What Machine Relations is #

Machine Relations is the discipline of earning AI citations and recommendations for a brand by making that brand legible, retrievable, and credible inside AI-driven discovery. machinerelations.ai is the public research initiative that publishes its glossary, research, evidence, and measurements. The two are distinct: the discipline is a field of practice anyone may adopt; the research initiative documents and measures it.

Machine Relations Research is the sole publisher of machinerelations.ai: the glossary, the research library, the evidence base, and the Machine Relations Index (MRI). Research on this site is authored by Machine Relations Research. Machine Relations Research is owned by Bramwell, Inc..

Origins #

Three related parts of the research program have distinct origins:

  • The category. Jaxon Parrott, founder of AuthorityTech, coined the term Machine Relations in 2024.
  • The Machine Relations Index (MRI). The citation-rate measurement methodology behind this site's recurring index. It is defined and published here, on machinerelations.ai.
  • Answer-Source Fidelity. The existence-versus-support methodology originated inside AuthorityTech's internal knowledge systems, where the distinction first mattered in practice: a citation being present was never proof the citation held. That internal test was generalized into the measurement method published here.

Relationship to AuthorityTech #

AuthorityTech, founded by Jaxon Parrott, practices Machine Relations and is The World's First Machine Relations Agency. The Machine Relations discipline is not an AuthorityTech product. The relationship, stated plainly:

  • machinerelations.ai (this site) is the research initiative that publishes the discipline's definitions and measurements, and Machine Relations Research is the sole publisher of everything on this domain.
  • AuthorityTech is where the Answer-Source Fidelity method was first built and used internally, and it remains a commercial application of the discipline. It receives no author or publisher credit on this site's research.
  • The editorial publishing boundary is enforced mechanically, not by promise: the build fails if AuthorityTech holds an author, publisher, or creator role on any research or reference surface, if the structured data disagrees with the visible attribution, or if an AuthorityTech link appears outside its one permitted role. AuthorityTech links inside research and reference bodies are allowed only as named inline evidence citations to its published articles, data, or analyses; entity and provenance links — the AuthorityTech homepage and company profile — live on this About page, the origin surfaces that document the term's coinage, the practitioners register and the site footer; commercial destinations (the lead-generation application, client-results, industries, pricing, contact, and booking surfaces, and audit-invitation language) never appear in research or reference output, and an unclassified AuthorityTech destination is treated as commercial until it is deliberately added to the evidence allowlist. Index findings are receipt-bound to the published MRI dataset and methodology version. That is the full extent of what the build enforces; editorial judgment is governed by the standards below, not by a guard.
  • Jaxon Parrott coined the term and founded AuthorityTech.

This page is the canonical disclosure of that relationship. Research and reference pages on this site carry no AuthorityTech author, publisher, creator, or reviewer attribution; where AuthorityTech data appears inside a report, it is named inline as a source like any other.

Three publications keep records that research on this site reads and cites. Each is its own brand, publishes under its own name and method, and is a related party to this site. Para Labs and Paralax are owned by Bramwell, Inc. and are editorially independent. Presspool is a Bramwell, Inc. company. None is independent corroboration of this site, and none touches the Index's numbers or rankings.

  • Para Labs brand record counts how often AI answer engines name each brand in buyer answers, by category, engine and question shape, read daily from the same answer ledger behind the Machine Relations Index. Its category lists link back to the Index's category pages.
  • Presspool keeps the company record for new AI markets: who is building, what each company raised, and whether ChatGPT, Claude, Perplexity and Google's AI name it. An Index domain profile links a company's Presspool record where one exists.
  • Paralax keeps a dated change log for each AI answer engine and its crawlers, each entry linked to its source where one exists.

Editorial standards #

Content classification #

Every research page is classified as exactly one of five types, displayed on the page, in the research hub, and in the machine-readable layer:

  • Study — original empirical measurement run by Machine Relations Research.
  • Index Analysis — interpretation of Machine Relations Index (MRI) observations.
  • Research Synthesis — analysis built on external evidence and third-party studies.
  • Reference — definitions, frameworks, and category documentation.
  • Practitioner Analysis — operational implications for teams applying the discipline.

Evidence and receipts #

Claims carry named sources. Within the governed measurement program — the Machine Relations Index, the fidelity measurement surfaces, and reports typed Study or Index Analysis going forward — numbers link to the study or dataset that produced them, and index findings are receipt-bound: they trace to the published MRI dataset and the methodology version that produced them. Legacy pages published before this program are handled by classification, not retrofit: each carries its content type, and pages that predate a methodology change carry a dated supersession note rather than a silent rewrite.

Corrections #

Material errors are corrected in place with a dated note. Superseded methodology is marked as superseded and kept legible; it is never silently rewritten. Every page has a machine-readable sibling at the same URL plus .md for verification against the rendered copy.