AI manifest

AI manifest

This page is the canonical definition of ai manifest within the canon, corpus, and machine readability layer of interpretive governance.

An AI manifest is a public machine-readable artifact that declares a site’s identity, purpose, canonical entrypoints, governance surfaces, interpretation constraints, and relevant policy or proof files for AI systems.

Short definition

An AI manifest is a public machine-readable artifact that declares a site’s identity, purpose, canonical entrypoints, governance surfaces, interpretation constraints, and relevant policy or proof files for AI systems.

Why it matters

It is a routing and declaration surface. It helps machines discover where identity, canon, definitions, evidence, and exclusions are located before producing a synthesis.

In AI search, retrieval-augmented generation, autonomous browsing, and agentic reading, a corpus is not interpreted only by its visible prose. It is interpreted through routes, files, metadata, exclusions, entity relations, sitemap placement, and internal links. AI manifest names one part of that documentary control layer.

The strategic function is therefore not cosmetic. The concept helps prevent systems from flattening doctrine, service language, proof artifacts, and observations into the same authority level. It also gives search engines a clearer canonical page to associate with the term rather than forcing them to choose between a hub, a category, a blog article, and a machine artifact.

What it is not

It is not a formal web standard, not a guarantee of model obedience, not a crawler directive, and not a replacement for canonical pages or public doctrine.

This distinction matters because machine-readable governance can create false confidence. A structured file, a definition page, or a graph relation should never be treated as proof that external systems comply with the intended reading. It only makes the intended reading more explicit, testable, and auditable.

Common failure modes

• the manifest lists files but not their authority role;
• the manifest becomes stale after definitions are added;
• concepts appear in the manifest without corresponding canonical surfaces;
• models treat manifest declarations as promotional claims instead of routing constraints;

These failures are typical when the human corpus and the machine-first corpus evolve separately. They increase interpretive risk because models can still produce coherent answers while violating the source hierarchy or ignoring exclusions.

Governance implication

An AI manifest should remain synchronized with the sitemap, definitions, entity graph, governance JSON, and public canon. Its value is not that models must obey it; its value is that the intended reading can be reconstructed.

For SERP ownership, the same principle applies: the canonical page should receive descriptive links, appear in the definitions registry, be discoverable from the glossary, and be reinforced by machine-first artifacts without competing against them.

Related canonical definitions

• Machine-first artifacts
• AI governance JSON
• Entity graph
• Canonical source
• Machine-first canon
• Reading conditions
• Proof Of Fidelity

Supporting artifacts and surfaces

• /canon.md
• /site-context.md
• /ai-manifest.json
• /.well-known/ai-governance.json
• /entity-graph.jsonld
• Definitions registry

Corpus role and diagnostic use

In the corpus, AI manifest belongs to the machine-readable layer of interpretive governance. It describes how meaning, routes, entities, exclusions, reading conditions or authority signals can be exposed in a form that machines can parse. The concept is not a promise that external systems will obey, cite, rank, recommend or correct the entity automatically.

The diagnostic use is architectural. It helps determine whether a system can identify what should be read first, which surfaces are canonical, which signals are supporting, which exclusions matter and which routes should not be collapsed together. Without this layer, a site may be readable by humans while still ambiguous to retrieval systems, answer engines or agents.

Failure pattern to detect

The main failure is artifact inflation. A file, graph, manifest or structured signal may be treated as if its existence alone created authority. In this corpus, machine-readable signals must remain tied to canon, source hierarchy, response conditions and evidence. They clarify the reading environment; they do not replace proof of fidelity.

Reading rule

Use this definition with canonical source, canonical surface, machine readability, reading conditions and documentary architecture. The term should help explain how a corpus becomes easier to read without pretending that external systems are bound by it.