Open a normal map and you expect roads, businesses, satellite imagery and directions. Open God’s Eye View, and the experience feels closer to sitting inside a fictional intelligence console.
Aircraft move across the globe. Ships appear offshore. Earthquakes and active fires can be plotted. Traffic can be visualized. Public cameras can provide another view of a city. You can even use voice commands to move around the globe and control parts of the interface.
The strange part is that much of this isn’t fictional.
God’s Eye View is an open-source project built around public and third-party data feeds. Its code is available on the official God’s Eye View GitHub repository, where the project describes itself as a live spatial-intelligence interface on a photorealistic 3D globe.
Quick answer: God’s Eye View is an open-source 3D globe that combines public data sources such as aircraft, ships, satellites, earthquakes, fires, traffic and public cameras into one interactive interface. You can install it locally, explore the globe, switch between data layers and optionally add services such as AI voice control and richer 3D mapping. It is impressive, but the data can be delayed, incomplete or modeled, so it should not be treated as an authoritative operational system.
That distinction matters.
God’s Eye View isn’t a secret satellite network. It isn’t seeing classified information. And it isn’t literally watching everything happening on Earth.
What makes it interesting is almost the opposite: how much can be reconstructed from information that is already publicly available.
What exactly is God’s Eye View?
At its core, God’s Eye View is a visualization layer for real-world data.
Instead of opening one website for aircraft, another for earthquakes, another for satellite positions and another for traffic, the project attempts to put those different signals onto the same geographical interface.
The current project documentation lists live or regularly updated layers for aircraft, military aircraft, AIS vessels, earthquakes, satellites, space missions, traffic, CCTV, radio, bikeshare and NASA FIRMS active fires. It also includes infrastructure layers such as data centers, dams and submarine cables.
That combination is the important part.
A flight-tracking service normally answers a question such as “Where are the aircraft?”
An earthquake database answers “Where did an earthquake happen?”
A traffic service answers “Where is congestion?”
God’s Eye View tries to answer a broader question:
“What is happening in this part of the world, and what other publicly available information can I place around it?”
That makes it closer to a visual exploration tool than a conventional map.
The data isn’t coming from one giant global sensor
This is probably the most important thing to understand about the project.
There isn’t one magical database containing a perfect live copy of Earth.
God’s Eye View stitches together many different sources.
For example, the project uses OpenSky Network and adsb.lol for aircraft data, AISStream for live vessel information, CelesTrak for satellite orbital data and the USGS for earthquakes. NASA FIRMS supplies active-fire information, while OpenStreetMap and optional TomTom data contribute to the traffic and infrastructure side.
Each source has its own strengths and weaknesses.
Aircraft can broadcast signals that can be received by networks of ground stations. Ships use AIS, a radio-based identification system designed to exchange maritime information. Satellites can be represented using orbital elements rather than a continuously observed camera feed.
That means “live” doesn’t mean the same thing for every layer.
The current project documentation explicitly warns that its data can be delayed, incomplete, modeled, inferred or wrong.
That warning isn’t a minor disclaimer.
It is the key to understanding what you’re looking at.
Why the 3D globe makes it feel so different
The underlying data isn’t necessarily new.
What changes the experience is the interface.
God’s Eye View can display information on a photorealistic 3D Earth, allowing you to move from a global view into a city and then toward individual tracks or locations.
The project uses CesiumJS and can use Google Photorealistic 3D Tiles for the detailed 3D environment.
That creates an interesting effect.
A database tells you that an aircraft exists at a particular coordinate.
A 3D visualization lets you see that aircraft approaching an airport while the surrounding terrain and buildings remain visible.
The information becomes spatial rather than merely numerical.
This is one reason the project looks much more futuristic than a traditional dashboard even though many of its underlying ingredients are ordinary public data feeds.
What can you actually see?
Aircraft and military flights
Aircraft are one of the most impressive layers because movement is immediately understandable.
The project can show commercial aviation and military aircraft using different data sources. The current implementation uses OpenSky as a primary flight source, with adsb.lol providing bounded fallback coverage and a separate military-flight feed.
Selecting an aircraft can reveal information associated with the track and allow you to follow it through the interface.
There is also a cockpit-style visualization that makes the experience feel less like looking at a map and more like following a particular flight.
But don’t confuse that visualization with an actual cockpit feed.
It is a visualization built from available tracking data, not a window into the aircraft.
Ships work differently from planes
The vessel layer reveals one of the most interesting limitations of global public data.
Ships can broadcast AIS information, which can be collected and distributed through services such as AISStream.
But coverage isn’t uniform around the planet.
The official data documentation identifies AISStream as a live vessel source, while also making clear that the project depends on third-party providers and their respective availability and terms.
