Jensen Huang says quantum computers could simulate biology and chemistry to generate synthetic training data that more closely reflects the physical world for AI models. On October 19, 2026, Nvidia announced a record simulation of a 104-qubit fault-tolerant quantum processor on the JUPITER supercomputer using 8,192 GH200 superchips. The simulation validated an error-correction method called magic state cultivation, which is said to require 90% fewer physical qubits than traditional approaches. The experiment serves as a dress rehearsal for testing quantum designs before chip fabrication and illustrates Nvidia’s strategy of pairing GPUs with quantum processors. The potential gains include molecular simulation, drug discovery, and materials science, but the method has so far been validated only in simulation.
Source: Read the original article
