AnatomyStrong architectural inferencev1.22.1

In plain English

This page explains where an AI behavior can live. It may be in a model, but it may also be in a prompt, memory record, adapter, dataset, tool setting, evaluator rule, or human workflow.

  • Why this matters: AI risk can come from the whole arrangement, not one obvious model.
  • What to look for: data, memory, routes, adapters, tools, evaluators, updates, and rollback paths.
  • Technical version below: the expert terminology remains available and is linked through the glossary.

Browser-Native Evolution Laboratory

Evidence levelStrong architectural inferenceTechnical label: Architectural inference

A browser-native evolution laboratory is a visibility surface. It can show model topology, candidate The parent-child history of models, adapters, datasets, or releases. Open glossary definition, simulation state, and review metadata without becoming a hosted AI product.

Core parts

PartRole
Visual canvasShows topology, population, routes, or lineage.
Local stateHolds worksheet inputs and review notes in the browser session.
Source report pointersConnects every concept to a preserved report summary.
Genome recordDescribes the variant being reviewed.
Fitness vectorDescribes why the variant was retained, rejected, archived, or promoted.
Boundary controlsKeep external data, tool execution, and irreversible actions out of the page.

Why this belongs in anatomy

The current site already treats models, adapters, prompts, memory, tools, evaluators, and human workflows as carriers. The browser-native lab adds another carrier: the interface that makes those carriers legible. A good interface can reduce confusion; a bad one can hide composition changes behind a single impressive score.

Guardrails for Cognivirus

v1.22.0 risk-side boundary

Evidence levelStrong architectural inferenceTechnical label: Strong architectural inference

This page preserves possibility-side source material. The A behavior pattern that can survive, move, or reappear across a changing AI system. Open glossary definition use is narrower: translate controlled-evolution vocabulary into risk review. For that translation, use ModelBreeder Risk Side, ModelBreeder Risk Translation, and Model Breeding Risk Boundaries.