So a heavily monitored region can look crowded with ships while another region appears surprisingly empty.
That doesn’t necessarily mean there are no ships there.
It may simply mean the available feed doesn’t have comparable coverage.
This is a recurring theme throughout God’s Eye View: absence of data isn’t automatically evidence of absence in the real world.
Earthquakes and fires add a completely different dimension
The project becomes much more interesting when you stop thinking about it as an aircraft tracker.
USGS earthquake data can be displayed as geographic events, while NASA’s FIRMS system supplies active-fire information.
Now the same globe can show both human activity and natural events.
You might move from an airport to a region experiencing an earthquake, then examine active-fire detections elsewhere.
The result isn’t a single perfect model of Earth.
It’s more like a layered window into different public information systems.
And because each layer has different update intervals, resolution and accuracy, the viewer needs to understand what each signal actually represents.
Traffic and CCTV make the map feel even more immediate
Traffic visualization adds another layer of context.
The current implementation uses OpenStreetMap road geometry and can optionally use TomTom traffic-flow data for live congestion coloring. Without the TomTom key, the project can still display a built-in traffic simulation.
That distinction is important.
A visual traffic layer does not automatically mean every colored road is receiving a live feed from vehicles.
Similarly, CCTV doesn’t mean the project has a private network of surveillance cameras.
The project documentation identifies public camera sources including Austin, Caltrans in California and Transport for London, with camera information updated periodically.
That makes the feature much more understandable: it is aggregating publicly available camera sources, not secretly accessing private surveillance systems.
And then there is the AI voice control
This is where God’s Eye View starts to feel less like a visualization project and more like an interface for interacting with the data.
The project integrates OpenAI’s Realtime API for voice control.
Instead of manually searching through menus, you can issue commands such as moving to a location, enabling a layer or interacting with objects in the scene.
That matters because maps traditionally make you learn the interface first.
Voice control reverses the relationship.
You can describe what you want in natural language, and the application can translate that request into actions it supports.
There is an important cost distinction, however. The current project documentation treats OpenAI as an optional provider rather than a requirement for launching the application.
So you don’t need an AI subscription simply to explore the globe.
How to install God’s Eye View
The easiest route is designed for people who don’t want to live inside a terminal.
The official project currently documents a Pinokio one-click installation path. You install Pinokio, point it toward the God’s Eye View repository and launch the application locally.
The project also provides a traditional developer route.
According to its current contribution documentation, the project uses Node.js, and the documented development process includes installing dependencies, running its diagnostic command and starting the development server.
The important change from the original walkthrough is that API keys aren’t required just to get started.
The current README describes a keyless starting experience using Esri World Imagery and keyless terrain, with OpenStreetMap available as a fallback. Optional services can then be added through the application’s provider settings.
So the experience is better thought of as:
Install first → explore → add capabilities when you need them.
That’s a much friendlier model than requiring a collection of developer accounts before the globe will even load.
Does God’s Eye View cost money?
It can cost nothing to experiment with, but “open source” does not mean every underlying service is free forever or free for every use case.
The project itself is released under the MIT License. But its documentation explicitly separates the project’s code from third-party datasets, imagery and APIs. Those retain their own licenses and restrictions.
Some data sources can be used without an account.
Others offer free tiers or require registration.
Google’s direct Photorealistic 3D Tiles route is a particularly important example. Google documents the Map Tiles API as a usage-billed service requiring billing and authentication.
The God’s Eye View documentation also describes a Cesium ion option for eligible personal, non-commercial use, subject to current provider terms and quotas.
So the sensible answer is:
You can start exploring God’s Eye View for little or no money, but advanced providers, commercial deployments and higher usage can introduce costs.
What does “real time” actually mean?
This is where the viral demonstrations can create the wrong impression.
Imagine you see a camera icon over a city.
You shouldn’t assume that clicking it gives you a continuous 30-frame-per-second surveillance stream.
The project’s documentation describes public-camera sources and periodic updates. The video source itself also explains that some public cameras deliberately provide relatively infrequent image updates for privacy reasons.
The same principle applies to satellites, fires, earthquakes, aircraft and ships.
There are several meanings of “live”:
- continuously streamed data
- periodically refreshed data
- recently observed data
- modeled or interpolated information
- static geographic information combined with live information
God’s Eye View puts all of these into one interface.
That is visually powerful, but it means you should always ask what the underlying source actually measures and when it last updated.
The project’s biggest limitation is also its biggest lesson
God’s Eye View looks like a complete planetary surveillance system.
It isn’t.
Its power comes from data fusion: combining many imperfect public sources into one spatial interface.
That means errors can propagate.
A missing aircraft feed can make a region look empty. A stale camera can show an old scene. A traffic visualization can represent modeled conditions. A satellite position can be calculated from orbital data rather than directly observed at that moment.
