On this page
This guide follows directly from the train of thought in “How Credibility Works.” We’ve established what credibility is, how it works offline and online, and what amplification requires. Here we take the next step: given that AI is the condition we’re in—both the force that makes credibility harder to see and a tool that can help credibility travel—how are we to live with it?
Where We Left Off: The Condition, Not the Hero
Credibility is a judgment: someone holds you credible for something, to someone, in a context. Offline it travels through institutions and through people who vouch. Online, format is flattened—everything looks the same—so the signal becomes legibility (who created this, what have they done, who points to them) and topology (your position in a graph of trusted voices). Content that’s optimized and linked isn’t “more true”; it’s more legible and more connected. The playbook is: one home, author legibility (E-E-A-T), content as nodes, linking in and out, restraint. Success is faithful reach—the right people finding you and trusting you—not going viral.
AI doesn’t change that definition. It intensifies the problem and enables part of the solution. It makes more content look authoritative without any of the human backing. And it can reduce the friction that used to require agencies and endless time to make your work findable and circulating. So the question isn’t “is AI good or bad?” It’s: How do we live with AI in a way that protects credibility and helps it travel—without losing ourselves in the process?
This guide is about that. Not a list of prompts or tools. A way of life with AI: posture, boundaries, and practices that follow from how credibility actually works.
Reframe: Not a Tech Challenge—An Adaptive One
The first thing to get right is what kind of problem this is.
AI and credibility is not mainly a technological problem. You don’t need to understand models or APIs to lead here. You need to understand people: how trust is built, how it’s broken, and how you show up in a world where machines can mimic so many of the old signals. The challenge is anthropological—what it means to be a human who creates, vouches, and is vouched for—and therefore adaptive. You’re not solving an equation. You’re navigating a new environment with old wisdom: credibility is relational, conferred by others, and moral. The question is how to preserve and amplify that when the environment is full of synthetic content and algorithmic sorting.
So “how we are to live with AI” is an adaptive leadership task. It requires discernment, not just technique. It requires knowing what to refuse as well as what to use. It requires holding two truths: AI makes the credibility crisis worse and AI can serve credibility when it’s put in the right place—in the background, reducing friction, not in the foreground, replacing you.
The Tension We Hold: AI as Problem and as Solution
We don’t resolve this. We live in it.
AI as problem.
AI can generate text that sounds expert, consistent, and confident. It can fill the internet with content that has no author, no track record, and no one to vouch for it—but that looks as polished as yours. So the indistinguishability problem gets worse. Credibility was already hard to see online because format was flattened. Now the volume of plausible-but-unsourced content is enormous. When everything looks the same, trust collapses. AI also tempts us to substitute—to let the machine write what we would have written, to optimize for output instead of formation, to present generated work as if it came from our own formation and relationships. When we do that, we undermine the very thing that makes us credible: that we are a particular person, with a particular history and a particular network, saying what we actually believe.
AI as solution.
AI can reduce friction that used to block credibility amplification. Structuring content for findability, turning talks into articles, maintaining metadata, translating for other languages, suggesting internal links—these used to take agencies, budgets, and time that most movement leaders don’t have. AI can do some of that work in the background. So your existing work can become discoverable and connected without you having to become a full-time content operator. In that role, AI doesn’t replace you; it translates your work into the forms the digital environment can read. It serves circulation, not creation. And circulation is what makes credibility travel.
How we hold both.
We don’t choose “AI is good” or “AI is bad.” We choose where AI is allowed to go and where it must stop. We use it for tasks that don’t require our voice, our judgment, or our relationship—and we refuse it for tasks that do. We put it in the background (discoverability, structure, translation) and keep ourselves in the foreground (author, voice, vouching). That’s not a technical decision. It’s a moral one. Credibility is moral. So is how we live with AI.
How We Are to Live with AI: Posture
1. You are not competing with AI.
Your credibility doesn’t come from producing more or sounding smoother than a machine. It comes from being a particular person with a particular body of work and a particular network. AI can’t have your history, your relationships, or your formation. So the goal isn’t to out-output the bots. The goal is to make you—and what you’ve already done—legible and connected. AI can help with legibility (structure, findability) and connection (circulation). It can’t be you. Live from that.
