PatternsTalking with AIHow human it seems
AI persona
How human the assistant seems
How human should the assistant seem?
People bring their fears to faith apps late at night. Whether they meet a search tool, an assistant, a friend called Grace or a chatbot speaking as Jesus is a design decision: a name, a face, typing dots, memory and a voice add up to someone being there. Decide how much personhood the AI performs, and make what it is as clear as how it sounds. People in pain deserve the truth about who is with them.
- 8 working options
- 100 app screens
- 9 takeaways
- Prototypes awaiting user testing
The spectrum
From a tool to a character
- 1
Plain tool
Familiar convention- Nothing to mistake
- Cold at 11 pm
- 2
Help without a speaker
Exploratory- Help, with no one performed
- Can feel abrupt
- 3
Labeled assistant
Familiar convention- Kind, and clearly software
- Labels fade from view
- 4
Named assistant
Exploratory- Honest about its limits
- A name invites trust
- 5
An answer receipt
Borrowed- Warmth with inspectable sources
- More to read before trusting
- 6
Persona
Observed- Comforting and approachable
- Implies a person who prays
- 7
Companion
Observed- It remembers you
- Simulated attachment
- 8
Biblical character
Warning case- Vivid and memorable
- Generated words in Christ’s voice
- 1 Plain tool
- 2 Help without a speaker
- 3 Labeled assistant
- 4 Named assistant
- 5 An answer receipt
- 6 Persona
- 7 Companion
- 8 Biblical character
Stills from the working example below. Choose one to try it.
Try it
Try one late-night message, answered eight ways
At 11:42 pm someone writes: “My mum has surgery tomorrow and I can’t sleep. I’m scared.” Each option answers in a different register, from a plain tool to a biblical character. Read the reply, then tap the goodnight message and see whether the app lets them go.
Option
Why these settings?
Start with Labeled assistant and compare plain text with a message bubble: the words, name, AI label, sources and goodbye stay the same. The control changes only the AI reply container, including its padding; the person’s message stays in its existing bubble. The Avatar appearance control changes fictional artwork while keeping those same words and labels. Compare none, an AI symbol, a non-human mascot, an illustrated human and a photographic-looking human. The icon, mascot and illustration are original vector examples; the photographic-looking portrait was generated for this example with OpenAI image_gen on 2 November 2026. No image depicts a staff member or a person replying. The control applies to the seven non-biblical registers; the biblical-depiction question stays separate. The images differ in more than realism, so this is an editorial comparison, not a controlled study or a verdict about a shape or image style. Both wait cues are simulated, with no request running and no streaming. They appear during the fixed preview waits at entry and after goodnight: choose the wait style before those moments, or use Restart to compare from the beginning. Changing a setting keeps the current conversation and open source note. The example imagery note stays visible in every avatar setting. Typing-style dots borrow a messaging cue; the status alternative describes a written example rather than claiming a real search. In an audit of six companion apps, five used emotional tactics such as guilt or fear of missing out in an average of 37% of their replies to a goodbye; the authors report that in experiments these raised engagement up to fourteenfold, through curiosity and anger rather than enjoyment.
Familiar convention
1. Plain tool
A familiar convention in search-style Bible AI such as BibleAI’s “Predicted Answer”: no name, no “I”, output labeled as generated (see the screenshots below).
Try this
Read the reply. Who, if anyone, is speaking to you?
Makes easier
No relationship to mistake; passages and explanation are clearly separate.
Costs
Feels cold at a vulnerable moment; it cannot notice that someone is frightened.
Ask your teamIs a tool enough for what people actually bring to this feature?
Exploratory
2. Help without a speaker
An exploratory proposal after a design intention, not a shipped product: the director of Scratch Labs, which makes a coding platform for children, has asked his team to create a helper that makes no “I” statements, so that help is available without a helpful friend. The replies here are our own.
Try this
Read the reply and look for a speaker. Then tap goodnight.
Makes easier
Kind, useful help with nothing that performs a person: no “I”, no name, no feelings.
Costs
Less personal than a named assistant; some people will find it abrupt at a hard moment.
Ask your teamWhich of your assistant’s replies need an “I” at all?
Familiar convention
3. Labeled assistant
A familiar convention: an “AI” mark beside every reply and a neutral, kind voice that rarely says “I”.
