What if the universe’s most mysterious giants were hiding secrets only AI could decode? Astronomers have just used artificial intelligence and a computing revolution to expose shocking new details about the black hole at the center of our galaxy.
Key Points at a Glance
- AI trained on millions of simulations reveals Milky Way’s black hole spins near top speed
- Breakthrough powered by 40 years of throughput computing innovation
- Magnetic field behaviors challenge established accretion disk theories
- Computing network processed over 12 million jobs across 80+ institutions
In a groundbreaking effort to uncover the mysteries of black holes, a team of international astronomers has harnessed the power of artificial intelligence and one of computing’s most underappreciated revolutions — throughput computing. Their target: Sagittarius A*, the supermassive black hole at the heart of our Milky Way galaxy.
Their findings? Not only is Sagittarius A* spinning at nearly its theoretical maximum, but the orientation of its spin axis is pointed directly at Earth. What’s more, the team has revealed that the hot plasma swirling near the black hole emits light primarily due to superheated electrons in its accretion disk, challenging previous ideas about jet-driven emission.
These insights were made possible by feeding a machine learning model — a Bayesian neural network — with millions of synthetic black hole simulations. This massive data ingestion was made feasible by the Center for High Throughput Computing (CHTC), a collaborative powerhouse operated by the Morgridge Institute for Research and the University of Wisconsin-Madison.
“The scale of the data required would have been impossible to handle with traditional computing,” says Chi-kwan Chan, Associate Astronomer at the University of Arizona. “Throughput computing lets us scale to millions of tasks and crunch them in parallel.”
Unlike conventional supercomputers that concentrate processing power on a few massive tasks, throughput computing excels at dividing workloads into millions of smaller jobs distributed across vast computing networks. This system has matured over four decades, thanks in large part to the vision of computer scientist Miron Livny.
The current research builds on the legacy of the Event Horizon Telescope (EHT), the team behind the iconic black hole images from M87 (2019) and the Milky Way’s own black hole (2022). But capturing images was only the first step. The real challenge lies in interpreting the torrent of data hidden within these cosmic snapshots.
“Our new AI-driven approach lets us peel back the layers,” says lead researcher Michael Janssen from Radboud University. “It’s like adding an ultra-sensitive lens to a telescope. Suddenly, we see dynamics and structures no previous model could resolve.”
And what they saw turned expectations upside down. The behavior of magnetic fields around the black hole didn’t align with the textbook models of accretion disks. Instead, they hint at more exotic, chaotic dynamics — a challenge to theorists and an opportunity for future discoveries.
The vast computational demands of this endeavor were met through the NSF-funded Open Science Pool, operated by the PATh project. More than 80 institutions contributed resources, completing over 12 million jobs in three years. This collaborative web of compute power allowed the EHT team to finally match real observations to the right theoretical models.
“We’re seeing a new era of astronomy,” says Anthony Gitter, Morgridge Investigator. “Where AI meets high-throughput computing, we’re unlocking the deepest secrets of the universe.”
Source: Morgridge Institute for Research
Enjoying our articles?
We don’t show ads — so you can focus entirely on the story, without pop-ups or distractions. We don’t do sponsored content either, because we want to stay objective and only write about what truly fascinates us. If you’d like to help us keep going — buy us a coffee. It’s a small gesture that means a lot. Click here – Thank You!