2. Discernment over recipe.
No one can give you a fixed list: “use AI for this, never for that.” Contexts differ. Stakes differ. What you’re willing to delegate and what you’re not will shift. So we aim for discernment—the capacity to ask, in each situation, “Does this use preserve or undermine my credibility? Does it amplify me or replace me? Does it serve my people or confuse them?” That takes experimentation (you have to try things to learn what AI does well and poorly) and slowing down (you can’t discern at full speed). We don’t replace that with a manual. We build the habit of asking.
3. Formation and integrity over efficiency.
If the only question is “can AI do it faster?” we’ll hand over everything. But the question is also “does doing it myself form me and my people? Does it preserve what only I can offer?” Sometimes the right answer is to do it slowly, yourself. Sometimes the right answer is to let AI handle the scaffolding so you can focus on what only you can do. We optimize for faithful reach and credibility that travels, not for maximum output. That may mean less content and more impact.
4. Responsible agency.
You are capable of discernment. You are obliged to integrity. You are entitled to clarity—including the clarity that you don’t have to adopt everything, and you don’t have to refuse everything. You get to decide, in your context, where the line is. This guide doesn’t decide for you. It orients you so you can decide well.
How We Are to Live with AI: Boundaries
What we refuse.
We refuse uses of AI that undermine credibility or betray relationship.
- Deception. We don’t present AI-generated content as if we wrote it when we didn’t, or hide that we used AI when the reader has a reasonable expectation of human authorship. Transparency is part of trust. Where disclosure matters (e.g. substantive articles, formation content), we disclose or we don’t use AI for the parts that carry our voice.
- Replacement of relationship. We don’t use AI to simulate pastoral presence, mentoring, or care. Formation happens in relationship. AI can support (e.g. draft a follow-up email); it can’t be the relationship.
- Formation without presence. We don’t hand over the work that forms us—preaching, teaching, writing that carries our convictions—to the machine without our full engagement. If the output is going out under our name and is meant to form people, we’ve read it, weighed it, and owned it. Human-in-the-loop isn’t optional there; it’s moral.
- Amplification without credibility. We don’t use AI to make ourselves look bigger than we are—more output, more polish, more reach—when we haven’t done the work or don’t have the network to back it up. Credibility amplification fails when it outruns the credibility we actually have. We refuse that.
What we’re free to do.
We’re free to use AI for friction reduction and circulation when it doesn’t replace our voice or our judgment.
- Structure and findability. Outlines, headings, metadata, internal linking suggestions, SEO-oriented structure—AI can help. These tasks don’t require your voice; they require clarity and consistency. You review; you don’t have to generate from scratch.
- Drafting and expansion. AI can draft from your notes, expand a talk into an article, or produce a first pass. The human-in-the-loop rule: you edit, correct, and own the final text. The output carries your name only when it actually carries your voice and your judgment. So we use AI to get to a draft faster; we don’t publish the draft without passing it through our own formation and integrity.
- Translation and adaptation. Turning your work into other languages or formats so it can circulate—AI can assist. Again, where the result represents you, human review is required. The goal is circulation of your work, not generic content that happens to use your ideas.
- Administrative and repetitive tasks. Summarizing, formatting, scheduling, tagging—tasks that don’t carry your voice or your pastoral authority. Free to delegate to AI, with the usual caveat: you’re still responsible for what goes out under your name or your organization.
The yellow light.
Some uses are context-dependent. Same tool, different stakes. E.g. a quick internal summary vs. a public-facing piece. A first draft you’ll heavily edit vs. a post you’ll publish with light changes. We don’t give a universal rule; we give a principle. When in doubt, slow down. When the output will form people or represent you publicly, the human in the loop is you—reading, weighing, owning. When it’s low-stakes and internal, the bar is lower. You discern.
How We Are to Live with AI: Practices
1. Experiment before you adopt or refuse.
You can’t responsibly answer “should I use AI here?” or “where do I draw the line?” without having experimented. Try it. Use it for low-stakes tasks. See what it does well (structure, expansion, variation, formatting) and what it does poorly (voice, nuance, theological depth, originality). Learn where it saves time and where it wastes it, where it amplifies and where it homogenizes. That learning is the foundation for discernment. Without it, you’re either refusing out of fear or adopting out of hype. With it, you can make a grounded choice. So: play is prerequisite to discernment. Not a luxury. A requirement.
