IBM managed to deliver a 433-qubit machine this year, and hopes to reach 1,121 qubits next year. These are the goals that IBM has set for itself in its hardware roadmap for 2022 and 2023. These announcements, if they reflect the hype around quantum, suggest that use cases will begin to emerge within a few years.
However, it is a difficult thing to predict, explains Pierre Jaeger, quantum leader IBM France. Because the promises of exponential computing capacities come up against the pragmatic vision of companies, which do not rely on technology without financial gain at stake.
“Without business gain, quantum experimentation stops”
IBM takes an approach focused on business value for the company. After the technical analysis, the question of the financial gain to be made for the company very quickly arises, which justifies going through quantum rather than classical calculation.
Some use cases go as far as PoC, or even a little further in their development, when in the specific case, quantum technology is able to do as well as conventional technology. Work progresses “faster when companies are accompanied by experts, or when they have reached a better level of maturity”, comments Pierre Jaeger.
There are two areas where the use cases are particularly mature, on fundamental physical science and improving AI models.
“Not everything can be solved by quantum”
But “everything is not solved by the quantum”.
“There are situations where we may have to backtrack and not test the use case. It’s not the fault of the technology as such, that’s how the segmentation of use cases is done,” explains Pierre Jaeger.
The IBM Quantum Network connects more than 200 members to date, and the number of participations is “increasing”, mainly on the side of businesses and service companies, notes Pierre Jaeger. Training modules are offered to professionals who join the network so that quantum is easier to understand for non-specialists.
“We have left the laboratory, but we are not yet in industrialization”
IBM still wants to unravel the mysteries of technology. In the premises of IBM Research in Zurich (Switzerland), the first laboratory established in Europe of the American company, the company’s researchers do not hesitate to explore the technology of semiconductors applied to quantum computing. While IBM bases its hardware policy on superconducting materials, it does not hide its scientific interest in silicon on which other competitors, such as the very recent French deep tech Siquance, are also working.
The group is working in parallel on “all kinds of qubit technologies,” Pierre Jaeger told ZDNet. And to add: “we forbid ourselves to say that superconductor technology is the only one”, even if it is one of those which will pass “the fastest on the scale”.
If the “hype” for quantum has contributed to the emergence of a host of new players in recent years, Pierre Jaeger insists on the fact that time is “long” in the world of quantum. “We have left the field of the laboratory, but we are not yet in the industrialization phase”, he nuances.