News

AI exposed the universe’s secrets

Scientists built an AI called SimBIG that looks at galaxy patterns like it’s playing cosmic Sudoku, and it’s crushing it. It figured out the “settings” of the universe (like how much dark matter exists) way more accurately, with way less data.

Old methods needed four times more galaxies to do the same thing. Bonus: This could help settle the universe’s biggest nerd fight: How fast it’s actually expanding. If the numbers still don’t add up? Congrats, we might’ve just found new physics.

A new AI-powered method is changing how scientists measure the universe. Developed by researchers at the Flatiron Institute and their partners, this technique offers a far more accurate way to determine the cosmos’ key properties.

The approach, known as Simulation-Based Inference of Galaxies (SimBIG), pulls hidden clues from galaxy patterns. It goes beyond older techniques by uncovering information that was previously out of reach.

Using AI, the team cut uncertainty in critical parameters—like how clumpy matter is in the universe—to less than half. These results match closely with other cosmic measurements, including the light from the universe’s earliest moments.

Published in Nature Astronomy, the study marks a major leap forward. With SimBIG, researchers can now probe the universe’s structure with greater confidence and detail.

“These parameters are essentially the ‘settings’ of the universe that determine how it operates on the largest scales,” says Liam Parker, a research analyst at the Flatiron Institute’s Center for Computational Astrophysics and co-author of the study.

Older tools focused on large-scale galaxy clustering. While useful, those methods often missed subtle patterns. SimBIG changes that by digging deeper into galaxy data, capturing smaller features that fine-tune our cosmic understanding.

This leap in precision isn’t just a technical upgrade—it’s a new doorway. SimBIG opens fresh paths for studying how the universe formed, evolved, and continues to expand.