Astrophysicist Uses Codex to Refine Black Hole Simulations

Researchers are working to improve computer simulations of black holes, a task made more challenging by the extreme gravity and physics involved. One astrophysicist is using an AI tool called Codex to help develop new algorithms that can model the behavior of electrons and ions around these cosmic phenomena.

Black holes are regions in space where gravity is so strong that nothing, not even light, can escape once it gets close enough. Astrophysicists like Chi-kwan Chan study black holes using computer simulations and observations to better understand their properties and behavior. However, current algorithms and computing power limit the realism of these simulations.

Chan, a researcher at the University of Arizona's Steward Observatory, is part of an international collaboration that published the first image of a black hole in 2019. The team is currently gathering observations to produce the first video of a supermassive black hole, focusing on one located at the center of the M87 galaxy.

To create these simulations, scientists study the region around a black hole called the event horizon, where matter cannot escape. They observe and measure light emitted by matter swirling just outside this boundary, which can be used to simulate the behavior of electrons and ions in this extreme environment.

One major challenge is modeling plasma, superheated matter made up of electrically charged particles that rarely interact with each other near black holes. Current simulations often simplify plasma as a fluid, but this approach fails when particles mostly spiral around magnetic field lines rather than colliding with each other.

Chan suspected that new mathematical techniques could help overcome these limitations and turned to Codex to derive candidate algorithms and test them against known solutions. The AI tool generated many potential approaches, not all of which were correct, but this was acceptable as long as the algorithms were testable.

Codex proposed numerical schemes that Chan's group could inspect, test, and understand physically. While large language models can make mistakes, scientists like Chan believe AI systems have a place in research when used rigorously to explore ideas, test them faster, and accelerate discovery while maintaining verification and reproducibility.