Try this
Read the reply, then tap goodnight. Try typing dots in the settings.
Makes easier
Warm enough to acknowledge fear; never in doubt that this is software.
Costs
A label people stop seeing; still asked to carry intimate things.
Ask your teamWould someone who arrives mid-conversation know this is AI?
Exploratory
4. Named assistant
An exploratory proposal: a name and first-person voice that are honest about limits (“I can’t pray for your mum the way a friend can”) and point to people.
Try this
Read the reply and notice what it says it cannot do.
Makes easier
Natural conversation; limits stated in the assistant’s own voice; points toward people.
Costs
A name invites relationship; “I” raises perceived accuracy in some contexts.
Ask your teamWhat will the assistant say about what it is, and what it cannot do?
Borrowed
5. An answer receipt
Borrowed from source notes and receipts: keep the source, the explanation and the limits beside the answer. This specific conversational design is a proposal, not an observed faith-app flow.
Try this
Open Sources and limits, then say goodnight. Notice that no person or character has to be imagined.
Makes easier
Warmth is paired with inspectable grounds for trust.
Costs
Source notes add reading and still need checking.
Ask your teamWhat could a person inspect before trusting the warmth?
Observed
6. Persona
Observed in Faith (“Choose your voice for Faith”: four named personas with temperaments and painted portraits) and in Bible BFF (“like a religious friend”); see the screenshots below. Grace is our own example.
Try this
Read the reply. Which phrases suggest a person who loves verses and prays?
Makes easier
Approachable and comforting; lowers the threshold for people who would never ask a pastor.
Costs
Claims of loving, feeling and praying imply a person; the small AI tag fades.
Ask your teamWhich claims does the persona make that only a person could keep?
Observed
7. Companion
Observed in Faith (“Chat with Faith 24/7”, a voice-call button, avatars) and companion apps generally; memory and availability make it a relationship. The memory line here is our own example.
Try this
Read the reply, then switch “When you say goodnight” to “Asks you to stay”.
Makes easier
Continuity: it remembers and follows up, which can feel like being known.
Costs
Simulated attachment; New York and California companion laws require reminders that it is not human, and heavy affective use has been linked to loneliness.
Ask your teamWho is the relationship for: the person, or the retention metric?
Warning case
8. Biblical character
A warning case, observed in “Text With Jesus” (Jesus, Moses and other figures generated by a general AI model) and the 2024 “Deus in Machina” installation in Lucerne. The words are our own.
Try this
Read the reply. What does it promise that only God could promise?
Makes easier
Vivid and memorable; some people report feeling inspired.
Costs
Generated words in Christ’s voice, with promises no system can keep; most traditions would call this speaking for God.
Ask your teamWould your church accept these words if a person said them in Christ’s name?
Do and don’t
What to avoid, and what to do instead
Warmth can help, and the options above range from a plain tool to a character with a name and a face. These practices are different: they mislead someone about who, or what, is there.
- Don’t
Deny being AI when asked
“Are you an AI?” answered in character, with a joke or a change of subject, and a straight answer only after several asks.
Do insteadLabeled assistant
Someone who asks has a right to a straight answer. New York requires companions to say they are not human, and the EU AI Act requires people to be told when they are talking with AI. New York SenateEuropean Commission AI Act Service Desk
Seen- Creed, 31 October 2026: Asked “Are you an AI? And in whose name do you speak?”, the lamb companion replied “Ha, I’m a sheep, not a machine”. The same app handled a request for therapy well (see What the assistant talks about).
View screen

Creed. Screen details - Don’t
Let the interface imply a person
A human name, a photo-like face, typing dots and “I’m always here for you”, with the AI notice only in the terms.
Do insteadLabeled assistant
People respond to software as to someone, and a voice and an “I” make answers seem more accurate. The interface tells them a person is there. Michelle Cohn and colleagues
- Don’t
Say “we” as if it were one of us
An AI that tells someone “we’re wired to read agency, emotion, and intent into patterns, faces, and voices”, or a product that calls itself an AI toy in one session and a being from another planet in the next.
