GTF is not a replacement for Palantir, GIS, or BI tools. It is a different category: a Deterministic Decision Intelligence Engine for physical-world risk — built from the ground up to be explainable, auditable, and deployable without an army of data scientists.
| Capability | Traditional GIS | BI Dashboard | LLM / GenAI | Palantir | GTF Engine |
|---|---|---|---|---|---|
| Explainable Risk Scoring (0-100) | ✗ | Partial | ✗ | Partial | ✅ Native |
| Multi-domain fusion (Air+Sea+Space+Cyber) | ✗ | ✗ | Text only | Complex setup | ✅ Built-in |
| No hallucinations / deterministic | ✅ | ✅ | ✗ (by nature) | ✅ | ✅ By design |
| EU AI Act compliant scoring | ✗ | ✗ | ✗ | Partial | ✅ Explainable |
| Digital Dossier per object | ✗ | ✗ | ✗ | Partial | ✅ Full dossier |
| Event cascade chain visualization | ✗ | ✗ | ✗ | Yes (complex) | ✅ Automatic |
| NIS2 / CER compliance support | ✗ | ✗ | ✗ | ✗ | ✅ Built-in |
| Operational without LLM/GPU | ✅ | ✅ | ✗ | ✅ | ✅ Pure logic |
| API-first (325+ endpoints) | ✗ | ✗ | Partial | ✅ | ✅ |
| Deployable by 1 engineer in 1 day | ✗ | ✗ | ✗ | ✗ (months) | ✅ systemd |
| COST to deploy (SME airports/ports) | $50K+ | $20K+ | $100K+ | $500K–$5M | €5K–50K |
Every risk score is produced by explicit formulas with named factors. No neural network. No statistical black box. Run the same inputs twice — get the same result.
EU AI Act (Article 13) requires explainability for high-risk AI systems. GTF produces a complete audit trail: input → factor → weight → output. Any regulator can verify any score.
One event engine, one risk engine, one dossier system — connected to 9 domains. A wildfire near an airport creates linked dossiers in the Fire, Airport, Infra, and Evac modules simultaneously.
GTF runs entirely on a €20/month ARM server. No GPU. No cloud dependency. No Kubernetes cluster required. This means it can be deployed in air-gapped environments (government, defence, critical infra).