AI builds interactive ASCII universe
An AI project has created an explorable ASCII universe that can be zoomed from a moon to the observable universe. The interface uses printable ASCII characters, with denser glyphs adding light and physics‑based color derived from blackbody temperature, Doppler shift, and emission lines.
Key points
- Interactive ASCII universe zooms from moon to observable universe
- Physics‑based color from blackbody temperature, Doppler shift, emission lines
- Real positions and sizes of 1,982 stars, Milky Way, and cosmic web
Users can drag to orbit, scroll or pinch to zoom, tap to fly to objects, and search the entire atlas. The interface shows real positions and sizes, including 1,982 naked‑eye stars, the Milky Way, its neighbours, and the cosmic web. Brightness is compressed to keep faint objects visible, and fast processes are sped up. Music is generated in the browser.
The project requires WebGL 2, which must be enabled in the browser or device. It showcases how AI can generate interactive, physics‑accurate visualizations using only text characters.
An AI built this explorable interactive ASCII universe in scale. A dream project finally come true
gcdatlas.com · 25 September 2026Loading the full article…
This text was published by gcdatlas.com. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
Coverage and discussion
2sources- Reddit discussionreddit.com
- Reddit discussionreddit.com
- Built this explorable interactive ASCII universe in scale with AIPress · gcdatlas.com ·
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Research
All →- OpenAI’s Astra and Anthropic’s Opus break two unsolved Enigma messages · 1 src
- OpenAI fires contractors for using AI to train its models · 2 src
- OpenAI’s research chief discusses AI safety and self-improvement goals · 1 src
- AWS boosts MoE RL training throughput by 40% with EFA and DeepEP on EKS · 1 src
- AWS shows how to train Qwen3-VL-8B with SkyRL on SageMaker HyperPod · 1 src
Comments
via GitHub Discussions