THE PLATFORM
A whole investigation, connected to one object.
Columbus AI System uses spatial objects as a common reference for screening, modeling and evidence research.
OBJECT
One place. A shared research context.
A map object connects data, assumptions, documents and results. The investigation moves beyond isolated layers and files.
- Location and boundaries
- Conditions and assumptions
- Documents and sources
Organize around the research object, not just the file directory.
COMPUTATION
Explicit computation for explicit questions.
Spatial algorithms screen conditions and calculate geometry. Model calls process documents. Each tool has a distinct task, with results connected to the same object.
- GIS spatial analysis
- Scenario comparison
- Model calls and extraction
Spatial computation and model interpretation are distinct responsibilities.
EVIDENCE
Keep a path back to the evidence.
Interpret candidates, scenarios and extracted records against their sources. Researchers examine support, counter-evidence and gaps before choosing the next action.
- Check original material
- Compare competing evidence
- Choose the next investigation
Traceability enables review; model output is not automatically a fact.
SYSTEM LOGIC
Clear roles. Accountable judgments.
Map workbench
2D maps and 3D scenes
Organize objects, documents and resultsSpatial computation
GIS and asynchronous geoprocessing
Execute explicit spatial algorithmsModel calls
Backend model APIs and structured output
Extract and organize document informationResearcher review
Source checks and assumption review
Interpret results and choose an actionAI-assisted software development and runtime model calls are separate activities. Cross-domain work reuses the organization of research, not its conclusions.