Anthropic's Claude Science released a complete ultraviolet map of the sky, marking the first time such a map has been created. The project aims to demonstrate how AI can automate tedious data work that researchers often delay. The map reveals details about dust illuminated by starlight, such as clouds around young stars or rings from stellar explosions.
The map was made possible by combining data from multiple space missions, with AI agents calibrating and merging the information. Missing areas were filled using inpainting, a technique where models learn from existing data to reconstruct gaps. In tests, predictions averaged about ten percent deviation from actual measurements.
"UV light reveals dust lit up by starlight, things like clouds around young stars or rings left behind by stellar explosions," said Brice Ménard, an astrophysicist at Johns Hopkins. Ménard suspects many scientists have been putting off similar projects that AI could now make possible.
The map is intended as teaching material, highlighting how AI can assist in scientific research. It follows NASA's GALEX mission, which covered about two-thirds of the sky but skipped bright star-forming regions. The project underscores the growing role of AI in scientific discovery.
Anthropic did not say how the map will be used beyond its educational purpose, and it remains unclear how widely it will be adopted. The team plans to continue refining the technique for future applications.
Source: thedecoder
