Respect for Life – Not really our problem beyond a vigilant, prayerful, awareness.
- Not participating in a culture of creative theft – Doing all we can not to use AI features that are trained using content that has been effectively stolen from creators and authors.
- Care for God’s Creation – Being mindful of our share of the earth’s resources that our activities are going to consume.
- Personal Privacy and Autonomy – Being aware of the end use of information that is given to outside users and vendors, allowing people opportunities to opt out of such systems, maintaining an awareness that most digital media providers are incentivized to hold your attention through unethical practices and do so.
- Authentic Communication and Personal Development – Being careful that ideas, thoughts, words, and processes, are generated by people. Not using AI tools as a crutch or shortcut around critical thinking, learning, and problem solving. Maintaining vigilance against false information and other “AI Slop.”
Grace Episcopal Church Operational Policies Regarding AI:
For our purposes and most business applications, we can categorize AI applications by what they are built to do:
- Generative AI: Creates new content (text, images, audio, code) from learned data patterns.
- Predictive AI: Forecasts future outcomes by analyzing trends in existing data.
- Agentic AI: Autonomous systems that can make decisions, plan multi-step processes, and execute tasks without human oversight.
- Assistive AI: Tools designed to support and enhance human workflows (e.g., transcribing conversations in realtime, summarizing documents, or drafting emails).
Generative AI – Due to the ethical concerns inherent in virtually all generative AI applications, we will avoid using AI to create original content.
This, in practice, means avoiding AI that creates new content, documents, and products “from scratch,” and not feeding prompts into AI applications such as ChatGPT, Claude, etc.
Predictive and Agentic AI – Agentic AI products are one of the chief ways that corporations and other organizations displace human workers in the name of cost savings and efficiency. They are also often built on analytical data that may not be creative theft but include derived user information that is ethically dubious. Like all AI, Predictive and Agentic applications require a lot of power, electricity, and infrastructure.
We will use these applications thoughtfully and sparingly, keeping in mind the resources used by AI and the ethical questions around the dignity of human labor, data privacy, and personal manipulation.
We will make sure that anyone whose words or work could potentially be integrated into an AI summary, note, or other use is notified and gives explicit permission before activation of an AI process, feature, or application.
Assistive AI – These are likely the applications that we will make the most use of at Grace. Apps that can take notes, summarize information in original content, help edit, and make grammatical corrections to documents are appearing everywhere. While the models that train them are usually narrower and more specialized, they are not completely free of the ethical implications that come with other forms of AI. Additionally, there is a privacy concern when a person’s words, spoken or written, are recorded by an AI for the purpose of documenting.
We will use these applications thoughtfully and sparingly, keeping in mind the resources used by AI.
We will make sure that anyone who could potentially have their words or work integrated into an AI summary, note, or other use is notified and gives explicit permission before activation of an AI process, feature, or application.
We will be mindful of respecting the work and efforts of our fellow humans to communicate authentically by not using AI to summarize unread documents or to create communications (emails, memos, etc.) on our behalf
Products we are currently using that include AI features (not exhaustive)
- Opus AI: Uses predictive analytics and Large Language Models to extract clips from sermons and other original content that are optimized and formatted for use in social media.
- Canva: We use Canva’s basic features extensively, but avoid as much as possible any features that generate content.
- Grammarly: What used to be a fairly straightforward grammar-checking application now uses AI to make editing suggestions for clarity, sentence structure, and even tone.
- Zoom: While we have not yet used zoom’s AI features, such as note taking and transcription, they are available for staff and volunteers to use in keeping with our ethical guidelines.
- Be aware however that AI often fails to transcribe accurately, and frequently misunderstands conversational intent, noting the opposite conclusion between multiple debated options. Aggressive and immediate proofing is strongly recommended.
Example Parish AI Policy 2: Grace & St. Stephen’s Episcopal Church AI Policy
Understandings around the Use of AI
O God, who wonderfully created, and yet more wonderfully restored, the dignity of human nature: Grant that we may share the divine life of him who humbled himself to share our humanity, your Son Jesus Christ; who lives and reigns with you, in the unity of the Holy Spirit, one God, for ever and ever. Amen.
–Collect for the Second Sunday after Christmas
Purpose
The purpose of this document is to provide a theological framework and clear policies around the technology commonly referred to as “AI.” This document is intended to inspire all members of the Parish and followers of Christ to be thoughtful and intentional about how they choose to interact with and use technology, particularly this new technology that is rapidly changing our culture and world.
Theological Commitments
Anglican theology is often described as being “Incarnational” and grounded in the claim of John 1:14—“And the Word became flesh and lived among us, and we have seen his glory, the glory as of a father’s only son, full of grace and truth.”
As we consider how to use technology in ministry, we are ever–mindful of God’s intentional choice to work through the challenges and limits of human flesh.
At Grace & Saint Stephen’s, we have three values that serve as the foundation of our mission and identity: Beauty, Wisdom, and Witness.
Beauty
Then God said, “Let us make humans in our image, according to our likeness.”
–Genesis 1:26
From Genesis, we learn that God is generative and creative. To be made in the image of God means that humanity is made as creative creatures. In valuing Beauty, we prioritize this aspect of our nature and recognize that our sacred vocation is to be generative. In contrast, much of generative AI is not beautiful, but rather functional, predictable, uninspired, and bland.