2. Human-in-the-loop for anything that carries your voice or your name.
When the output will be read as you—your teaching, your writing, your formation—you stay in the loop. You don’t publish AI draft as-is. You edit. You correct. You ask: “Does this sound like me? Do I believe this? Would I say this to someone I care about?” If not, you change it. The system can suggest; you decide. Credibility is moral. So is the decision to put your name on a piece of content.
3. Voice as constraint, not vibe.
Your voice isn’t a feeling; it’s something you can test. Does this paragraph sound like you? Would your people recognize you? We treat voice as a constraint that we preserve through feedback: we compare AI output to our prior work, we revise until it passes the “sounds like me” test, and we don’t publish until we’re satisfied. That takes time. It’s worth it. Credibility depends on readers trusting that the words are yours—that they come from your formation and your relationships, not from a generic language model.
4. Transparency where it matters.
When readers have a reasonable expectation that they’re hearing from you directly, we don’t hide that we used AI. We can say “I used AI to draft this; I’ve edited and own it” or “This piece was written with AI assistance.” We don’t have to label every sentence. We do have to avoid deception. Transparency supports trust. Secrecy about AI use, when the use is substantive, undermines it.
5. AI in the background, you in the foreground.
The sustainable pattern: AI handles discoverability and circulation—structure, metadata, translation, formatting, linking—so your work can be found and can move. You handle creation and authority—what you teach, what you write, what you put your name on. You’re the node. AI helps the node be legible and connected. It doesn’t become the node. Live with AI that way: as infrastructure for your credibility, not as your substitute.
6. Slowing down.
Discernment takes time. Experimentation takes time. Editing so that output actually carries your voice takes time. The culture will push you to move faster—to post more, to automate more, to “keep up.” We push back. Formation and integrity matter more than efficiency. Sometimes the right move is to do less and do it well. Sometimes the right move is to use AI so you can focus on what only you can do—and that might mean you slow down in one place (less time on formatting) so you can slow down in another (more time with people). The goal isn’t speed. The goal is faithfulness and credibility that travels.
How This Connects to Credibility
Everything above is just the credibility playbook applied to AI.
- One home, author legible. AI can help you maintain that home (structure, updates). It can’t be the author. You stay legible as you.
- Content as nodes. AI can help turn your existing work into findable, structured content. It can’t replace the work. The content that matters is still yours—your ideas, your formation, your voice—made discoverable, not generated from scratch to fill space.
- Graph position. AI can’t vouch for you. Only people can. So living with AI in a credibility-preserving way means we don’t let AI replace the human network. We use AI so our work can circulate within the network. We link, we cite, we show who we’re connected to. The graph is still made of people. AI just helps the graph read our work.
- Restraint. We don’t use AI to inflate our output or our reach beyond what we can back up. Credibility amplification fails when it outruns credibility. So we live with AI under the same rule: we make what’s true about us visible and connected. We don’t fake it.
When we live with AI this way, we’re not “doing AI.” We’re protecting and amplifying credibility in a world where AI is part of the environment. The aim is the same as in the credibility doc: the right people find you, trust you, and carry your work forward. AI serves that. It doesn’t replace that.
A Short Liturgy for Living with AI
You don’t have to say this out loud. But you can use it as a check.
- Before I use AI: Am I clear what I’m asking it to do? Is this a task that carries my voice and my authority, or a task that doesn’t? If it carries my voice, am I committed to being in the loop before anything goes out under my name?
- After I use AI: Does this output sound like me? Do I believe it? Would I say this to someone I’m forming? If not, I edit or I don’t publish.
- When I’m unsure: I slow down. I don’t have to decide today. I can experiment in low-stakes ways until I have enough experience to discern.
- When the pressure is to do more, faster: I remember that formation and integrity matter more than efficiency. Faithful reach is the goal, not maximum output. I can refuse the rush.
We don’t have to have it all figured out. We have to experiment (so we know what we’re deciding about), slow down (so we can decide well), and refuse the obvious wrongs (deception, replacement of relationship, formation without presence). The rest is discernment—and that takes time. That’s how we are to live with AI in 2026.