AI creates the first 100-billion-star Milky Way simulation

Simulating every single star in the Milky Way used to require 36 years of computing time for just one billion years of galactic evolution. Now, researchers in Japan have cracked that cosmic puzzle in just 115 days.

Keiya Hirashima and his team at RIKEN created the first simulation capable of tracking more than 100 billion individual stars across 10,000 years of our galaxy's life. Their secret weapon combines artificial intelligence with traditional physics calculations, creating a hybrid model that's 100 times faster than previous attempts.

The breakthrough centers on supernovae, those stellar explosions that scatter elements across space. Rather than calculating every detail of these cosmic fireworks in real time, the AI component learned to predict how gas spreads during the 100,000 years following each explosion.

"This achievement shows that AI-accelerated simulations can move beyond pattern recognition to become a genuine tool for scientific discovery, helping us trace how the elements that formed life itself emerged within our galaxy," Hirashima says.

The technique could revolutionize other fields wrestling with similar computational challenges, from climate modeling to weather prediction.