The technology giant Microsoft claims it has made a significant breakthrough in its mission to develop commercially viable quantum computing – a technology seen by many as one that will make even AI developments look pedestrian.

Quantum computing technology represents almost unlimited computing power on a level so far beyond traditional computing that it is hard for the human mind to comprehend. The prospect of quantum computers being able to find hidden patterns in colossal data sets in lightning quick time is seen as paving the way to the kind of rapid advancements in areas like drugs discovery, materials science and our understanding of physics.

If quantum computing advocates are right, the groundbreaking technology would help us quickly solve many of the most stubborn mysteries in science. Any almost certainly many more we haven’t even yet considered.

We would become almost godlike in our ability to understand and master a physical world whose best kept secrets would be laid bare.

But how close are we to mastering quantum computing itself? Microsoft says a lot closer than we were before thanks to the recent achievements of its scientists in developing stable qubits – the quantum equivalent of the binary bits computer science has relied on so far.

Quantum computing – the final frontier towards unlimited knowledge?

Understanding the potential of quantum computing first requires a good overview of how the classical computers we currently use work. While modern microchip designs and cloud computing have hugely increased the volumes and speed at which computers can process information, the basics of how it is done have not changed since the very first computers.

Classical computing relies on bits, which hold information based on a binary system: each bit is either 0 or 1. Think of it like a light switch that can be either on or off. Data in this system is processed through binary logic operations, such as AND, OR, and NOT.

When a classical computer performs calculations, it does so bit by bit. For example, if you wanted to add two binary numbers, the computer would start at one end and add each pair of bits together, carrying over any values as needed. The process is similar to how you might perform long addition on paper.

This step-by-step method of computation works well for many tasks, but it can be relatively slow for very large or complex problems. Classical computers must process all possible combinations of bits one after another, which can be time-consuming for problems that involve a lot of data or require exploring many possible solutions, even with the most sophisticated modern chip designs.

Quantum computers use qubits that can simultaneously represent both 1 and 0

Unlike the binary bits used by classical computing systems, quantum computing uses qubits. Mirroring the complexity of the subatomic world and possessing a quality called superposition, qubits can be 1 or 0, or any proportion of both, at the same time. In this way, they can mirror the complexity of the subatomic world.

Another quantum principle, entanglement, lets qubits influence each other instantaneously, regardless of distance, allowing complex computations to be performed simultaneously.

So, quantum computers can potentially process enormous amounts of data and solve complex problems much faster than classical computers.

Work on quantum computing research by companies like Microsoft and IBM started as far back as the 1990s. They’ve since been joined by representatives of later generations of tech giants including Google. A huge amount of academic research in quantum computing has also been conducted over the years.

However, until relatively recently the majority of quantum computing research has remained at the level of theoretical physics.

What’s held quantum computing back so far?

Decades of quantum computing research have resulted in some progress. Scientists at IBM, Microsoft and academic researchers have, at great effort, successfully managed to create qubits and hardware that can interact with them.

However, scientists have struggled with a number of fundamental problems that have proven a bottleneck to progress beyond proving that quantum computing is theoretically possible.

One major challenge is that quantum states are extremely fragile and prone to “decoherence.” This means that external disturbances, like heat, electromagnetic radiation, or even cosmic rays, can easily disturb qubits, causing them to lose their quantum properties and introduce errors into calculations.

Another issue is quantum error correction. In classical computing, error correction is relatively simple, but quantum computing’s complexity and the superposition property make it more challenging. Techniques have been developed, but they require many more qubits, often thousands, for each ‘logical’ qubit we want to protect.

Getting enough stable qubits is itself a challenge. As the number of qubits increases, managing and maintaining their quantum state becomes exponentially more difficult. Many technologies are being explored to create qubits, including superconducting circuits and trapped ions, but each has its advantages and limitations.

Lastly, there’s a significant problem of ‘readout’ error, where the process of measuring the qubits (which is how we extract computation results) can disturb their state and introduce more errors.

What does Microsoft claim its quantum computing researchers have achieved?

In late June, Microsoft chief executive Satya Nadella announced that the company’s physisists had made a long pursued discovery that would allow them to make more reliable qubits. The announcement came after Microsoft scientists published the peer-reviewed results the company says demonstrates proof of the underlying science required to create a new kind of more stable qubit.

Microsoft claims to have shown the ability to create and control an elusive quantum mechanical system known as a Majorana particle.

Most research into qubits is focused on minimising, correcting and compensating for their errors. While good progress has been made in this direction, most notably and recently by  IBM, Microsoft has taken a different approach – one that most scientists see as more difficult but potentially more promising – Majorana Zero Modes (MZM).

Until now, mathematics has told scientists that MZMs should exist but physicists have been unable to pin them down.

MZMs can only exist in very specific conditions – at a hundredth of a degree above absolute zero, in a magnetic field 10,000 times the Earth’s, in a wire a thousandth the width of a human hair. They are split electrons which hold information in two self-reinforcing halves separated by an almost incomprehensibly thin wire.

They are error resistant on the same principle as the double-entry ledger system used to prevent accountancy errors.

Scepticism remains

Despite the bold claim based on peer-reviewed research, the wider scientific community remains unconvinced that Microsoft’s breakthrough will hold up to further scrutiny.

Some of that scepticism stems from the fact that this is the second time Microsoft has made such an announcement. In a 2018 report in the journal Nature, the tech giant claimed to have found evidence of a Majorana particle. However, a subsequent a review found that key data had been omitted.

Other previous announcements that MZMs had been found and created have also had to be retracted after different explanations for the data produced by researchers were uncovered.

For now, the expectation is that Microsoft’s claims will be similarly refuted.

But even if they are, progress towards quantum hardware that can be applied to a range of real life computational problems classical computers struggle with, or are far beyond them, is being made. And at pace.

When it does happen, the outcomes will be revolutionary.

In cryptography, quantum computers could crack codes that stump classical machines, but they could also help develop new, unbreakable encryption methods.

In drug discovery and material science, they could simulate complex molecules, reactions, and materials, which is currently impractical even with the most powerful classical supercomputers. This could lead to breakthroughs like new medicines or super-efficient solar panels.

Quantum computers could also greatly speed up machine learning and artificial intelligence algorithms, enhancing their capabilities.

Optimisation problems are another obvious use case. Tasks such as finding the shortest route for a delivery truck visiting many locations or optimising airline schedules are currently hard for classical computers when the number of variables is large. But will be expected to be a walk in the park for even first generation quantum computers.

Even if Microsoft’s announcement of proving and harnessing MZMs proves premature this time, it may not be next time. Others like IBM and Google are also making progress in qubit error reduction and correction in more conventional ways.

When quantum computing does arrive, the generative AI that has caused such a commotion this year will look like a typewriter sitting beside a supercomputer.