A superconducting neuromorphic chip passed tests processing fusion-reactor data under the Genesis Mission project.

HPCwire reported that a superconducting neural-network chip successfully passed a battery of tests involving data from a fusion reactor. The article describes a neuromorphic chip that borrows techniques from quantum computers, rather than identifying it as a general-purpose quantum processor.

A related announcement names Great Sky, Brookhaven National Laboratory, Princeton Plasma Physics Laboratory, and the DIII-D National Fusion Facility as collaborators developing superconducting AI chips for plasma control. Their stated development objective is hardware fast enough to sense and respond to the plasmas present in fusion reactors.

The Genesis Mission report describes a planned larger chip for additional fusion testing in 2027. The supplied excerpts do not disclose numerical latency, accuracy, power consumption, or evidence that the tested chip operated a reactor in closed-loop control.