Coding Horizon

This Open Source AI Map Feels Like A Command Room

Every figure, name and mark this video puts on screen, chased to a primary source. Checked 26 August 2026.

The repository

God’s Eye View, by Bilawal Sidhu, at github.com/bilawalsidhu/gods-eye-view. Described by its author as “A spy satellite simulator in your browser, except the data is real. Live open source spatial intelligence on a photorealistic 3D globe.”

On screen Value Where it comes from
stars 5,031 GitHub API, stargazers_count, 26 August 2026
forks 1,277 GitHub API, forks_count, 26 August 2026
licence MIT The repository’s LICENSE file and its README
language JavaScript GitHub API, language
first commit June 2026 GitHub API, created_at, 22 June 2026
topics geospatial-intelligence, osint, cesium, 3d-globe, flight-tracking, webgl GitHub API, topics

Stars and forks move. They are shown as the count on the date above and the shot carries that date in its source line.

On the licence. The LICENSE file is the MIT licence, with a note that the third party data and assets the app reaches for are not covered by it and stay under their own terms. GitHub’s own licence detector reports NOASSERTION for the repository because of that added note, so the API field and the file disagree. The file is what governs the code, so the shot says MIT.

What it puts on the map

Thirteen live layers, per the repository’s README:

Layer Provider named by the repo Access
Map stack Google Photorealistic 3D Tiles, Bing aerial, OpenStreetMap Google key required, metered billing
Live flights OpenSky, adsb.lol anonymous, credits optional
Military flights adsb.lol open
Live vessels AISStream free key
Satellites CelesTrak open
Earthquakes USGS open
Traffic TomTom, OpenStreetMap approximate without TomTom
CCTV mesh city APIs open
Radio Radio Browser and broadcasters open
Bikeshare GBFS open
Active fires NASA FIRMS free key
Space missions Launch Library 2 open, token optional
Mapped installations OpenStreetMap open

Bundled static datasets, per the same README:

On screen Value
datacentres 4,351
dams 704
submarine cables 712
public cameras about 800, across Austin, California and London

What it is built out of

What it costs to run

The README separates the keys three ways, and the video’s gate diagrams follow it exactly:

What the project says it will not do

The repository states that it models events, assets, infrastructure and systems, and that it does not build features for named person search, face recognition or tracking individuals. It also carries a data disclaimer that its data may be delayed, incomplete, modelled, inferred or wrong, and is not for navigation, emergency response, medical, or investment decisions.

That is why the video’s OSINT chapter draws vessels, aircraft, cameras and weather rather than people, and says so in the narration.

Marks used on screen

Every product named on screen carries its own published mark, taken as geometry from the simple-icons package: GitHub, Google Earth, Google Maps, OpenStreetMap, NASA, TomTom, Cesium, ESRI. The OpenAI mark is not in simple-icons any more and is carried as the published blossom geometry rather than approximated.

Not checked