Quantum Computing Remains Experimental Despite 2024 Advances: Forrester

Quantum computing made significant strides in 2024, but it’s yet to demonstrate a practical advantage over classical digital computers, according to a recent trends report released by Forrester Research.

“Despite improvements in qubit count, coherence time, and gate fidelity, the technology remains experimental, with broad-scale applications likely still a decade away,” maintained the report written by Forrester Vice President for Emerging Technologies Brian Hopkins and Principal Analyst David Mooter, with Stephanie Balaouras, Mike Gualtieri, Charlie Dai, James McGlynn, and Jen Barton.

“Key developments in optimization, quantum simulation, and quantum machine learning show promise for specific industries like finance and pharmaceuticals, yet challenges such as high error rates and scalability persist,” the analysts added.

Roger A. Grimes, author of “Cryptography Apocalypse: Preparing for the Day When Quantum Computing Breaks Today’s Crypto,” published by Wiley, acknowledged that practical, usable quantum computers haven’t made a public appearance yet, but some useful applications of quantum have been deployed, such as quantum random number generators, networks, sensors, and all sorts of other devices.

“No one has publicly shown a problem solved by a quantum computer that is super usable in the real world,” he told TechNewsWorld. “Today, your wristwatch has more power than what we have in quantum computers. But that is changing. We are making steady progress, and the day when quantum computers turn into sufficiently-capable quantum computers is not far away.”

Quantum Solving Problems Now

Trevor Lanting, chief development officer at D-Wave Systems in Vancouver, B.C., Canada, agreed that gate-model quantum computing has not yet demonstrated a practical quantum advantage but pointed out that annealing quantum computing is delivering value over classical computing today.

Gate-model quantum computers use quantum logic gates to perform operations on qubits, similar to how classical computers use logic gates to perform operations with digital bits. The gate model is more suitable for general-purpose computing, while annealing quantum computing is more narrowly focused on optimization problems, such as workforce scheduling and portfolio optimization.

D-Wave has been using annealing quantum technology in a hybrid solution to solve complex optimization problems. For example, it was able to optimize the mobile network resources of Japan’s largest telecom provider, NTT Docomo, in 40 seconds, compared to 27 hours using classical methods.

Forrester’s report predicted that gate-based quantum computing platforms will likely remain experimental for 10 to 15 years, a prediction that Lanting agrees with. “However, annealing quantum computing — which is uniquely suited for solving complex optimization problems — is here now,” he told TechNewsWorld.

“Optimization problems are everywhere — from workforce scheduling to vehicle routing — and D-Wave’s annealing quantum computers are already delivering measurable results for customers,” he said.

Lanting maintained that D-Wave’s technology helped Pattison Food Group, a Canadian grocery chain, reduce an 80-hour scheduling task to just 15 hours — an 80% time savings — and at the Port of Los Angeles, working with SavantX, cargo handling efficiency was improved by 60%.

“These aren’t theoretical use cases,” he said. “They’re real businesses solving real problems right now with quantum and hybrid quantum computing.”

Optimization Apps Will Lead Way

While D-Wave’s annealing platform has been touted by the company for years as superior to gate-based solutions for optimization problems, Forrester pointed out that those claims were challenged in 2024.

“Q-CTRL has challenged D-Wave’s claim by using IBM’s gate-based quantum computers to outperform D-Wave for an optimization problem,” the analysts wrote. “Gate-based algorithms offer the potential for greater solution speedups as qubit counts and quality increase.”

“This makes Q-CTRL’s claim an interesting challenge to D-Wave’s self-proclaimed lead in optimization,” they continued.

Optimization is an important application for quantum computing because it matters to most industries, Forrester noted. “For finance, representative areas include risk modeling, trading strategy optimization, asset pricing optimization, and portfolio optimization,” it explained.

“Health care use cases include optimizing radiotherapy treatments, generating targeted cancer drug therapies, and creating protein models,” it added. “And for energy, use cases include energy exploration, seismic survey optimization, reserve and spot trading optimization, and reservoir optimization.”

In the near term, if quantum is going to go beyond experimentation and generate a return on investment for users, it’s going to be through optimization tools, explained Erik Garcell, director of quantum enterprise development for North America for Classiq, a global quantum computer software maker.

“Optimization offers more near-term benefits because it scales so well on quantum computers,” he told TechNewsWorld. “Even having a few quantum resources, you know, 100 qubits on your quantum chip, means a lot for that kind of problem.”

“The bigger the problem, the harder it is for a classical computer to solve,” he continued, “but that many more resources aren’t needed for a quantum computer because of how it scales. So, you’re going to actually see applications in quantum for very large optimization problems that are causing classical computers to chug.”

Quantum Machine Learning Shows Signs of Life

Grimes, though, contends focusing on optimization can be too limiting for quantum. “Optimization completely rules out the chances of brand-new advancements,” he said. “I’m not sure if quantum is like the AI world, but it is getting the same sort of feel.”

“It really feels like we are possibly on the cusp of tremendous, sudden improvement,” he continued. “There are so many organizations making steady improvement. It seems strange to me to think that not one of those vendors out of the thousands wouldn’t make a substantial breakthrough.”

Another 2024 quantum breakthrough cited by the Forrester analysts is the emergence of quantum machine learning. They explained that quantum-as-a-service (QaaS) has expanded access to quantum computing, enabling breakthroughs in quantum machine learning. Researchers are now developing quantum neural networks, quantum support vector machines, and quantum algorithms for complex tasks like image and natural language processing.