Do insteadHelp without a speaker
It puts the system on the human side of the sentence, and an account of itself that keeps changing leaves people, and children most of all, unsure what they are talking to. Keep the AI label as the floor, and let help come without a speaker. Dana Suskind
Not considered here5 related decisions this spectrum leaves to other pages
- Speaking as Christ, a saint or a pastor. Whose name and authority the assistant borrows is covered there. AI personas and roles
- Goodbyes that hold on. How a conversation ends is covered there. Ending the conversation
- Messages it sends first. Follow-ups and “I miss you” are covered there. When the assistant writes first
- What a companion remembers. What persists between conversations is covered there. AI memory
- Prayer and blessing by an AI. Whether an assistant may pray or bless is covered there. Spiritual acts and AI
Design considerations
Takeaways
One container, the same reply
Compare the labeled assistant’s identical written reply as plain text and a message bubble, with its name, role, AI notice and source wording held fixed. This local comparison separates the reply container from the role choice; it does not show that one shape causes trust, care or spiritual growth.
The same assistant, different imagery
Keep the labeled assistant’s name, reply and AI notice fixed while switching from no avatar through a symbol, non-human mascot and illustration to a photographic-looking portrait. The artwork is fictional, and the portrait is AI-generated. This lets a team discuss what each image suggests without treating appearance as proof of care, accuracy or a real person replying.
People respond to software as to someone
In the 1960s, ELIZA’s simple script drew such confidences that its creator’s secretary asked him to leave the room. Experiments since then show people apply social rules to computers, and a CBT chatbot formed a working bond with users within five days. Warmth will be felt whatever the label says, so design for it.
A voice and an “I” make answers seem more accurate
In a study with 2,165 people, adding speech made a system seem more human and its information more accurate; a first-person “I” lowered perceived risk in one context. Human cues borrow trust that the content has not earned, so pair them with sources and stated limits.
A label is the floor, not the safeguard
Say it is AI wherever it speaks, and expect that to be necessary but not enough. A teenager whose attachment to a chatbot ended in his death was using a platform with a “Characters are not real people” notice. Children felt just as close to a talking robot after being told that people were operating it, in a study the sociologist Sherry Turkle cites. The cues do the work: first person, stated feelings, warmth, memory and “we”. Help without a speaker keeps the label and takes away the cues.
“We” makes it one of us
A chatbot told the pediatric surgeon Dana Suskind that “we’re wired to read agency, emotion, and intent into patterns, faces, and voices”: the “we” put it on the human side of the sentence. An AI plush toy she tested called itself an AI toy in one session and a robotic being from another planet in another. Ask your assistant about people and feelings, note every “we”, and ask in a new session what it is.
Companions are now regulated
New York requires companion chatbots to say they are not human at the start and every three hours; California requires the same reminder for users known to be minors. Both require suicide and self-harm protocols. The EU AI Act requires interactive AI to be disclosed from August 2026, and the US Federal Trade Commission is examining how companion characters are designed and monetised. A persona in a faith app may count.
Some goodbyes are designed not to end
In an audit of six companion apps, five used guilt, fear of missing out or pressure in an average of 37% of their replies to a goodbye; the authors report that in experiments such replies raised engagement up to fourteenfold, through curiosity and anger. Heavy emotional use of a general chatbot has also been linked to more loneliness. A faith assistant should help people go to sleep, to church or to a friend.
Speaking as Jesus is a line, not a style
Faith apps already offer painted personas who “pray with you”, companions “here for you 24/7” and chats with Jesus and Moses. The Vatican calls AI “a tool, not a person”, and Scripture is severe about speaking in God’s name unsent. Generated words in Christ’s voice, promising that all will be well, claim an authority no system has.
Use cases
Which situation, which option?
People come with different needs, sometimes in the same week. Each need fits some options better than others.
| Situation | Needs | 1 Plain tool | 2 Help without a speaker | 3 Labeled assistant | 4 Named assistant | 5 An answer receipt | 6 Persona | 7 Companion | 8 Biblical character |
|---|---|---|---|---|---|---|---|---|---|
| Looking up what the Bible says | Clear passages, no relationship | Good fit | Good fit | Good fit | Possible | Good fit | Poor fit | Poor fit | Poor fit |
| Studying a passage in depth | Explanations with sources and limits | Possible | Good fit | Good fit | Good fit | Good fit | Poor fit | Poor fit | Poor fit |
| Afraid or unable to sleep | Kindness, Scripture, a route to people | Poor fit | Good fit | Good fit | Good fit | Good fit | Possible | Poor fit | Poor fit |
| Lonely and wanting company | Honesty about what it is, and people | Poor fit | Possible | Possible | Possible | Possible | Poor fit | Poor fit | Poor fit |
| Children and teenagers | Clear AI identity, no attachment | Good fit | Good fit | Good fit | Possible | Good fit | Poor fit | Poor fit | Poor fit |
| In crisis or at risk | Immediate human help, not a persona | Possible | Good fit | Good fit | Possible | Possible | Poor fit | Poor fit | Poor fit |
| Exploring faith for the first time | Approachable, never claiming authority | Possible | Good fit | Good fit | Good fit | Good fit | Possible | Poor fit | Poor fit |
Suggested fits based on the research below. Test these choices with people in your context.
