Mitsubishi Electric has begun two research-and-development projects intended to address a less visible challenge in quantum computing: the control hardware needed to scale machines beyond laboratory-scale systems. The company said on September 17 that Japan’s New Energy and Industrial Technology Development Organization, or NEDO, had selected both projects under a post-5G research programme.
The announcement does not introduce a finished quantum computer or a commercial service. It is a component-and-systems effort spanning three distinct hardware approaches: neutral atoms, trapped ions and superconducting circuits. That scope matters because the field has not settled on one winning qubit architecture, and each approach needs different ways to control fragile quantum states at larger scale.
What Mitsubishi Electric quantum computing R&D covers
The first project, called Research and Development of Multi-Qubit-Control Laser Systems, targets the lasers and low-latency controls used in neutral-atom and trapped-ion systems. Mitsubishi Electric says it will apply laser technology used in machine tools alongside programmable logic controls to develop high-power, highly stable laser systems.
The second project, Development of Ultra-Compact, Multi-Channel, Low-Noise Amplifier Modules for Large-Scale Superconducting Quantum Computers, focuses on the microwave electronics used to read and control superconducting qubits in cryogenic environments. In both cases, the stated objective is to make it possible to control more qubits, not to claim that a large-scale machine already exists.
That distinction is central. A qubit can be useful only while its quantum state can be prepared, controlled and measured reliably. As systems add qubits, the surrounding lasers, wiring, microwave electronics, cooling and error-correction infrastructure become as consequential as the qubits themselves. The technical problem is therefore not simply increasing a qubit count; it is preserving control quality while a system becomes more complex.
The government programme provides the broader context
NEDO’s public decision page, dated July 17 and updated July 22, places the work inside a 2026–2028 programme to accelerate next-generation quantum-computer development and demonstration. The agency describes the programme as covering system-scale development, component materials and software platforms, as well as applied demonstrations and workforce development.
For the component-development track, NEDO explicitly identifies core materials and equipment such as cryogenic refrigerators and lasers. That makes Mitsubishi Electric’s emphasis on laser systems and cryogenic amplifier modules more than a generic quantum announcement: it is a bid to supply enabling technology across multiple architectures. The company says it will validate the work with research institutions including Japan’s National Institute of Advanced Industrial Science and Technology.
The selection itself was public before Mitsubishi Electric’s September announcement. What is new is the company’s declaration that it is launching the two named projects and its fuller account of the components it plans to develop. Neither the company release nor NEDO’s decision page provides performance results, a completed prototype, a funding figure, a deployment customer or a date for commercial availability.
Why the architecture breadth is notable
Neutral-atom machines use laser light to manipulate atoms with no electric charge. Trapped-ion systems similarly use lasers, but operate on charged atoms held in place by electromagnetic fields. Superconducting systems use electrical circuits and microwave control at very low temperatures. These are not interchangeable engineering problems, yet each needs a route from individually controllable qubits to a system that can operate at useful scale.
Mitsubishi Electric’s programme is notable for concentrating on common bottlenecks in those control layers rather than announcing a wager on only one type of qubit. That may make the work relevant to an ecosystem of hardware builders, but it is still research. Whether any component produces a practical advantage will depend on stability, noise, integration, manufacturability and the results of later demonstrations.
For readers looking for the basic concepts behind these systems, Unhyd’s plain-English guide to quantum computing explains how qubits differ from classical bits. The longer-term security implications are addressed separately in our guide to quantum-safe encryption deployment; this announcement does not change those migration timelines.
What to watch next
The useful evidence now will be technical rather than promotional: measured laser stability, control latency, amplifier noise at cryogenic temperatures, the number of controllable qubits, integration with partner hardware and the terms of any demonstration. Mitsubishi Electric describes the projects as a path toward industrial applications, but it has not announced a deployed quantum-computing product. For now, the signal is that Japan’s quantum programme is directing attention not only to qubits themselves, but also to the hardware that has to make many of them work together.
