The quantum revolution is quietly gaining momentum. While on the surface it looks like nothing is changing, if we take a deep dive, we see that the first quantum processors are already tackling some of the world’s most intractable problems. Quantum processing is going to hit all industries and will change them forever. We are fortunate to live during this historic moment in which quantum technologies, beyond a physicist’s dream, are real and taking off.
Many experts nonetheless fear the “quantum winter,” a period in which no dramatic improvements are made in the capabilities of quantum computers, so that the technology is “frozen.”
Today’s quantum computers have hundreds of qubits—the quantum version of the binary bit (a variable that can be either 0 or 1). These qubits are prone to noise and, therefore, to error, so the computations carried out by today’s machines are far from perfect.
The long-term vision for quantum computing is the ability to leverage millions of noise-free qubits to carry out extremely hard calculations in a reliable way. As of today, however, we still do not know for sure how to scale up from hundreds of error-prone qubits to millions of error-free ones.
This is perhaps the most relevant experimental problem that we face in quantum computing, and to many, it is also a signal that it is still too early to find useful applications of quantum computers in real-life scenarios.
As an academic and entrepreneur, I respectfully disagree with this point of view. I am not saying that the problem does not exist; it certainly does. But in my opinion, we are asking the wrong question.
Asking the right quantum question
Rather than ask when quantum computers will be more powerful, we should be asking, What can we do now with today’s limited and noisy quantum computers?
This question is the right one from a pragmatic point of view. It’s also remarkably hard to answer. But if we manage to answer it, at least in part, then we can prove that quantum computers have value now.
This situation is not new; indeed, history teaches us that, when it comes to technology, asking the right question is more than half the challenge. Imagine going back to the 1970s and asking Apple whether its first computer had a speech recognition system. Obviously not—but that did not mean the machine was useless! You could do lots of useful things with the Apple-1, such as develop programs, play videogames and more.
Go back even further and ask Alan Turing whether the machine he invented to decipher crypto codes could be used to play videogames. Complete nonsense—at the time.
Today’s quantum computers have obvious limitations. We cannot run many of our favorite quantum algorithms beyond toy models that don’t represent real use cases. We cannot yet crack the RSA cryptosystem or search huge unstructured databases. At the same time, today’s quantum machines are not useless.
Quantum use cases in finance, operations, and energy
So, what can we do with a quantum computer today that is useful in practice? You might be surprised to discover that we’re already doing a lot. In fact, the first prototypes of quantum computers are offering real industrial value in specific, well-targeted business applications. Some of these solutions are already moving into production.
What matters in industry is finding business advantage, i.e., applications for which, for whatever reason (perhaps even illogical ones), quantum computers can be more practical to use than classical resources. This is what matters in a business, and finding such applications is an art in itself.
As the World Economic Forum explains, quantum computing is expected to work best across three specific areas: materials science, optimization and artificial intelligence. With optimization in particular, companies across multiple industries are using quantum and quantum-inspired algorithms to test new solutions now.
Banks, including JPMorgan Chase and Credit Agricole CIB, are applying quantum algorithms against optimization problems that cover portfolio optimization, index tracking, ETFs, trading decision systems, fraud detection, option pricing and credit scoring.
In another optimization use case, the Port of Los Angeles used a quantum-powered artificial intelligence engine to increase the capability and velocity of cargo movement. The AI engine doubled the productivity of cargo handling equipment, produced more predictable cargo flows, optimized scheduling and streamlined container handling for trucking companies.
Green energy is another vertical with lots of applications. We have already seen business advantages in energy market optimization, energy production forecasting, windfarm design and more. Even small gains can return significant results. Forecasting renewable energy production is one of the industry’s most difficult challenges. Quantum and quantum-inspired machine learning algorithms can reduce the error rate in these predictions, in turn saving millions of euros per year.
Build once, deploy many times
In fact, the number of use cases is unending, because the core of the algorithms is universal and the technology transversal. Consider predictive maintenance, for example: The same algorithm used to determine the probability of a machine failure in a factory can be used in a hospital to predict a patient’s probability of having specific health problems, and even in soccer clubs to predict the probability of player injuries. And this is not sci-fi: All these use cases are already being implemented as you read this article.
The implications are enormous. When it comes to industry processes, quantum is not an alternative, but a necessity. And the quantum processors that we have available right now are already able to provide business value, beating internal benchmarks by using a combination of limited quantum hardware and clever algorithms.
Companies need to investigate where their advantage might be today from quantum before their competition does. It’s not an easy task, but with the right partners and elbow grease, it can be done.
Find Roman Orus' column about finding value in quantum computing on EE Times here.