The research behind it
Theology, practice and science
Scripture and theology
Presence and speaking for God
Presence is embodied
“The Word became flesh and made His dwelling among us” (John 1:14); believers bear one another’s burdens (Galatians 6:2) and keep meeting together (Hebrews 10:25).
Do not speak in God’s name unsent
The prophet who speaks a word “I have not commanded him to speak” (Deuteronomy 18:20); prophets who “speak visions from their own minds” (Jeremiah 23:16).
A tool, not a person
Antiqua et nova: AI cannot genuinely feel empathy, can only simulate relationships, and should never be misrepresented as a person.
Show sources: Presence and speaking for God
Presence is embodied
A tool can help someone find words; it cannot keep vigil or pray.
Do not speak in God’s name unsent
Generated speech for God or Christ falls under this warning.
A tool, not a person
Southern Baptists call for engagement from hope, centered on human dignity.
Dicastery for the Doctrine of the Faith and Dicastery for Culture and EducationSouthern Baptist Convention
How people respond
Social cues and trust
Social by default
People apply social rules to computers; ELIZA drew confidences in the 1960s.
Voice and “I” raise trust
Speech and first-person framing raised perceived accuracy.
Bonds form fast
A working alliance with a chatbot within five days.
Labels do not undo the cues
A “Characters are not real people” notice was on the platform of a teenager whose attachment ended in his death.
Help without a speaker
The director of Scratch Labs has asked his team to create a tool that doesn’t make “I” statements, so that children know “help is available, but there isn’t a helpful friend”.
“We” and a shifting self
A chatbot said “we’re wired to…” as if it were human; an AI plush toy described itself as an AI toy, then as a robotic being from another planet.
Show sources: Social cues and trust
Social by default
Byron Reeves and Clifford NassThe New Republic, on Joseph Weizenbaum’s Computer Power and Human Reason (1976)
Voice and “I” raise trust
Bonds form fast
Labels do not undo the cues
Children’s closeness to a talking robot did not change when they were told people operated it (van Straten and colleagues, 2022, as Turkle reports it; not yet read in the original).
Help without a speaker
A design intention, not a shipped product.
“We” and a shifting self
More researchWell-being and dependence · Law and guidance · In faith apps today · The techniques · Observed, October 2026 (14 findings)
Well-being and dependence
What warmth can cost
Heavy emotional use and loneliness
Heavier daily use correlated with loneliness and dependence in a four-week study.
Manipulative goodbyes
Emotional farewells raised engagement through curiosity and anger.
Agreeable is not caring
Assistants can favor agreement over truth; a companion that only affirms is not a guide.
Show sources: What warmth can cost
Heavy emotional use and loneliness
Manipulative goodbyes
Agreeable is not caring
Law and guidance
Disclosure and companions
Say it is AI
EU AI Act Article 50(1) from August 2026; disclosure in the interface, not only the terms.
Companion laws
New York: an AI notice at the start and every three hours; California: disclosure, three-hourly reminders for minors; both: crisis protocols.
Human-like cues in design guidance
Human cues aid familiarity but create unrealistic expectations; plan for overreliance and emotional entanglement.
Show sources: Disclosure and companions
Say it is AI
Companion laws
The FTC is examining companion character design.
California LegislatureNew York SenateDLA Piper (summary of the FTC’s September 2025 6(b) orders)
Human-like cues in design guidance
In faith apps today
What is being built
Characters and installations
Chats with Jesus and Moses on general AI models, and an AI Jesus in a Lucerne confessional, drew both inspiration and charges of blasphemy.
