Japanese technology outfit NEC has stopped developing physical quantum computers after nearly 30 years, leaving rival Fujitsu to carry the expensive torch.
According to TechRadar, NEC has reportedly ended hardware development involving superconducting circuits and solid-state qubits, shifting its attention towards quantum applications instead.
Several engineers involved in the programme, including a senior researcher, have taken their potentially dead-or-alive cats and moved to Fujitsu, where they can keep working on quantum hardware.
It marks a retreat for a company that helped establish the foundations of modern quantum computing before today’s industry had acquired its impressive collection of roadmaps. NEC was once a semiconductor heavyweight, ranking first globally for chip sales in 1990. Hitachi, Toshiba, Fujitsu, Motorola and Intel were all trailing it at the time.
In 1999, NEC researchers demonstrated what is regarded as the earliest solid-state qubit using superconducting technology. That work became an important milestone in quantum computing and helped establish techniques now reflected in superconducting processors developed by Google, IBM and others.
NEC’s retreat from the hardware business did not happen overnight. The company effectively left DRAM production in 1999 when it folded those operations into a joint venture that later became Elpida Memory.
In November 2002, NEC spun off its remaining semiconductor businesses, including microcontrollers and system-on-chip products, into NEC Electronics. That operation eventually became part of Renesas Electronics.
Today’s NEC focuses on IT services, enterprise systems, aerospace, defence, telecommunications infrastructure, AI, and security. Building quantum computers apparently no longer sits comfortably within that collection, particularly given the substantial cost and uncertain timetable for commercial returns.
NEC has not described the decision as a complete withdrawal from quantum computing research. The company intends to continue studying practical quantum applications and possible commercial uses, meaning it has abandoned the particularly expensive business of building the machines rather than the technology altogether.
Fujitsu is taking the opposite approach. Working with the Japanese research institute RIKEN, Fujitsu launched a system containing 256 superconducting qubits in 2025. It has targeted a superconducting quantum computer with more than 1,000 qubits during fiscal 2026.
Fujitsu’s longer-term plans are considerably more ambitious. The company is developing a machine with more than 10,000 physical superconducting qubits, to be completed around 2030. That project is intended to implement Fujitsu’s early fault-tolerant STAR architecture and initially provide around 250 logical qubits.
The company believes those systems could eventually tackle useful workloads in areas such as materials science, chemistry and finance.
NEC’s departure provides a less glamorous counterpoint to an industry where quantum computing regularly seems five years away from changing civilisation. Even a company with decades of semiconductor expertise and one of the field’s earliest qubit breakthroughs has decided that building the hardware no longer makes commercial sense.
