Satellite fire detection
Near-real-time active-fire hotspots from NASA FIRMS and Digital Earth Australia, mapped against your registered assets.
Capabilities
The full catalogue, from live features to what's built and activating, in preview, or planned next. Every claim is inspectable — the honesty is the point.
Capabilities
Near-real-time active-fire hotspots from NASA FIRMS and Digital Earth Australia, mapped against your registered assets.
Scattered detections are grouped into tracked fires, so you follow a fire's movement over time rather than loose pixels.
Ten-minute fire-growth tracking from the geostationary satellite feed — a data-partnership item.
Capabilities
Wind-adjusted McArthur (FFDI) spread modelling, with optional WindNinja terrain-aware wind where terrain processing is enabled. Strongest over grassland.
Per-asset estimated time-to-impact that updates as conditions change.
A projection that advances through the hourly forecast and bends with the wind over time — shown in-app as a clearly-labelled preview, not yet accuracy-measured.
An honest 'prove-it' loop that scores modelled spread against observed fire perimeters — the mechanism that moves prediction from preview to measured.
Per-state forest and shrubland fuel models, and probabilistic burn-probability maps.
Capabilities
Register the homes, sites, infrastructure and land that matter, and see modelled risk against each.
Download saved map areas so the map works without connectivity — the tile logic is built; on-device wiring is activating.
Household members mark themselves safe, leaving, or needing help during an incident.
Evacuation routing that accounts for modelled fire and nearby refuges, water, fire stations and hospitals (coverage expands as place data grows).
Capabilities
Alerts as risk to a registered asset changes, delivered to the people responsible — quiet-hours aware, with Emergency warnings prioritised.
Native mobile push works today; web push and an Emergency email fallback are built and activate once delivery keys are configured.
An Emergency warning that isn't confirmed delivered is escalated to an operator queue, so a life-safety alert is never silently lost.
Capabilities
Vetted agency officers get role-based accounts and an operations console — invite, verify and manage staff. Ships dark until an agency onboards.
An incident-command situational view over live fire data, with an operator-selectable spread preview.
Verified officers flag no-go / restricted areas, or suppress a known false-positive — fully audited, and never silencing an Emergency Warning.
Send a warning to the registered asset-owners inside a drawn area or radius — preview the exact audience count, confirm, then send.
Agency admins assign staff authority by jurisdiction (State → region) and capability — least-privilege, and every action audited.
Capabilities
One platform across the web app and native iOS and Android apps.
Organisations with five roles and tenant isolation, so teams see and manage only what they should.
Personal data is held in Australian infrastructure (AWS ap-southeast-2, Sydney).
Predictions come from published physics and rules-based models — the same inputs produce the same, inspectable result, with an audit trail.
What's included
The modules a Fire Path AI deployment includes today.
The offering
A deployable Australian bushfire-intelligence platform — physics-based spread prediction, per-asset alerting, and an emergency-services control-plane, across web and native apps.
Cloud-hosted in Australia (AWS Sydney) with a cost-scalable footprint. The resident-facing platform works now; the emergency-services suite activates with an agency partner.
Deterministic, auditable predictions and Australian data residency — designed for government and enterprise review, with an honest, published maturity gradient.
A clear line between what is live, what is built and activating, and what is being measured next — no accuracy or certification claims that aren't yet earned.
Fire Path AI is in active development and is available to partner with, deploy with, or acquire and take forward.
Method & maturity. Spread modelling is physics-based (McArthur), strongest for grassland; forest and shrubland fuel-type models are being refined, and predictive accuracy is being measured rather than claimed. Fire Path AI is designed to support official decision-making, not replace it.
Request a briefing
We can walk through what's live, what's activating, and the roadmap with your team.