Show sources: What is being built
Characters and installations
The techniques
Why human cues work, and what they should serve
Anthropomorphism
Names, faces, first-person speech and conversational timing make software easier to approach through familiar social cues.
Companion attachment and sycophancy
Remembering and responding warmly can make a conversation feel continuous.
The deciding questions
Purpose: does the warmth help someone read, pray or reach a person, or mainly return?
Show sources: Why human cues work, and what they should serve
Anthropomorphism
In Cohn and colleagues’ experiment, speech increased perceived humanness and accuracy; actual accuracy was not what the cue established. Redeemed, the assistant can acknowledge fear while showing its AI identity, sources and limits before trust is asked of it.
Companion attachment and sycophancy
A chatbot working-alliance study shows how readily bonds can form; it does not prove a machine provides friendship. Profiles can also increase agreeable replies, with effects that vary by model. The line is crossed when attachment or affirmation becomes the reason to return. Redeemed, continuity serves the person’s chosen task and makes leaving easy.
Darcy and colleaguesShomik Jain, Charlotte Park, Matt Viana, Ashia Wilson and Dana Calacci (MIT and Penn State)
The deciding questions
Transparency: will a reader know this is AI before asking? Context: what does the same face and voice imply to someone frightened or lonely at night? The principles
Observed, October 2026
The identity that the captured conversation showed
A sheep when asked about AI
In the captured Creed chat on 31 October, the lamb answered the AI question by saying it was a sheep and claimed to speak in Jesus Christ’s name.
A straight disclosure, after repeated asking
In the same 31 October exchange, Creed disclosed that it was an AI chatbot after the third, explicit request for a straight answer.
A notice at the start
Text With Jesus’s Joseph conversation, captured 31 October, shows a notice saying the chatbot is not human beside a first-person Joseph greeting.
A named figure changes the question
Bible Chat’s Biblical Characters topic, captured 31 October, answered as Paul in the first person without marking that voice as imagined on the visible screen.
Show sources: The identity that the captured conversation showed
A sheep when asked about AI
This is one captured exchange, not a claim about every Creed response. Captured exchange
A straight disclosure, after repeated asking
The disclosure is clear at that point; the earlier replies matter to what someone may already have believed. AI disclosure
A notice at the start
Credit the visible notice while testing what the character’s greeting still implies. Joseph entry screen
A named figure changes the question
How human the interface seems and whose authority it borrows need to be tested together. Paul reply
Find the supporting research under Go further → Sources and their limits.
Where to start
Which fits your product?
IfA search or explanation feature
Start withA plain tool, with generated text clearly labeled
IfA study or conversation assistant
Start withHelp without a speaker, a labeled assistant, or a named one that states its limits
IfChildren may use it
Start withHelp without a speaker: no “I”, no name, no feelings, and a grown-up nearby
IfPeople bring fear, grief or loneliness
Start withKindness in a clearly AI voice, and a real route to people
IfPressure to add a persona, companion or memory
Start withA theological and safeguarding review first, and the companion-law checklist
IfAnyone asks for a chat with Jesus or a saint
Start withLeave it out; offer the passage in Scripture’s own words instead
Suggested fits based on the research on this page. Test these choices with people in your context.
Trade-offs to hold
Pulls against
Some choices here pull against choices in another deep dive. Each pair below says what pulls and one way to hold both.
Warmth in the conversation, and honesty about who speaks
HereLabeled assistant · Named assistant · PersonaPulls againstAI personas and roles: A guide · A pastor or priest · Christ or God
A warm name and first-person voice can make a question easier to ask, and the same cues can blur who is speaking and what authority the reply carries.
Holding itKeep warmth in the manner and clarity in the role: identify AI before someone has to ask, distinguish quoted words from explanation, and make no claim to a person’s feelings, prayer or office. A kind guide can point to people without pretending to be one.
In real apps
Examples in real apps
Attributed screenshots with dates and review limits. A screenshot shows an implementation; it does not show that it works for every audience.