“These advancements are pushing the boundaries of what machine learning can achieve, making it a critical area of growth,” they wrote.

Training AI models on classical computers is time-intensive and computationally expensive, especially with deep learning networks, observed Skip Sanzeri, co-founder and COO of QuSecure, a maker of quantum-safe security solutions, in San Mateo, Calif.

“Using an algorithm like the Quantum Approximate Optimization Algorithm, along with other quantum enhancements like gradient descent, could speed up the training of machine learning models by orders of magnitude,” he told TechNewsWorld.

Sanzeri also pointed out that AI on classical systems is not exponentially scalable, so classical machines tend to struggle with combinatorial problems like optimization. “Since quantum computers are exponential in nature, they will be able to handle these combinatorial problems much better,” he explained.

Quantum algorithms can also be used to process and analyze large data sets more efficiently using the Quantum Fourier Transform, leading to faster insights and real-time decision-making, he noted.

Generative AI models, too, can be challenging for classical computing systems, he added. “Superposition and entanglement — quantum properties — can be used to generate data distributions more efficiently and accurately,” he said.

Preparation for Quantum Threats Becoming Urgent

As quantum advances push the boundaries of what machine learning can achieve, they’re also driving a focus on quantum security. “With NIST [the U.S. Department of Commerce’s National Institute of Standards and Technology] setting standards for quantum-resistant algorithms, the need to safeguard data against future quantum threats is becoming more urgent,” the Forrester analysts wrote.

They added that cryptography and machine learning hold substantial potential, but their benefits remain years away. Shor’s algorithm could one day break today’s PKI cryptography, they noted, although this is likely a decade or more in the future.

Sanzeri disagreed with Forrester. “With the use of ever increasingly powerful AI, combined with other breakthroughs in the quantum computing industry — like Google’s Willow chip — we could see that 10-year time frame getting cut in half,” he argued.

Meanwhile, Grimes cautioned that government intelligence agencies working to break quantum-susceptible encryption don’t need a fully-capable quantum computer to break today’s cryptography. “They will make quantum devices that are specialized in breaking encryption, just like they do with today’s regular non-quantum encryption breaking,” he said.

“The NSA isn’t using laptops, servers, and regular cloud computing to break crypto,” he continued. “They use specialized crypto-devices, maximized for crypto-breaking efficiency. Certainly, they are doing the exact same thing with quantum cracking. It would be insane to do anything else if I was in their shoes.”

He also warned about using Shor’s algorithm, created in 1994, as a benchmark for what kind of quantum power is needed to crack PKI cryptography. “I think there is a good chance that the U.S. government has access to another algorithm that is far more efficient than Shor’s,” he contended. “If you’re fixated on Shor as the standard to meet, you’re probably not focused on the right algorithm.”

Even if a quantum solution that can crack PKI is 10 years away, the time to act on that possibility is now, declared Tomas Gustavsson, chief PKI officer at Keyfactor, a digital identity management company, in Cleveland. “A decade is a short time for this type of migration, as it is an immense undertaking,” he told TechNewsWorld.

“We need to act now for the migration to be completed in a decade. Organizations must not start in a decade,” Gustavsson said. “So, when saying that a practical Shor’s algorithm is at least a decade away, Forrester is reiterating what NIST and others are saying. I also hope it’s at least a decade away. If not, we are in trouble.”

Winter of Quantum Investor Discontent?

Despite quantum’s promise, Forrester predicts a “winter” setting in on investment in the technology. “Although the number of quantum computing deals hit a record in 2023, investment dollar totals peaked in 2021 and have declined sharply since as generative AI excitement has claimed investor funding,” Forrester’s analysts wrote.

They added that geopolitical pressures, like vendors in China transferring their IP to academia, are also at play. “This will put pressure on startups, causing many to seek exits with little to show,” they noted.

On the plus side, the investment winter will delay the time when quantum computing platforms become powerful enough for mainstream use, which means a delay of Y2Q: the day when quantum computers break state-of-the-art asymmetric cryptography, the analysts reasoned.

However, they warned about procrastinating about Y2Q. Although this development may buy more time to implement post-quantum encryption, they urged security leaders to begin planning now on how to protect against “harvest now, decrypt later” vulnerabilities, where encrypted data is gobbled up by adversaries now with an eye on decrypting it later with a quantum computer.

Forrester’s glum weather forecast isn’t shared by everyone. Growth estimates for the quantum computing market by industry forecasters range from a low CAGR of 27.04% over the next eight years to a high CAGR of 32.7% over the next five years, growing from slightly over a billion dollar market in 2024 to US$6.95 billion to $16.22 billion market in the early 2030s.

“We would not expect a quantum winter but may see the same level of investment or increased investment in 2025,” Sanzeri predicted. “There has been significant progress in quantum computing in 2024 with some breakthroughs that are fundamental. We cannot find a reason why there would be at least the same level of investment in 2025 as in 2024.”

“The recent relative downturn in quantum investment is really just a factor of the original, overstated quantum hype dying down at the same time as AI took off,” added Grimes. “Quantum will have plenty of investment, and as the steady improvements turn into sufficiently-capable quantum computers, the needed investment will flood back. I’m not concerned.”