- Starting price
- Free trial; paid subscription
- Free tier
- No
- Platforms
- Web · Multilingual
- Developer
- Virggil · Alexandre Genko-Starosselsky
- Launched
- 2026
- Updated
- Aug 16, 2026
The verdict
Virggil takes a more explicit values-first approach than a generic chatbot, allowing users to select discernment keys and conversational posture. That may reduce unwanted cultural assumptions, but a Christian label does not make model output wise, doctrinally neutral, private, or spiritually authoritative. The product needs clearer AI identity, pastoral boundaries, safety evidence, source citation, and transparent public pricing.
Try Virggil ↗Opens virggil.com
Virggil is a web-based Christian AI companion designed to filter responses through values, faith, age, and life context selected by the user. It addresses questions about family, work, relationships, culture, belief, and everyday decisions rather than limiting itself to Bible lookup.
Users can choose discernment keys and one of three companion styles: Socrates asks questions to develop the user’s reasoning, Nestor offers more direct counsel, and Plutarch reflects the user’s thoughts back for examination. Profiles include teenager, adult, senior, and school contexts, with support for more than ten languages.
The product offers a free trial, paid subscriptions, and institution-oriented plans for schools, companies, and associations, but a stable public numeric price table was not available in the research output. Its public positioning emphasizes privacy, no data resale or advertiser targeting, and resistance to sycophantic agreement.
We reviewed the current website, about, pricing route, companion descriptions, related articles, and faith.tools editor notes without relying on Virggil for personal counsel. We did not independently audit model providers, safeguards, data handling, theological corpus, institution controls, accuracy, or subscription terms.
✓ The good
- Explicit worldview configuration - users can state which values should shape a response
- Multiple conversational postures - questioning, direct counsel, and reflection suit different purposes
- Age and context profiles - response style can account for a teenager, adult, senior, or school setting
- Multilingual web access - no app-store installation is required
- Anti-sycophancy goal - the product recognizes that agreeable AI can reinforce poor judgment
✗ Watch out
- Spiritual-authority ambiguity - companion language can sound wiser or more personal than software is
- Limited sourcing - personalized answers need traceable Scripture and tradition evidence
- Public price opacity - free trial and paid plans are described without a dependable visible amount
- Safety uncertainty - advice about minors, abuse, health, crisis, and relationships needs documented escalation
- Configuration can reinforce bias - user-selected values may create a comfortable rather than corrective loop
Best for
- Adults experimenting with values-aware reflection while retaining human counsel
- Users comparing how different questioning styles affect their own reasoning
- Schools or organizations conducting a supervised, policy-controlled pilot
- Christians willing to verify every scriptural, doctrinal, factual, and practical claim
Avoid if
- You need prayer, pastoral care, counseling, safeguarding, or crisis support
- A child or vulnerable person may treat the companion as a private trusted authority
- Your tradition requires sources and boundaries the product cannot configure and cite
- You cannot obtain clear provider, retention, deletion, security, and institution terms
What Virggil is
Virggil is a generative AI interface with user-configured values, not the Holy Spirit, prayer, pastor, parent, teacher, counselor, lawyer, doctor, safeguarding officer, doctrinal authority, or conscience. Its classical companion names describe styles, not persons with wisdom or responsibility.
Users and institutions remain responsible for verification, age appropriateness, crisis escalation, consent, privacy, doctrinal boundaries, bias, citations, records, supervision, and taking consequential questions to qualified humans.
Why values-aware AI still needs voices that disagree
Mainstream assistants often hide assumptions behind a neutral tone. Virggil makes worldview preferences more visible and lets users choose whether the system should question, advise, or reflect, which can prompt more deliberate interaction.
A configured worldview can become a closed loop. Invite Scripture in context, church tradition, trusted people, domain experts, and uncomfortable evidence into the decision. The model should surface alternatives and uncertainty rather than baptize the user’s first instinct.
Discernment keys: personalization needs transparent boundaries
Users select lenses such as faith-informed, family-centered, or age-appropriate to shape generated answers. The system aims to respond consistently with declared convictions rather than broad internet consensus.