Wisdom
For the wisdom about the cross is foolishness to those who are perishing,
but to us who are being saved it is the power of God.
–1 Corinthians 1:17
As we know from the Cross and from Jacob’s struggling with God (Genesis 32), we are blessed through challenges and struggles. Some of our modern idols are efficiency and perfectionism, both of which are promised through wider adoption of AI. However, in avoiding the limitations and struggles of our humanity, we avoid the growth and blessing that comes through wrestling with, instead of avoiding, the difficulties posed by discernment, creativity, and failure.
Witness
God reconciled us to himself through Christ and has given us the ministry of reconciliation.
–2 Corinthians 5:18
The Church is decidedly pro-humanity and our mission is to convey Christ’s reconciling love to all people and situations. This is a mission driven not by profits or efficiency, but comes through human interactions. Replacing human interactions with technological interfaces threatens the relationships that make up the Body of Christ. In our Baptismal Covenant, we affirm the “dignity of every human being.”
Over-reliance on “artificial intelligence” erodes human dignity, adds to environmental degradation, and outsources human potential, agency, and value to technology.
Furthermore, reliance on AI can lead to the unintended and undesired outcome of making human labor expendable.
Much of AI functions by scraping the Internet for information in a fashion akin to plagiarism and intellectual theft, which devalues human creativity and severs the author/sources from their context.
The Church is called to be an embodiment and sign of Christ’s reconciling alternative to the ways of empire, whether technological, political, or otherwise.
Understandings
AI can be a helpful tool in some situations. Like any tool, it can also be inappropriately used. This document is intended to set forth polices for appropriate usage in the context of ministry. AI is a helpful tool that has been in existence for decades in spellcheckers, automation, spam filters, virtual assistants, fraud detection, driving safety systems, facial recognition, and predictive text. There is nothing inherently anti-human about these technologies. The impetus for this document is the usage of what is often called “generative AI.”
Definitions
In many current conversations about AI, there is confusion because the blanket term “AI” is being used to describe an incredibly diverse set of technologies. For the purposes of this document, we recognize several different types of AI:
- Reactive AI: Technology that follows simple “If-then” instructions, often used in automation practices.
- Contextual AI: Technology that takes into account several factors before producing an output (self-driving cars, for example).
- Narrow Intelligence: AI that is trained for one specific task such as medical diagnostics or data analysis.
- Predictive AI: Can forecast outcomes based on historical data and patterns.
- Generative AI: Large–language models (Claude, Gemini, ChatGPT, etc.) fall into this category that is focused on mimicking human-like text and can also be used to create images, video, code, speech, and music. Generative AI produces the content, whereas other types of AI still rely on human created content and follows predictable rules.
Policy
Given the above theological commitments, understandings, and definitions, the policy for the usage of Generative AI is:
No AI created images, text, or content should be disseminated from any staff member to the parish. This includes prayers, counseling sessions, sermons, presentations, reports, flyers, banners, announcements, or graphics.
Though generally discouraged, if generative AI is used in the research and evaluative process for brainstorming or considering different and broader perspectives, the information should be reviewed, discerned, and presented using the God-given gifts of human wisdom and creativity. AI generated data should not be “cut and pasted” into a document, email, or presentation, as such actions are inauthentic and a form of plagiarism.
Any research or information obtained through AI should be verified by non-AI sources before being shared or presented.
Other types of AI (other than generative) can be used, though should not be overly relied on. Using the basic tools in Canva, Microsoft Office, or Google Workspace such as “smart crop,” spell checker, image alignment guides are acceptable.
Rationale
The policy against using AI to generate content that will be used in the ministry of this parish is grounded in our valuing of human potential and creativity, our emphasis on Beauty, and our trust that working through challenges is a virtue and blessing of being human.
Mitigations
Because the use of generative AI has become so common place, it may be necessary to be more intentional in avoiding over–reliance on this technology.
Employees are encouraged to have conversations with other staff or church members about ideas instead of turning to a Large Language Model. If an employee does not have time for such conversations, they can speak with the Rector about making sure they are given enough time and space for human dialogue.
If generative AI is being used to overcome insufficient resources or staffing, please speak with the Rector.
If additional training is needed to build skills that would allow for a lessened reliance on generative AI, please also speak with the Rector about continuing education opportunities.
We also have the opportunity to ask more of our members to share their time and gifts instead of using generative AI to do tasks that could be a practice of ministry for a church member.
Consider how the task would be handled if generative AI was not an option—what people, skills, our resources would be needed? Focus on using these more humanity– centric options instead of turning to AI as a shortcut.
Conclusion
Anytime a new technology is introduced, humanity must consider and contend with how that technology will change us. The adage is “We shape our tools and then our tools shape us.” Over most of human history, tools are introduced slowly, often over a period of years, or even decades. Generative AI has become nearly ubiquitous in a relative instant. The Christian practice of discernment takes time, and our primary concern is safeguarding human dignity, allowing for creativity to flourish through struggle, and avoiding the idols of efficiency and perfectionism. Just as God became flesh in order to save humanity, we commit to growing further into the humanity which Jesus has redeemed and blessed by his very nature.
