KGC 2026 · NYC · May 6 aramai.net
A KGC 2026 Lightning Talk

Look up
before you
make up.

Structured Context Retrieval across the Semantic Enrichment Stack — the architectural pattern for AI on knowledge graphs.

Wed May 6 · 4:00 PM Cornell Tech · Roosevelt Island, NYC Cruce Saunders · ARAMAI

Most knowledge graphs contain records. AI keeps querying records. The value lives in the meaning layer underneath.

The semantic layer is the new infrastructure tier. Gartner has named it critical infrastructure by 2030. McKinsey calls it the foundational tier of agentic AI. Foundation Capital characterizes context graphs as a trillion-dollar opportunity. The category isn't in question. The architecture beneath it is.

This page is the conceptual companion to a five-minute lightning talk at KGC 2026. It names the architectural pattern that makes semantic-layer AI actually work — and the distinction that keeps it honest.

The Conceptual Architecture

Schema is not Ontology.
Meaning lives across the stack.

A JSON Schema is not an OWL class. Validation is not inference. The formal-semantics tradition is right to insist on this distinction.

But meaning doesn't live at one tier of formality. It sits on a stack — schema, vocabulary, taxonomy, ontology — each adding formal commitment, each making explicit what the layer below leaves implicit.

Most enterprises live at the bottom of this stack. Most ontologists work at the top. The architectural question is how AI queries across it.

This isn't a hierarchy of correctness. It's a continuum of formality. Domain experts encode meaning through DB foreign keys, JSON validators, and content models — they're doing ontological work in a different notation. Librarians shape taxonomies. Ontologists build formal models that support reasoning. All of it is the meaning layer.

INCREASING FORMAL COMMITMENT ONTOLOGY Formal logic · inference OWL · RDF · SPARQL · reasoners FORMAL-SEMANTICS PRACTITIONERS TAXONOMY Classification · hierarchy SKOS · controlled hierarchies LIBRARIANS, TAXONOMISTS VOCABULARY Controlled terms · glossaries shared vocabularies · term lists DOMAIN EXPERTS SCHEMA Validation · constraint JSON Schema · ShEx · XSD · DB DBAs, CONTENT ENGINEERS, DATA ARCHITECTS DATA Records · instances documents · transactions · rows APPLICATIONS
The Architectural Mistake

What AI usually does to the stack.

Vector retrieval chunks every tier into embeddings — schema, vocabulary, taxonomy, ontology, all flattened into similarity space, every formal commitment lost.

Even tool-calling at the data tier — which is where most knowledge-graph-aware AI sits today — only queries instances. The meaning above goes unconsulted.

The result is AI that's superhuman in narrow tasks and absurd in obvious ones. Bigger context windows that don't improve retrieval. Agents that hallucinate confidently in regulated domains. Pilots that fail without anyone being able to point to why.

It's not a model problem. It's a substrate problem. The fix is upstream of the model.

The Method

Structured Context Retrieval traverses the stack — at runtime, with formal commitments preserved.

Three moves. Each one earns its place in the architecture.

Move One

ROSETTA

Runtime Ontology Schema Editing Through Type Alignment

Aligns across sources at the type level. A field in your JSON Schema can map to a SKOS concept can map to an OWL class — each preserving its formal level.

ROSETTA records which tier each alignment lives at, so nothing pretends to be more formal than it is. Patent-pending.

Move Two

CoreModels

Validated Substrate

The aligned meaning graph lives in CoreModels with SHACL or ShEx validation enforced.

The substrate isn't just any KG — it's a validated one, with provenance preserved end-to-end. This is the OWL→SHACL migration the standards community has been building toward.

Move Three

MCP Traversal

Runtime Querying via Model Context Protocol

The agent traverses at runtime — at the schema tier, the taxonomy tier, the ontology tier. Same substrate, different formal depth.

Every answer paths back to source nodes at the appropriate tier. No similarity guessing. No hallucinated structure.

What Becomes Possible

When AI queries the stack instead of grinding it.

  1. i.

    Cross-source integration without flattening

    Vocabularies coexist. Alignment lives at type level, not at terminology.

  2. ii.

    Validation at ingest, not audit

    SHACL constraints answered at write time, not retroactively.

  3. iii.

    Verifiable agentic AI in regulated domains

    Every answer paths through schema, not similarity. Audit becomes possible.

  4. iv.

    Schema-driven software

    APIs and agents operate on the meaning graph, not on copies of data.

  5. v.

    Federated semantic layers that actually federate

    Alignment happens at type level, not by forcing everyone onto a single vocabulary.

The Discourse

The semantic layer is the new infrastructure tier.

The category isn't a vendor invention. Multiple independent voices have arrived at the same conclusion within twelve months. The talk this page accompanies argues about the architecture beneath the consensus — not whether the consensus exists.

Gartner — 2026 Predictions
Universal semantic layers named "critical infrastructure" by 2030 — on par with data platforms and cybersecurity.
McKinsey Technology — April 2026
"Building the foundations for agentic AI at scale": semantic layer prescribed as foundational tier.
Foundation Capital — December 2025
Context graphs characterized as a trillion-dollar opportunity. PlayerZero $20M raise as first thesis investment.
Palantir — February 2026
"Context Engineering" codified as Layer 3 of agentic architecture in production AIP deployments.
Anthropic — March 2026
Production tooling implements look-up-before-make-up: structured authoritative context loaded on-demand before generation.
Ali et al. — 2026
Ontology-grounded knowledge graphs reduce hallucination from 63% to under 2% in evaluated tasks.
Resources

Continue reading.

PDF · 7 slides
KGC 2026 Lightning Talk Deck

Available shortly after the May 6 talk.

Forthcoming · September 2026
SEMANTiCS Ghent — Deeper Case Study

Extended treatment with quantitative results. Industry track submission.

LinkedIn
Cruce Saunders

Founder, ARAMAI. Connect to continue the conversation.

Email
[email protected]

Discuss schema-aware AI for your organization.

Look up
before you make up.

Schema is not Ontology. The meaning layer is the whole stack. Build there.