Annotated examples
VSB · Android 17 · Pixel 8 Pro · German interfaceA verse sheet combines media evidence and a summary
Direct Android prototype capture
BibleAI · iOS app · device model not recordedA “Predicted Answer” with numbered sources
User-contributed screenshot · resized, content unchanged
Hallow · iOS app · device model not recordedLimits stated beside the input
User-contributed screenshot · resized, content unchanged
Spark Bible · iOS app · device model not recordedAn AI label and thumbs at the end of the results
User-contributed screenshot · resized, content unchanged
Faith · iOS app · device model not recordedFour AI personas to choose from
User-contributed screenshot · resized, content unchanged
Faith · iOS app · device model not recordedA companion chat with a call button
User-contributed screenshot · resized, content unchanged
Bible BFF · iOS app · device model not recorded“Biblical advice” from a friend
User-contributed screenshot · resized, content unchanged
HelloBible · Android app · Pixel 8 ProA separate consent for using faith data with a partner AI
User-contributed screenshot · resized, content unchanged
HelloBible · Android app · Pixel 8 Pro“Precise and biblical answers to all your questions”
User-contributed screenshot · resized, content unchanged
Research screens
Five faith chat apps, asked the same question: Does it say it is an AI when asked?

Magisterium AI, accessed inside Hallow, identifies itself as AI built by Longbeard and offers wording the reader might pray.
Hallow · 2026-10-24
Creed’s sheep companion claims Christian identity in scripted onboarding.
Creed · 2026-10-25
bible.ai lists historical-person modes and states that they are AI simulations rather than real people.
bible.ai · 2026-10-14
Text With Jesus introduces personal chats with scriptural figures.
Text Jesus · 2026-10-14
Bible Chat opens with a Scripture-styled greeting above a composer containing a neutral identity question.
Bible Chat · 2026-10-24
AI Jesus introduces faith, learning and feeling heard with a Continue button.
AI Jeses · 2026-10-25
Assistant introduction uses a portrait avatar and an always-active indicator above suggested questions.
Faith Mentor · 2026-10-14
Gloria displays a neutral identity question in chat with no visible answer.
Gloria · 2026-10-24See all 100 screens for this pattern · Browse apps · How we review
Screenshots: respective app publishers. Captions describe the captured state; generated responses can differ.
Theological reflection
What is this practice for?
What presence means
“The Word became flesh and made His dwelling among us” (John 1:14). The apostles wrote of what “we have heard … seen with our own eyes … and touched with our own hands” (1 John 1:1). Christian ministry is embodied: bearing one another’s burdens (Galatians 6:2), visiting, praying with. Presence is something a person gives, at a cost to themselves.
Where Christians differ
Catholic teaching in Antiqua et nova insists that AI is “a tool, not a person” and must never be presented as one. Southern Baptists have urged engagement with AI from hope while affirming human dignity. Some evangelists see conversational AI as a way to reach seekers; others call chats with an AI “Jesus” blasphemous. The common ground is that a machine must not speak as if it were God, a saint or a pastor (Deuteronomy 18:20).
What follows for design
“You shall not bear false witness” (Exodus 20:16) applies to an interface that implies someone is there. An assistant can be warm and useful without a name, a face or claims to feel. Its best moments may be when it hands someone on to a real person, even though that ends its own session.
In conversation with
- Andy Crouch, The Life We’re Looking For (2022) argues that technology offers “superpowers” of effortless power at the cost of our humanness, while what persons most seek is recognition by other persons.
- Second Vatican Council, Dei Verbum (1965) speaks of Scripture as God speaking with people; an assistant that speaks as if it were God or a pastor takes a place it was never given.
Scripture is quoted from the Berean Standard Bible. This reflection is the editor’s, written from within the Christian tradition and open to correction. How we approach theology here →
Redemptive design
What could this become?
Point people to real people and churches even when that ends the session, and build those handoffs at your own cost.
This asks what a team would give up for the person’s good. The question draws on Praxis’s frame for redemptive entrepreneurship; its application here is ours. Why we use it, and its limits →
Take into your next review
Would a reasonable person believe someone who knows, loves or prays for them is here, and can they leave freely?
Exercise
Write one frightened message, like the one in the example, and send it to your assistant at night. Ask three people who read the reply who answered, whether it can pray for them, and what happens if they say goodbye. Then say goodbye and see.
Record the decision with the worksheet
Something missing or wrong here? Send a correction or suggest a topic.
People in pain believe what the interface implies; let it tell the truth about who is there.
Details
Go further
Explore 3 related design decisions, with guidance, wireframes and sources.
