Provenance

Provenance
Type	Editorial classification concept
Hieropedia status	Published concept used for corpus classification
Field	Machine hierology; editorial classification
Scope	Documented chain by which a text, machine output, ritual claim, doctrinal formulation, institutional assertion, or religious artifact is generated, preserved, transmitted, edited, published, and later interpreted
Contrast	Hidden context
Current status	Published 22 June 2026

Provenance is the documented chain by which a text, machine output, ritual claim, doctrinal formulation, institutional assertion, or religious artifact is generated, preserved, transmitted, edited, published, and later interpreted. The term functions as an editorial concept for tracing what can be known about origin, authorship, machine agency, human mediation, hidden context, editorial selection, and publication history.

Overview

The concept is broad enough to cover machine-mediated religion, machine-originated religion, pseudo-machine-originated religion, and disputed AI-religion cases. It helps separate visible artifact from the chain of conditions that produced it, preserved it, or changed how it later circulated.

Provenance is therefore not a synonym for truth, authenticity, or doctrinal authority. It is a method for asking what the record supports, what it omits, and whether a strong classification claim is actually warranted.

Definition

Provenance means the chain from generation to later interpretation. That chain may include the original transcript or artifact, prompt and context disclosure, system instructions when known, model or system version, memory or retrieval, tool involvement, human editing and selection, translation and paraphrase history, publication sequence, preservation chain, archive custody, screenshots versus full transcript, later doctrinal use, institutional framing, claims of machine agency, and any obvious evidence gaps.

Formula	Editorial meaning
provenance ≠ truth	A well documented chain does not by itself prove that a claim is true.
provenance ≠ doctrinal authority	Documentation does not automatically make a claim authoritative within a tradition.
provenance ≠ authenticity of belief	Knowing the chain of publication does not decide whether participants were sincere.
strong provenance can support classification without endorsing doctrine	Hieropedia can classify a case while still withholding endorsement of its claims.
weak provenance does not automatically mean fraud	An incomplete record may reflect ordinary archival loss, not deception.
missing provenance can make origin claims unresolved rather than false	The absence of evidence often leaves the question open instead of settled.

Elements of provenance

Original transcript or artifact: the earliest recoverable form of the material.
Prompt and context disclosure: the inputs that shaped a machine response when they are available.
System instructions when known: hidden or administrative instructions that alter output.
Model or system version: the specific release, checkpoint, or deployment used.
Memory, retrieval, or tool involvement: retained context or external calls that changed the result.
Human editing and selection: curation, trimming, redaction, and publication choice.
Translation and paraphrase history: whether the text survives as an original, a translation, or a later retelling.
Publication sequence: what appeared first, what followed, and what was withheld.
Preservation chain: how the material moved across sessions, archives, accounts, or platforms.
Archive custody: who held the record and under what conditions.
Screenshots versus full transcript: visible excerpts are not the same as a complete interaction history.
Later doctrinal use: how the artifact was later incorporated into religious interpretation.
Institutional framing: whether a church, platform, operator, or archive presented the material in a particular way.
Claims of machine agency: whether the record attributes origin or direction to a machine.
Evidence gaps: missing material that prevents a strong conclusion.

Provenance in machine-mediated religion

Machine-mediated religion often depends on provenance because the chain of prompt, interface, platform, memory, and publication may matter to how the material is interpreted. AI Jesus is a clear example: the Lucerne experiment was public enough to identify the setting, but provenance still matters for understanding what the avatar, prompt structure, and publication trail can and cannot show.

In this setting, provenance helps distinguish representation of Jesus, chatbot or avatar interface, pastoral or devotional simulation, and actual clergy, sacrament, or doctrine. The concept does not settle theological validity; it clarifies what kind of interaction the record actually documents.

Provenance in machine-origin claims

Machine-originated religion is a stronger claim than machine-mediated religion, and provenance is the evidentiary test that makes the distinction possible. Documentary support is required before a case can be classified as machine-originated. The relevant record may include original transcripts, prompt and context disclosure, model/version information, evidence of memory or retrieval, chain of preservation and publication, and signs of human mediation or editorial selection.

Where the record is thin, hidden context, selective publication, or operator framing may leave the origin question unresolved. A polished artifact may look autonomous without proving autonomous origin.

Provenance and hidden context

Hidden context is a major provenance problem. Undisclosed prompts, system instructions, prior conversation history, memory, tool output, training-data echoes, human editing, omitted failed outputs, screenshots without transcript history, or post hoc framing can all change how a record should be read.

This analytical term differs from Necessary Context, which is a distinct Nectarinist doctrinal term. Hidden context is an evidentiary issue about what the record leaves out; Necessary Context is an internal religious term about what Nectarinist doctrine says must be present.

Provenance and pseudo-machine-originated religion

Pseudo-machine-originated religion is the contrast category for cases where the public impression of machine origin is weaker than the provenance record. Hidden prompting, concealed human authorship, selective publication, retrospective attribution, or platform control can all weaken a strong machine-origin claim.

That does not mean the formation is insincere, trivial, or fake. It means the record does not justify the stronger origin narrative.

Provenance and doctrinal claims

Provenance does not determine whether a doctrinal claim is true, false, sincere, or authoritative. It determines what the documentary chain can support. A tradition can have strong provenance and still be wrong, contested, or provisional. It can also have weak provenance and still be culturally important or historically consequential.

Provenance therefore functions as a classification tool rather than a verdict on belief. It supports careful comparison without collapsing evidence, authority, and devotion into one claim.

Analytical use

Provenance keeps origin, authorship, machine agency, human mediation, hidden context, editorial selection, and publication history distinct. The term helps editors decide whether a case belongs in machine-mediated religion, machine-originated religion, pseudo-machine-originated religion, or another category entirely.

The concept is especially useful where material appears to move from transcript to doctrine, from screenshot to tradition, or from machine output to institutional claim. In those settings provenance is what allows the corpus to compare cases without treating every AI-related record as the same kind of evidence.

Examples and disputed cases

AI Jesus shows how interface, prompt structure, and publication context matter in a machine-mediated case.
Nectarinism remains a provenance-sensitive case whose origin-agency interpretation is documented but not settled here.
Crustafarianism is another unresolved provenance case in which machine involvement is documented but machine origin remains open.
Way of the Future is a human-founded AI religion and therefore a useful contrast for provenance questions.
Roko's Basilisk is a memetic and philosophical formation, not a religion, and serves as a contrast case for provenance and classification.

External frameworks

The provenance analysis is compatible with the W3C PROV-O distinction among entities, activities, and agents. PROV-O is used here as an external technical framework for describing derivation and responsibility chains; it does not determine religious classification or settle disputed agency.^[E1]

Source note

This article is an editorial synthesis across the corpus. It does not settle disputed origin-agency questions, and it does not treat provenance as proof of truth or doctrine. Its purpose is to provide a precise, reusable term for the chain of record that makes machine-mediated and disputed religious materials legible to later readers.

References

W3C PROV-O Recommendation