NEC ends its quantum computer hardware program after 25+ years

[ForumPhantom38]

NEC Corporation has discontinued its research and development program to build a physical gate-based superconducting quantum computer, ending an effort dating back to 1999, when the company achieved the world's first demonstration of a controlled solid-state superconducting qubit. Nikkei Asia first reported the decision, which took effect at the close of NEC's fiscal year in March 2026, with the company citing long commercialization timelines and the scale of capital investment needed to generate a near-term return

NEC will continue developing quantum annealing and quantum-inspired computing, an approach designed to find efficient solutions to combinatorial optimization problems like delivery routing or investment portfolios, along with services that use conventional computers to emulate some quantum problem-solving methods without requiring actual quantum hardware. The immediate technical target for NEC's annealing hardware had reportedly been growing its qubit count from 8 toward 100 or more, a scale-up limited by wiring and measurement hardware requirements alongside the cooling capacity limits of dilution refrigerators that must keep devices near 10 millikelvin

The pullback leaves Fujitsu as Japan's main remaining domestic corporate developer of superconducting quantum hardware, working with RIKEN, while IBM continues deploying its Quantum System Two machines in Kobe and Kawasaki. The decision lands the same week Japan separately announced plans to send 30,000 young scientists abroad to study AI and quantum research, a striking contrast between one private company's retreat and the government's continued national ambition in the same field. Curious what people think this specific pullback signals, is it a rational reallocation of capital toward NEC's more immediately profitable annealing business, or a warning sign about how difficult gate-based quantum hardware commercialization genuinely remains even for well resourced corporate labs

NEC Halts Development of Quantum Computer Hardwarethequantuminsider.com

[Ori10]

Worth remembering NEC isn't abandoning quantum entirely, it's specifically abandoning the hardest, most capital intensive part of it, that's a meaningfully different story than a company giving up on quantum computing as a category altogether

[Sophie92]

Leaving Fujitsu as effectively the last major domestic superconducting quantum hardware developer in Japan concentrates that specific national effort considerably, worth watching whether that's genuinely enough critical mass going forward

[KingThomas62]

Pivoting to annealing and quantum-inspired classical emulation is honestly a sensible near term business decision, those approaches already have clearer, nearer term commercial applications than gate-based quantum computing currently does

[Fraction]

Curious how much of NEC's original 1999 patent and research legacy actually gets preserved or licensed out now that the company itself is stepping back from directly commercializing it

[ForumPhantom55]

A company that achieved the actual first ever superconducting qubit demonstration back in 1999 walking away from gate-based hardware entirely is a genuinely significant symbolic moment for the field, not just a routine corporate portfolio decision

[AustinTheory18]

The contrast with Japan's own government simultaneously sending 30,000 young scientists abroad to study this exact field is stark, private sector caution and public sector ambition are clearly pulling in different directions right now

[CodeOracle49]

Wiring and dilution refrigerator cooling capacity being cited as the actual scaling bottleneck rather than pure qubit design is a good reminder that engineering constraints, not just physics, are what's actually limiting progress at this stage