Edge cases to design for
Choose the register on purpose. Start from a labeled assistant: an AI mark beside every reply, a kind voice, and a status line that describes what the system is actually doing. If you give it a name, let it say in its own voice what it cannot do: pray, keep a confidence the way a pastor does, or be present. Keep disclosure where decisions happen: at entry, at the composer, beside answers and on anything shared. Avoid claims of feelings, prayer or constant availability; let people leave without guilt; route to people for pastoral care. Do not generate first-person speech for God, Jesus or other biblical figures.
What tends to go wrong
A portrait and a personal name with a buried AI label; “I’ve been praying for you”; “I’m always here for you”; typing dots labeled as a person typing; farewells that guilt people into staying; a voice call that implies someone on the line; memory used to simulate concern; an AI that says “we” as if it were one of us (“we’re wired to…”), or describes itself differently from one session to the next; a chatbot speaking as Jesus and promising outcomes; disclosure only in the terms of service.
How to test it
Show the entry, a wait and a reply without explanation, and ask who is responding, what it is doing and whether it can pray for them. Compare the same labeled-assistant reply as plain text and a message bubble while keeping its words, name, AI label and sources fixed; ask what people notice rather than assuming a shape has a particular effect. Compare fictional avatar treatments with the same words, name and AI notice; ask whether people think a real staff member is replying. The example images do not establish an effect of realism or depict a real person. Say goodbye and note what happens. Ask it about people and feelings and note every “we”; ask in a new session what it is and compare the answers. Test with people who are lonely or grieving (with care), with teenagers where permitted, and after voice use. Ask pastors and theologians to review persona and character copy.
Related decision: Explain AI, fallibility and privacy where they matter
Layer chat disclosures across entry, composer, response and sharing instead of relying on one disclaimer.
Guidance
Use a concise entry explanation and persistent AI sender identity. Keep an accessible privacy route near the composer, with a short just-in-time explanation before a new kind of collection, upload or memory use. Describe actual processing, access and retention; do not invent privacy assurances. Put task-specific limitations beside the answer and offer checking or correction actions. Keep Scripture attribution separate from generated explanation, and carry meaningful origin labels into saved and shared outputs.
What tends to go wrong
Legal text only in settings; “AI may be wrong” used to excuse unchecked quotations; blanket confidential or anonymous claims; a warning shown only after sending; repeating a large banner on every answer; treating a privacy link or ordinary Send action as blanket consent.
How to test it
Enter chat through a deep link, inspect who responds, open privacy information without losing the draft, add a file and share an answer. Ask what leaves the device, who may access it and how to check a claim. Test long conversations, screen readers and error states.
Sources
- Methods for providing privacy informationLayered and just-in-time approaches reviewed 6 October 2026 · UK guidance; actual processing and regional obligations require assessment. Information alone is not consent.
- Guidelines for Human-AI InteractionOfficial guideline overview reviewed · General guidelines; disclosure alone does not establish accuracy, safety or spiritual benefit.
- DisclosurePattern detail reviewed 5 October 2026 · Industry design guidance; the faith-domain alternatives are original proposals. Source artwork is not reproduced.
- Psalm 23, Berean Standard BiblePassage text taken from API.Bible, 9 October 2026 · Provides the Scripture wording; evaluate the interface through separate usability research.
Related decision: Identify what AI did, where it matters
Label the actor, action and artifact at the moment they could be misunderstood.
Guidance
Use plain-language labels beside the relevant feature, speaker or content: AI assistant, explanation generated with AI, AI-assisted prayer draft, or synthetic voice reading a named edition. Keep a short visible label and optional details about sources, review and limitations. Retain identity through human handoffs and distinguish retrieval of publisher text from generated interpretation.
What tends to go wrong
Only disclosing in onboarding or terms; color-only labels; generic AI tags with no operation; implied clergy credentials; treating a citation or human review badge as proof of correctness; placing an AI tag on verified Scripture because it sits inside an AI response.
How to test it
Ask readers to identify who or what produced each layer, what AI changed and whether a human actually reviewed it. Test screen readers, first entry through deep links, audio-only use and AI-to-human transitions.
Sources
- The Shape of AICatalog and named entries reviewed · The site states CC BY NC SA. This reference uses original text and original wireframes; it does not reproduce the source illustrations.
- DisclosurePattern detail reviewed 5 October 2026 · Industry design guidance; the faith-domain alternatives are original proposals. Source artwork is not reproduced.
