Join our early access program and receive free access

Frequently Asked Questions

Straight answers on how the platform works, what it costs, how we handle security, and how cities get started.

Conflation Labs is an AI platform for local governments and AEC firms that automates both zoning research and permit plan review. Our Site Plan Agent checks uploaded site plans and concepts against local zoning and other development standards, while our Zoning Research Agent answers “What can I build here?” questions using ordinances and authoritative GIS data.

Together they give staff, residents, and applicants fast, map-backed answers on what is allowed on a property and whether a proposed design complies, without digging through PDFs or manuals.

ChatGPT is a general chatbot. It doesn't know your zoning code, your parcels, or your permit rules. Conflation Labs is grounded in your official documents and authoritative GIS data. Every answer cites the exact ordinance section or data source, and we connect language understanding to a live map, so you get a visual answer, not just text.

It handles the questions planning staff field every day, returning both a written explanation and a map instantly:

Zoning & property:
“What is the zoning for APN 010353120?” → “Explain what that zoning allows” → “Show me other parcels with the same zoning within half a mile.”

Business permitting:
“Can I open a tobacco shop at 300 Morningside Drive?” → “Why or why not?” → “Where else in the city can I open one?”

Residential permits & ADUs:
“Do I need a permit for a shade structure at 225 Charlotte Ct?” → “Walk me through the application” → “What are the ADU setback requirements here?”

Economic development:
“Show me available commercial properties downtown” → “What incentives are available for new businesses?”

Answers are grounded in your authoritative documents and cite the specific section, paragraph, or data record they came from, so you can verify in one click. We don't generate from general knowledge. If the document doesn't say it, we don't say it. AI outputs are designed to supplement, not replace, licensed-professional judgment.

Every answer links back to the exact source: the ordinance section, the parcel record, the sheet on the plan. You can verify in one click, and that audit trail is what makes an answer defensible in a council meeting, a FOIA response, or a permit appeal.

For site plans and drawings, we use vision-capable AI models to parse the sheets (title block, scale, north arrow, dimensions, setbacks, parking counts, landscape percentages, and so on) and check them against the applicable code. It's the same way a reviewer reads a sheet, just faster and more consistently.

Yes. We routinely work with municipal codes that are inconsistent, scanned as PDFs, full of conflicting overlays, or scattered across decades of amendments. Onboarding actually surfaces these gaps as a side benefit. Many cities learn where their own code is ambiguous after seeing what the AI flags.

There is a reason we call ourselves 'Conflation' Labs.

Yes. We plug into the data and documents you already have (zoning ordinances, building codes, parcel data, comprehensive plans, GIS layers) without replacing your system of record. We work alongside Tyler, Accela, OpenGov, Esri, and others. We don't ask you to migrate off them.

Yes. Data is encrypted in transit and at rest, hosted in US-based cloud, with role-based access, MFA for staff, audit logging, and SSO available for enterprise. Customer content is never used to train public AI models, and that is contractually enforced with our model providers.

Full details and a security review packet are in our Trust Center.

We work with any U.S. city that gives us access to its documents and GIS data. The platform isn't tied to a region. Today we have live deployments with municipalities from California to Florida, and we're onboarding new cities as design partners.

No. Conflation handles the tedious review (code lookups, map requests, repetitive permit questions) so your staff has time for the strategic work that actually needs human judgment. Cities are short-staffed, not over-staffed.

Everyone in the city ecosystem, with no GIS training required:

City staff: planning, building and permits, code enforcement, and economic development all get instant zoning, permit, and spatial answers.

Elected officials and leadership: quick district and development data, plus accurate answers for constituent inquiries.

Residents and the business community: self-service access to zoning, permits, and property information for residents, business owners, developers, and real estate professionals.

A single senior planner spending about 2 hours a day on routine zoning and permit questions costs roughly $50K a year, and GIS analyst map requests add another ~$30K. Most customers recoup the cost in the first month of saved staff time.

Yes. Our design-partner program is free for the first 2 months, and the post-pilot cost typically fits in existing IT or planning operating budgets without a new procurement cycle. We can also help you scope it into next year's budget request with ROI data from comparable cities.

You're not alone. Most of our cities didn't either when they started. We can share a template policy and the data-handling answers (training carve-outs, FOIA and public-records posture, audit logging) that legal and IT typically need to greenlight a pilot.

Minimal. We work with your existing documents and GIS data, so there's no data project to run first. Design-partner cities commit about 1 to 2 hours per week of staff time during onboarding. We do the heavy lifting.

About 1 to 2 hours per week from a key staff member during the design-partner phase, usually a senior planner or GIS lead. After that, ongoing involvement is minimal, mostly feedback as you use it.

Request a demo through our contact form. We'll set up a walkthrough with real city data and answer any implementation questions for your jurisdiction.