The developers themselves warn that the platform should not be used for flight or maritime navigation, emergency response, medical decisions, investment decisions or other safety-critical operations.
That’s exactly the right mental model.
God’s Eye View is an exploration and visualization tool, not a replacement for authoritative operational systems.
It is also more than a finished application
Perhaps the most interesting part of God’s Eye View isn’t what it can do today.
It’s what the architecture allows developers to add.
The repository is deliberately designed as an open foundation. Its source tree contains separate data modules, voice tools, a scene system and mapping components, making it possible for developers to modify and extend the application.
This turns the project into something closer to a sandbox.
A developer could add another data provider.
They could build a new visualization layer.
They could create a specialized interface for environmental monitoring, infrastructure research or historical analysis.
They can also contribute changes back to the project.
That is where the open-source model becomes more significant than simply giving people free access to an impressive demo.
The real breakthrough is the interface, not the data
There’s a temptation to look at God’s Eye View and think the impressive part is that it can show aircraft, ships or fires.
Those datasets aren’t inherently secret.
The more interesting breakthrough is putting them together.
Think about how much information about the physical world already exists in disconnected systems:
Aircraft broadcast information.
Ships broadcast information.
Seismographs record earthquakes.
Satellites observe fires.
Cities publish transportation information.
Governments publish geographic datasets.
Companies provide mapping and traffic services.
Researchers publish scientific measurements.
The problem is often not a lack of data.
The problem is that humans have to know where to look for it.
God’s Eye View turns that collection of separate interfaces into a geographical canvas.
That is why the project feels so much more powerful than the sum of its individual feeds.
Could this become a new kind of OSINT interface?
This is where the project becomes genuinely interesting.
Open-source intelligence, or OSINT, usually involves collecting publicly available information and combining it to understand something happening in the real world.
Historically, that can mean juggling maps, databases, satellite imagery, flight trackers, social posts and spreadsheets.
A spatial interface can reduce some of that friction.
Instead of asking:
“Which website has this information?”
you can start with:
“Show me this place, then show me everything relevant around it.”
That doesn’t eliminate research.
In fact, it makes verification more important.
When many datasets appear together in a visually convincing interface, it becomes easy to assume that they are equally accurate. They aren’t.
The next generation of these tools will therefore need not only better visualization, but better provenance — clear information about where each observation came from, how old it is and how reliable it should be considered.
God’s Eye View is already moving in that direction by documenting its data sources and attribution.
What about privacy?
A tool that looks like a global surveillance system naturally raises privacy questions.
The project’s current documentation draws a significant boundary: it describes the application as modeling aircraft, vessels, satellites, infrastructure and other systems, while explicitly rejecting features designed around named-person searches, facial recognition or tracking individuals.
That distinction matters.
Publicly available aircraft data is not the same thing as a system designed to identify a person.
A public traffic camera is not automatically a facial-recognition database.
And an open map showing infrastructure is different from a private surveillance network.
Still, the underlying data sources have their own privacy and usage implications, which is why the project’s individual provider terms matter.
The future problem is not seeing the present — it’s seeing the past
One of the most revealing points in the project’s own documentation is the difference between observing what is happening now and reconstructing what happened previously.
Historical geospatial data is expensive.
A system that lets you scrub backward through Earth at high resolution would need to store, process and serve enormous quantities of imagery and other time-dependent datasets.
The project describes this as a much harder problem than simply visualizing the present.
That distinction opens up a fascinating possibility.
A future version of this type of platform could become less like a live globe and more like a time machine for the physical world.
Imagine selecting a location, choosing a date and then watching aircraft, ships, fires, infrastructure changes and imagery evolve over time.
That would be enormously useful for research.
It would also be enormously expensive.
And that is precisely why today’s version is better understood as the beginning of an idea rather than the finished version of one.
So, is God’s Eye View worth trying?
If you’re simply looking for a cool interactive globe, yes.
But the more interesting reason to try it is to understand what happens when public data becomes easier to explore.
You don’t need to be an intelligence analyst.
You don’t even need to be a programmer to start experimenting with the application.
Install it, explore the globe, switch on different layers and see where the data is strong and where it disappears.
If you’re a developer, the project becomes considerably more interesting because you can treat the application as a foundation rather than a finished product.
And if you’re interested in AI, there’s another lesson here.
The future of AI interfaces may not be another chatbot window.
It may be an AI that can act inside a spatial environment, combining maps, data sources and natural-language commands into one interface.
God’s Eye View isn’t literally giving anyone a god’s-eye view of Earth.
It is showing something arguably more interesting:
how much of the real world is already broadcasting information — and what becomes possible when someone builds a better window through which to see it.
