Japan's NEC Halts Quantum Computer Project After Decades Of Research

NEC has ended its effort to develop a working quantum computer, reportedly deciding that the project would take too long to deliver an acceptable return on investment, according to a new report by Nikkei.

The move represents a significant change in direction for a company that has been involved in quantum computing research for more than three decades.

NEC was among the earliest companies to pursue the technology, beginning its research in the 1990s and achieving a major breakthrough in 1999 with the first demonstration of superconducting qubits. Those qubits remain one of the principal approaches used in quantum computer development today. Despite that early lead, NEC discontinued work on its own physical quantum computer at the end of March.

Nikkei reported that the company will continue pursuing quantum-related technologies and services, including quantum annealing, which is used to find efficient solutions to complex optimization problems. NEC also plans to expand services that use conventional computers to simulate quantum computing.

The shift suggests that management sees a more immediate commercial opportunity in applying quantum-related techniques than in funding the lengthy and expensive development of its own hardware.

NEC’s retreat comes as other major players continue investing heavily in the field. IBM and Google remain prominent competitors in the United States, while China is advancing quantum computing through coordinated public- and private-sector efforts.

In Japan, Fujitsu is continuing its research and signed an agreement in August with an Australian university and government research institution to collaborate on quantum-related projects.

The industry has nevertheless made meaningful technical progress over the past two to three years. Google’s Willow processor, introduced in 2024, demonstrated that error rates could decline as additional qubits were added, an important step toward building larger and more reliable systems.

Microsoft and Quantinuum have also reported advances in error correction, while Amazon’s Ocelot prototype, unveiled in 2025, was designed to reduce the hardware required to produce reliable logical qubits. Increasingly, the challenge is not simply to build machines with more qubits, but to make those qubits stable enough to perform useful calculations.

That progress has yet to resolve the commercial question. IBM’s experimental Loon chip, unveiled in November 2025, forms part of its effort to develop a fault-tolerant quantum computer by 2029, while Google reported another quantum-advantage demonstration in October.

Such milestones show that the technology is advancing, but they do not establish when quantum computers will become broadly useful or economically viable. NEC’s decision therefore illustrates the distinction between scientific progress and investment returns: the industry may be moving forward, but the timetable for turning those advances into a profitable business remains uncertain.