Claude Helps Astronomers Map 119 Million Stars in Ultraviolet Light

For the first time, humans have a complete ultraviolet map of the entire sky. And they didn't do it alone. Astrophysicists teamed up with Claude, Anthropic's large language model, to fill in the missing pieces of a puzzle that has stumped astronomers for half a century.

The new map contains 119 million stars. About one-third of the sky was filled in by Claude, which used existing multi-wavelength data to predict what the ultraviolet sky should look like in regions that had never been properly observed. The gap existed because ultraviolet telescopes have historically avoided the bright, dusty plane of the Milky Way, leaving a huge blind spot.

From Weeks to Days

Building an all-sky map is no small feat. It involves collecting data, calibrating it, cross-checking different sources, and then filling in the gaps. Normally, this would take a team of astrophysicists several weeks of tedious manual work. But with Claude, the entire workflow was completed in just a few days.

The AI didn't just speed things up—it also proved surprisingly accurate. When compared to real measurements, the areas completed by Claude had an error rate of only about 10%. To keep things scientifically rigorous, the model labeled every pixel in the final image, noting whether it was measured or calculated and giving the corresponding error range.

What This Means for Science

This isn't just a cool tech demo. It's a sign of things to come. Fundamental research problems that have been stuck for years because of limited human resources are now being tackled by AI agents. Instead of spending weeks on low-level data processing, researchers can focus on steering the science and asking the big questions.

"We can step away from the tedious stuff and focus more on controlling the direction of research and identifying core scientific issues," the team noted.

So what's next? If AI can help map the entire ultraviolet sky, it might soon help with other massive data challenges in astronomy and beyond. One thing is clear: the partnership between humans and AI is opening up new windows on the universe.

Key Points:

First complete ultraviolet all-sky map created with Claude's help
119 million stars included; AI filled in ~1/3 of the sky
Error rate ~10% compared to real measurements
Workflow reduced from weeks to days
AI handles tedious data tasks, freeing scientists for higher-level work

Claude Helps Astronomers Map 119 Million Stars in Ultraviolet Light