Ope nAI's Codex, an A I model that tran slates natural langu age into code, is now used by astrophysicist Chi-kwan Chan to build black hole simulations that test Einstein's theory of general relativity, according to a case study published by OpenAI.
Simulating black holes requires solving complex systems of equations derived from general relativity, a task that traditionally demands extensive manual coding. Per OpenAI's report, Chan uses Codex to translate natural-language descriptions of desired functions into working code, letting him focus on the physics rather than programming syntax.
Because black holes cannot be observed directly, simulation is the primary way scientists study phenomena such as gravitational waves and accretion disks. OpenAI's writeup frames Codex as a way to speed up writing and iterating on this simulation code, reducing the manual effort involved and potentially making high-performance computational work more accessible to physicists who are not specialist programmers.
The piece is a promotional case study from OpenAI itself, so its claims about efficiency gains should be read as the company's own characterization of the collaboration rather than independently verified results.