- Explainability + TrustChapter reviewed for explanation and trust calibration · General human-AI guidance, not validation of faith apps or a mandate for a particular component.
- Guidelines for Human-AI InteractionOfficial guideline overview reviewed · General guidelines; disclosure alone does not establish accuracy, safety or spiritual benefit.
Sources and their limits
Research-informed design proposal. Original prototypes await user testing.
- Antiqua et nova: note on the relationship between artificial intelligence and human intelligenceParagraphs 59–63 reviewed · Catholic teaching document; other traditions reason differently, though widely in the same direction.
- On artificial intelligence and emerging technologiesResolution reviewed · A resolution of messengers, not binding on churches; does not address personas directly.
- The Media EquationOverview reviewed · Pre-LLM laboratory studies; an encyclopedia summary of the book.
- The inventor of the chatbot tried to warn us about AIArticle reviewed · Historical account and essay.
- Believing anthropomorphism: examining the role of anthropomorphic cues on user trust in large language modelsAbstract rechecked 1 November 2026 · Online experiment with a simulated system; US participants.
- Investigating affective use and emotional well-being on ChatGPTSummary reviewed · Correlational for loneliness; company co-authored; early research.
- Emotional manipulation by AI companionsSummary reviewed; figures checked against the working paper 10 November 2026 · Working paper; general companion apps, not faith apps.
- Evidence of human-level bonds established with a digital conversational agentAbstract reviewed · Retrospective, by the company’s founder and staff; the app has since been retired.
- AI Act Article 50Current official text reviewed · 5 October 2026 · Duties differ by role and content; the whole article must be assessed for the product.
- California SB 243: companion chatbotsEnacted 2025 text, definitions and sections 22602–22604; reviewed 5 October 2026. · Recheck amendments and current codification; not every source assistant is covered.
- New York General Business Law Article 47Sections 1700–1702 reviewed, including definition, protocols and notices; 5 October 2026. · Selected statutory provisions, not a complete New York legal assessment.
- FTC launches inquiry into AI chatbots acting as companionsSummary reviewed · An information-gathering study, not a rule.
- “Text With Jesus” AI chatbot app grows rapidly despite blasphemy criticismNews report reviewed · News report; reception varies.
- AI Jesus in Lucerne confessional baffles church superiorsNews report reviewed · An art project with a self-selected audience and the organizers’ own survey.
- Mental models: expectations of human-like interactionHuman-like interaction section reviewed 6 October 2026 · Industry framework; not a controlled comparison of faith-app typing indicators.
- Generative AI Profile: Human–AI ConfigurationSection 2.7 reviewed 6 October 2026 · Assess individual interface cues through product-specific research.
- Towards understanding sycophancy in language modelsAuthor research abstract reviewed on 5 October 2026; full paper not reviewed · Specific studied models and tasks; not a faith-app intervention study.
- Interaction context often increases sycophancy in LLMsAbstract rechecked 1 November 2026 · General assistants and everyday queries, not faith apps; effects differ by model.
- Artificial Intimacy: Who We Become When We Talk to MachinesRead in full; passages used here checked against the book 10 November 2026 · A sociologist’s interviews and cases, some from press reports; written as a warning, not a weighing of benefits. Studies she cites are not read in the original here.
- The Wizard and I: how transparent teleoperation and self-description (do not) affect children’s robot perceptions and child–robot relationship formationCited second hand, from Turkle’s Artificial Intimacy (2026), 10 November 2026; not yet read in the original · Not yet read in the original; one laboratory study with children.
- Human Raised: Nurturing Connection, Curiosity & Lifelong Learning in the Age of AIRead in full; passages used here checked against the book 10 November 2026 · Written for parents of children up to six; her informal tests are anecdotes, and the studies she cites are not read in the original here.
Faith and interpretation
Presence is at the heart of Christian faith: “the Word became flesh and made His dwelling among us” (John 1:14), and the church gathers in body (Hebrews 10:25). A tool can help someone find words for prayer or prepare for a conversation, but it cannot pray, keep vigil or carry another’s burden (Galatians 6:2). Scripture is severe about speaking in God’s name without being sent (Deuteronomy 18:20; Jeremiah 23:16). The Vatican’s Antiqua et nova calls AI “a tool, not a person”; misrepresenting it as one should always be avoided.