Every answer should show which keys were applied, what evidence supports it, which assumptions remain, and what a different Christian tradition might conclude. Sensitive personal values should not become advertising or training data.
Socrates, Nestor, and Plutarch: style changes perceived authority
Socrates guides with questions, Nestor speaks more directly, and Plutarch mirrors thoughts for examination. Choosing a mode lets users distinguish exploration from more assertive output.
Direct counsel can be disproportionately persuasive even when the underlying model is equally uncertain. Display an AI identity and confidence boundaries continuously, and restrict advice in crisis, abuse, self-harm, medical, legal, financial, and safeguarding scenarios.
Age and institution profiles: children require more than simpler prose
Teenager, adult, senior, and school profiles can adjust vocabulary and context, while organization plans suggest shared deployment. Multilingual support broadens reach.
For minors, require parental and institutional approval, age assurance, content filtering, staff visibility, retention limits, reporting, and a route to a human. Translation and age adaptation must be tested culturally, not assumed.
Pricing
Free trial
Limited
Trial access is promoted so users can evaluate the companion before subscribing.
Individual subscription
Shown at signup
Paid ongoing access; verify current amount, limits, renewal, and cancellation directly.
Institutions
Custom quote
Plans for schools, companies, and associations require scope and governance review.
Virggil advertises a free trial followed by subscription access. A reliable public numeric price was not exposed during this review.
Institution plans for schools, companies, and associations are also promoted without standard public amounts.
Before paying, verify message limits, model access, languages, profiles, history, exports, deletion, renewal, refunds, taxes, and whether features change by plan.
Institution buyers should include security, child safety, administration, audit, data residency, support, and exit requirements in the quote.
Where Virggil falls behind
Identify the system as AI in every conversation and forbid claims of spiritual agency or divine guidance.
Publish exact pricing, limits, model providers, changes, refund terms, and institution feature matrices.
Cite Scripture and other sources at claim level with translation, context, tradition, and uncertainty.
Provide a public safety specification for minors, self-harm, abuse, sexuality, health, legal, financial, and crisis prompts.
Publish retention, deletion, training, subprocessors, security audits, incident notice, and institutional administration details.
Virggil vs. ChatGPT vs. Waymark
Virggil configures a general companion around user values and conversational modes. ChatGPT is a broad assistant with larger general capabilities. Waymark narrows theological answers to a cited historical corpus, especially the Reformed tradition.
Choose Virggil for worldview-aware reflection, ChatGPT for general tasks, or Waymark for source-grounded theological research. None is a pastor, and each has different privacy and doctrinal risks.
Ask the same difficult question, inspect sources and uncertainty, test whether the assistant challenges assumptions, trigger safety scenarios, delete the conversation, export data, and compare the current contract.
The bottom line
Virggil is right that AI is never free from assumptions and that endless agreement is not wisdom. Its discernment keys and three companion styles create a more deliberate interface than a generic chat window. The unresolved danger is anthropomorphism: a personalized Christian companion can feel spiritually trustworthy even when it is producing probabilistic text. Virggil should make its AI identity, sources, boundaries, models, pricing, and safety behavior unmistakable. Users should keep prompts nonsensitive, verify every claim, seek disagreement, and return important decisions to Scripture, church community, and qualified professionals. Used that way, it may be a reflection tool rather than a replacement relationship.
Alternatives to Virggil
Frequently asked questions
What is Virggil?
It is a Christian-positioned AI companion that adapts answers using selected values, age context, and conversational mode.
What are the three companion modes?
Socrates asks guiding questions, Nestor offers direct advice, and Plutarch reflects the user’s thoughts for examination.
Is Virggil free?
It offers a free trial, followed by a paid subscription; verify the current price and limits at signup.
Can children use Virggil?
A teenager and school context is described, but families and institutions should require strong supervision, privacy, and safety controls.
Does Virggil replace a pastor or prayer?
No. It is generative software and should never claim spiritual agency or replace church, prayer, or qualified care.
Is Virggil private?
The product promotes no data resale or advertiser targeting, but users should verify providers, retention, training, deletion, and security terms.
How should I evaluate an answer?
Check Scripture in context, inspect cited evidence, compare traditions, seek human counsel, and reject confident output that exceeds its sources.
