# GEAIRM 0.1 Canonical Example Set — A9

This directory contains the informative machine-readable example set for **GEAIRM-A9**.

The first fifteen examples project the human-readable **M16-P01 through M16-P15** canonical patterns into JSON / JSON-LD structures validated against the A7 structural schema. Two additional examples exercise the A8 governed Profile and Extension schema.

> **Conformance boundary:** structural validity of these examples does not establish complete GEAIRM conformance, AIPS Product qualification, governance approval, semantic-assessment PASS, or production readiness. The human-readable GEAIRM specification remains the normative semantic authority.

## Example inventory

| ID | Example | File | Schema |
|---|---|---|---|
| M16-P01 | Primary Learned-Model Fraud Detection AI Product | `m16-p01-primary-learned-model-fraud.json` | A7 |
| M16-P02 | Productized Model Boundary | `m16-p02-productized-model-boundary.json` | A7 |
| M16-P03 | Primary Symbolic Policy Reasoner | `m16-p03-symbolic-policy-reasoner.json` | A7 |
| M16-P04 | Grounded / Retrieval-Augmented Generation | `m16-p04-grounded-generation.json` | A7 |
| M16-P05 | Agent Uses an AI Product as a Tool | `m16-p05-agent-tool-use.json` | A7 |
| M16-P06 | Human-Reviewed Recommendation and Authorized Action | `m16-p06-human-reviewed-recommendation.json` | A7 |
| M16-P07 | Autonomous Digital Action Under Bounded Authority | `m16-p07-autonomous-digital-action.json` | A7 |
| M16-P08 | Closed-Loop Physical AI | `m16-p08-closed-loop-physical-ai.json` | A7 |
| M16-P09 | Derived / Composite Integrated Transaction Risk AI Product | `m16-p09-derived-composite-risk.json` | A7 |
| M16-P10 | Runtime Orchestration Without New AI Product Identity | `m16-p10-runtime-orchestration.json` | A7 |
| M16-P11 | Multi-Agent Composition Without Collective-Agent Inference | `m16-p11-multi-agent-composition.json` | A7 |
| M16-P12 | AI-Enabled Streaming Service Remains a Service Product | `m16-p12-ai-enabled-service-product.json` | A7 |
| M16-P13 | Evaluation and Assurance Without Risk Collapse | `m16-p13-evaluation-assurance-risk.json` | A7 |
| M16-P14 | Dependency Currentness Change Without Historical Mutation | `m16-p14-dependency-currentness.json` | A7 |
| M16-P15 | AI Primitive Removal and Product-Family Reclassification | `m16-p15-version-reclassification.json` | A7 |
| A8-PROFILE-01 | Regulated Autonomous AI Profile | `profile-regulated-autonomous-ai.json` | A8 |
| A8-EXTENSION-01 | Banking AI Semantic Extension | `extension-banking-ai.json` | A8 |

## Usage

- Validate `m16-p*.json` with `../geairm-0.1.schema.json`.
- Validate `profile-*.json` and `extension-*.json` with `../geairm-0.1.profile.schema.json`.
- `geairm-0.1.examples.json` is the informative A9 index.
- A9 examples deliberately preserve external authority through qualified terms such as `aips:*`, `agent:*`, `bps:*`, `gedrm:*`, `gearm:*`, and `prov:*` rather than reminting external semantics into the GEAIRM namespace.

## Semantic intent

The example set demonstrates learned-model AI, productized-model boundaries, symbolic reasoning, grounded generation, Agent/Tool semantics, human-reviewed decision support, autonomous digital action, Physical AI, Derived/Composite AI Products, runtime orchestration, multi-Agent Composition, AI-enabled non-AI Products, Evaluation/Assurance/Risk separation, currentness change, and version-specific Product reclassification.

The A8 examples additionally demonstrate a governed Profile and a namespace-separated local Extension without weakening core GEAIRM semantics.
