The thrill of quantum in 2025

What a time to be alive. It may not feel that way with the current global industry turmoil, but underneath there are fundamental changes happening. The shift to quantum computing will create fundamental changes in the industry, and the underlying building blocks are clearly visible. Of course von Neumann machines with a Harvard architecture built in CMOS will continue to be the dominant technology for the trillion dollar semiconductor industry by 2030, or thereabouts. But this feels very much like the equivalent of the late 1950s for classical electronics.

From 1955 there were many types of technology being developed for various applications, from germanium and gallium arsenide to bipolar silicon. University and industry labs such as Bell Labs around the world were racing to develop integrated circuits, from Jack Kilby at Texas Instruments to Kurt Lehovec at Sprague Electric and Bob Noyce at Fairchild Semiconductor. The development of complementary metal oxide semiconductors (CMOS) at Fairchild led directly to Intel, while companies such as Sperry, Remington Rand, Burroughs, Amdahl and IBM were all building large-scale, centralized computer systems before the PC revolution.

Today in quantum computing there are similarly many different technologies for quantum computing. Some are using superconducting Josephson junctions (another technology from the 1960s) while others are building spin qubits on CMOS. Others are trapping ions, moving neutral atoms around a 2D grid, building cat qubits or using nitrogen vacancies in diamond. “There are 500 quantum computing companies out there,” says Mandy Birch, CEO at TreQ in the UK which build quantum computers from a variety of technologies.

“We’ve seen the supply chain develop over the last five years, new cryogenics, room temperature control, wireless and laser systems, and we are now starting to see companies emerge that are developing the quantum processors,” she said. “There is innovation out there but there hasn’t been a landing place for those and that’s what we are doing.” There are also photonic systems that do not need to operate around zero Kelvin, and even room temperature systems using diamond vacancies. The supply chains for all of these are maturing, with various approaches to implementing error correction to ensure that the qubits can stick around for long enough to do some useful work.

What is different is that the industry has understood the importance of the full stack. IBM was founded in 1911 and has evolved through many different technologies, but has always had a ‘full stack approach’. So quantum computing companies such as Rigetti, D-Wave and Quantinuum are all partnering to run all kinds of algorithms on their hardware. But this is just like the early days of mainframes, with dedicated algorithms implemented in special ways by ‘quantum wranglers’ to find new materials, new drugs or new ways to optimize the power grid.

The next stage will be the general purpose quantum operating system, which is where Microsoft has been driving its own hardware technology. Where does Europe sit in this evolution? We have vibrant quantum clusters in the Netherlands with companies such as Quantware and in France with Alice and Bob, Pascal and Quobly as well as Finland with IQM successfully commercializing technology from research lab VTT. Infineon Technologies has been driving ion quantum processor technology from Villach, Austria, while the UK is also looking to boost quantum hardware through companies such as Oxford Quantum Machines, Orca Computing, Quantum Motion and Oxford Ionics, as well as Equal1 in Ireland, with quantum computing systems being benchmarked in Oxford later this year.