Straight answers
about what GUS actually does.

GUS, our Geospatial Understanding System, is the AI platform behind every SilaCities engagement. A brain for cities. You work in nine workspaces (Chat, Atlas, Map View, GUS Consult, Data Management, GUS Fieldnotes, GUS SQUID, GUS DAT, GUS Plan Assessment); inside each, agents call on 90+ tools, all drawing on the same enriched dataset and the same neighbourhood grid, so every answer agrees. Ask a question in any language. Get a cited answer.

GUS works in any language your team uses. Ask in plain English, French, Spanish, Arabic, or whatever fits your deployment. Arabic is native to GUS, with dialect-aware sentiment across 20+ dialects covering the Gulf, Levant and North Africa. Right-to-left layout is built in across every tool from day one.

The Sila Score. Our own 0 to 10 metric, running critical to excellent across 12 canonical topics and 68 subtopics, computed by our own NLP models for sentiment and emotion. One legend across every view in Atlas. The only 0 to 100 score in the product is GUS SQUID’s overall suitability score.

Seven lenses on your city: synthesis, satisfaction, community, urban pulse, urban intelligence, spatial economics, vulnerability — with an AI assistant alongside. You turn lenses on and off per deployment, so your team sees only what their role needs.

Goal-first, not form-first. State what you want to understand in plain English — for example, "an economic baseline of the city" or "housing supply risks across districts" — and GUS Consult finds the relevant documents in your library, asks a few short questions, one at a time, then shows a plan with the documents it will cite, the data it will use, and the sections it will write. Nothing generates until you approve the plan. You can pick a perspective (economist, demographer, transport planner, urban designer, environment, community, service delivery), refine in plain language, and export the finished report to PDF, Word, or PowerPoint.

Yes. PDF, Word and PowerPoint. Every export comes through a short-lived, signed download link.

No. When the data cannot support an answer honestly, GUS says so. "Not enough signal here" is a valid answer. When it does answer, every claim carries an inline citation you can click to see the exact source and date.

Yes. Every chat saves automatically. You can reopen it, share it with a colleague, or export it later. Conversations are private to your deployment. No public share URLs today.

Yes. GUS SQUID is live. The 0 to 100 suitability score, four supporting signals (population, activity, competitors, satisfaction), catchment drawing, market-gap analysis and PDF + PowerPoint export are all shipping. Save any analysis with its weights, reopen it later, share it with the team.

GUS Fieldnotes is cloud-backed. Every edit saves automatically and syncs across your team’s devices, so you pick up where you left off, wherever you are. Short offline stretches in the browser are fine. The save flushes when you reconnect.

No. Map View is the dedicated, full-featured GIS workspace for the days you want every lever. Atlas, GUS SQUID, GUS Fieldnotes, GUS Plan Assessment and the maps embedded in Chat are curated views over the same city data.

Spatial files: Shapefile, GeoJSON, CSV with geometry. Documents: PDF, Word, Excel, JSON, imagery and TIFF. Classify each upload by theme (planning, demographic, satisfaction, infrastructure, economic, environmental, transport, or other) so the right team finds it.

Yes. Every question you can ask through the UI is also available over a clean REST API. Scoped to your city’s deployment, rate-limited and documented. Contact us for the full reference.

About eight weeks from kickoff to first decision. Week 1: scope and ingest. Week 3: your data is indexed and baseline maps come online. Week 6: tools configured per role and report templates set up. Week 8: your team ships its first decision with GUS evidence. Single-district engagements go faster. Multi-city rollouts take longer.

No. We demo on a city we have already deployed. Once you are committed, we bring your data in while we set up your deployment. First questions on your own data are typically answerable by week 3 or 4.

Yes. We deploy in our cloud, in your cloud environment, or fully on-premises. On-premises adds roughly four weeks to the timeline.