Why Quantum Computers Won’t Replace Classical Computers Anytime Soon

It’s easy to understand the allure of the super-processing powers of quantum computing when you consider the explosion of data from artificial intelligence (AI), machine learning, and the Internet of Things (IoT).

IDC researchers predicted there will be more than 300 billion connected things by 2021. Business models are being disrupted overnight. Workforce diversity and empowered customers are rising while resources are getting scarce. The ability to manage and monetize large amounts of data is top of mind for leaders whose survival depends on connecting mountains of data ─ experience and operational within and beyond company walls ─ to make better, faster decisions. Quantum computing is one way of putting all this data to work.

“AI-fueled innovations like self-driving cars require tremendous amounts of computing power that’s miniature and almost invisible because it has to fit inside our cars,” said Andrey Hoursanov, lead for Quantum Security at SAP. “The trouble with further miniaturization is that smaller transistors will only work up to a point, after which quantum effects cannot be ignored.”

Industries that Could Benefit from Quantum Computers

Optimists think quantum computing will perform all tasks faster and smarter than classical machines. However, Hoursanov said that early experiments show how quantum could solve some problems faster, but not all. He mentioned transportation, navigation, and finance among the industries that could benefit first from quantum computing’s advantages. Industries like pharmaceuticals or battery manufacturing could benefit from quantum technologies even earlier, using quantum simulators before universal quantum computers become available.

“Route planning, supplier management, and financial portfolio management are places where quantum’s unique ability to quickly find the optimal solution by analyzing huge amounts of heterogeneous data would work well,” said Hoursanov. “Classical computers get overwhelmed by endless calculations when it comes to these enormous amounts of data or resort to approximations, which are little better than guess work.”

Transportation companies could map out the most cost-effective route using far more parameters than today. Procurement could more easily select the best suppliers for individualized demands. Quantum computing could help people with limited budgets distribute investments for the greatest returns. Manufacturing is another industry where quantum computers could help scale complex assembly, factoring in workers, equipment, raw materials, customer demand, or anything else relevant to producing goods, while saving time and maximizing resources.

Proof of Quantum Computing Progress

A sure-fire sign that quantum computing is moving beyond hype and experimentation will be when we see it repeatedly performing real-life tasks better than classical computing. However, the technology barriers to this remain huge.

One challenge is the inability to store data in a quantum state for reasonably long time periods. Right now, quantum data can be stored on a microsecond timescale. We need to get to hours, at least.

“By nature, a quantum bit loses its quantum properties over time and as soon as it interacts with other matter. This is called decoherence and it leads to errors,” said Laure Le Bars, project director for SAP Research & Innovation. “When this happens with classical computers, we correct errors behind the scenes. But for quantum bits, error correction technologies are not that advanced.”

Le Bars added that it’s difficult to control many quantum particles at once: “You need something much more universal and reprogrammable to manage the amount of data from advanced technologies like AI and machine learning.”

Another major obstacle is efficiently transferring data between classical and quantum computers. It’s time-consuming to convert classical computer data from places like social networks, the stock market, or internal company systems into the quantum state for processing. For now, companies would likely spend more time converting data than any benefits gained from quantum computing.

Developers Will Do the New Math

Although quantum computers may not be right around the corner, classical software developers will need to prepare for a different future, long term.

“Developers will need new skills to build software for quantum computing,” said Le Bars. “It will force programmers and technology architects to think differently about algorithmic problems than they have in the past. It’s not about more power. It’s a different approach and way of programming. Quantum computing could be good for certain problems, and developers need to begin exploring the possibilities.”

Le Bars shared that SAP is working with other industry leaders to explore quantum technology, including the company’s partnership in the Quantum Internet Alliance. While in these early days it’s difficult to predict the exact trajectory of quantum computing’s growth, the ongoing rise of Big Data means it will eventually impact future business.

Follow me: @smgaler.
A version of this story originally appeared on SAP BrandVoice on Forbes.