If “figure out quantum computing” is still in your future file, it’s time to update your timeline. The industry is nearing the end of the early adopter phase, according to one expert, and the time is now to get up to speed.
Denise Ruffner, the vice president of business development at IonQ, said that quantum computing is evolving much faster than many people realize.
“When I started five years ago, everyone said quantum computing was five to 10 years away and every year after that I’ve heard the same thing,” she said. “But four million quantum volume was not on the radar then and you can’t say it’s still 10 years away any more.”
IonQ announced in October 2020 its next generation quantum computer system with 32 qubits and an expected quantum volume greater than four million. Quantum volume is a measurement of the overall power of a quantum machine.
SEE: What CIOs need to know about quantum computing (ZDNet)
Christopher Savoie, CEO of Zapata Computing, said quantum computing is quickly approaching a turning point where it will start to generate real value for businesses that can’t be achieved with classical computing.
“IBM, Google, Honeywell, Rigetti, IonQ and others are all creating increasingly powerful quantum devices,” he said. “In the near term, these devices are noisy and prone to error, but the right software can correct for these errors and unlock applications.”
Ruffner recently joined IonQ and previously worked at Cambridge Quantum Computing and IBM. She recalled an event for quantum computing startups that IBM hosted in 2018. She said it was a struggle to find 10 companies then but now there are more than 650 young companies working in the sector, she said.
“It’s a call to action for all these companies and individuals to learn about quantum,” she said. “Companies need to understand their plans and they need to get moving now.”
Here is a look at what you need to understand to get a grasp of the basics of quantum computing as well as how it could impact business strategy and operations.
The basics of quantum computing
Just as quantum computing brings a whole new set of business opportunities, the field has its own language and ways of operating. Many online courses about quantum computing recommend a basic familiarity with linear algebra. As Microsoft describes it in its Quantum Documentation, “Linear algebra is the language of quantum computing.” Researchers and developers use linear algebra to describe qubit states and quantum operations as well as to predict what a quantum computer will do in response to a set of instructions.
It’s also helpful to have an understanding of quantum mechanics, which is the branch of physics that describes the behavior of very small particles.
Finally, quantum computing has its own vocabulary as well. Here is a short list of a few terms to know:
Qubits: These are the 1s and 0s of quantum computing. Making and managing these objects is one of the most challenging elements of quantum computing. Some companies use superconducting circuits cooled to very cold temperatures. Others trap individual atoms in electromagnetic fields on a silicon chip in ultra-high-vacuum chambers. Yet another approach is making qubits with photons.
Superposition: This describes the combination of two states that are normally independent, such the heads side and the tails side of a coin. With quantum computing, qubits can also be in more than one state at the same time—both heads and tails simultaneously. Superposition allows quantum computers to consider many possibilities at once, as compared to traditional computers which work through one scenario at a time.
Quantum entanglement: Particles that are entangled behave together as a system. By linking two quantum computers, the interactions between the two systems can reveal information about the physical properties of each system.
The current challenge with quantum computers is keeping the qubits stable for long enough to complete a calculation.
SEE: Research: Quantum computing in the enterprise; key vendors, anticipated benefits, and impact (TechRepublic Premium)
Paul Smith-Goodson, a quantum computing analyst with Moor Insights & Strategy, said that qubits are very susceptible to noise and researchers have not mastered error correction yet.
“The qubits can sense each other when they operate, and heat and radiation affect them, which causes errors,” he said.
Smith-Goodson compared the frequency of errors on a laptop—one error for every trillion operations for example—to the frequency with a quantum computer—every 100 or 200 operations
This is where the ability to control the operations of qubits and quantum computers becomes more important. Honeywell and Intel are focused on that ability, as much as increasing the number of qubits in a quantum computer. Smith-Goodson said that the total number of qubits in a quantum computer isn’t very meaningful if the controls don’t allow the circuit to run numerous times.
“When things progress, it will be more and more important to have finer control over operations,” he said.
Smith-Goodson said it will take millions of qubits to be able to complete calculations that have meaningful results for businesses, so the industry is still a few years away from achieving the quantum advantage.
While researchers add qubits to quantum machines and improve error correction, business leaders should be using this time to figure out how the technology fits into their industry. Developing internal expertise is another important to-do, because as Ruffener of IonQ said, you don’t write quantum software overnight.
“The day it’s available, there will be people ready to go,” she said.
Early use cases for quantum computing
As companies like IBM and Microsoft provide access to quantum computing via the cloud, other companies are working on software to use the quantum machines. That’s what Zapata Computing does with Orquestra, a programming tool that lets developers invent algorithms to be run on quantum hardware.
Zapata CEO Savoie said that this abstraction layer lets businesses focus on developing algorithms without being locked into a particular device or architecture, or overhauling their software to work on new devices as they come online.
For business leaders who are new to quantum computing, the overarching question is whether to invest the time and effort required to develop a quantum strategy, Savoie wrote in a recent column for Forbes. The business advantages could be significant, but developing this expertise is expensive and the ROI is still long term. Understanding early use cases for the technology can inform this decision.
Savoie said that one early use for quantum computing is optimization problems, such as the classic traveling salesman problem of trying to find the shortest route that connects multiple cities.
“Optimization problems hold enormous importance for finance, where quantum can be used to model complex financial problems with millions of variables, for instance to make stock market predictions and optimize portfolios,” he said.
Savoie said that one of the most valuable applications for quantum computing is to create synthetic data to fill gaps in data used to train machine learning models.
“For example, augmenting training data in this way could improve the ability of machine learning models to detect rare cancers or model rare events, such as pandemics,” he said.
Savoie also said that one thing no one is talking much about is how to use quantum as part of an analytics system in the enterprise that makes use of increasingly fragmented data and the increasingly fragmented possibilities for computation and backends.
“Quantum computing needs to be part of the analytics strategy,” he said, “making it an analytics discussion removes some of the hype of this edge technology.”