5 Misconceptions about quantum computers

• Quantum computers calculate faster than any computer before¹
• How quantum computing will change the future of the financial sector²
• Expert warns: Quantum computers can “undermine any blockchain security system”³

Some of these headlines are partly true, others are exaggerated. Below, I will try to address and clarify the five most common misconceptions. Although there are other misconceptions⁴, I think these points are the most commonly misrepresented.

1. Quantum computers are faster than normal computers

The story usually goes like this: normal computers use bits that can only represent 0 or 1. Quantum computers, on the other hand, use qubits, which have two interesting properties:

• The first is superposition, which allows qubits to be in the 0 and 1 state at the same time.
• The second effect is that qubits become interlocked and can influence each other. This effect is also known as entanglement.

These entangled qubits in a superposition state can (simplified explanation) calculate all input possibilities simultaneously and therefore behave like a classical computer with an infinite number of processors.

So much for the general explanation, but it ignores an important point: quantum computers, due to their physics, cannot deliver a finished result on the first computational run. Calculations must be carried out more frequently – often many thousands of times – before a reliable result is available. It can therefore happen that a quantum computer can solve a problem faster, but the necessary calculation runs take more time in total.

The advantage of quantum computers is not necessarily that they are faster than “normal” computers. The advantage lies in being able to solve problems at all. And that brings us to the next point.

2. Quantum computers will replace classical computers

A quantum computer can solve certain problems much faster than a normal computer. Quantum computers can determine solutions for which classical computers, such as those at home or even in computer centres, are not suitable. For this purpose, computer science divides problems into different complexity classes.

Complexity classes basically describe how difficult a problem is to solve. The most well-known classes among computer scientists are probably the P and the NP (there are of course many more, but I will omit the details here). Complexity class P stands for problems that can be solved relatively easily by programs. Complexity class NP contains problems that cannot be solved deterministically in finite time. Here one has to test and verify different possibilities to find a valid result. Basically by trial and error.

Tasks from class P can be solved excellently with conventional computers. These are, for example, tasks such as writing an e-mail, surfing the Internet, playing Wordle or even training a model of an AI.

Current computers fail with problems from the NP class, since the computational effort here quickly exceeds the available computing capacities. Thanks to the qubits, quantum computers can, for example, solve the problem of the travelling salesman⁵ or the problem of packing a backpack⁶, which classical computers have always struggled with. Quantum computers are therefore a supplement rather than a replacement for classical computers.

3. Quantum technology is currently only theory

Quantum technology is no longer just grey theory or pure science fiction. Many companies already offer access to their various quantum computers, including many well-known ones such as IBM, Amazon, Rigetti, D-Wave etc.

Since the currently available quantum computers are still technically very complex and very expensive to purchase as well as to operate, they are mostly used in clouds. Designs such as those from IBM already provide 127 qubits for calculations. Systems with up to 7 qubits, which are sufficient for simple experiments and first steps, are available free of charge after registration.

In addition to systems such as those from IBM, which are also called universal quantum computers because they can solve any problem thanks to free programming, there are also quantum computers that use the quantum annealing approach.

These computers are suitable for solving optimisation problems and are currently used in finance, production, logistics and the pharmaceutical industry. Instead of prescribing a programme to be executed, quantum annealing describes the problems in mathematical formulas for which solutions are subsequently determined. In contrast to classical computers, not all possibilities are tried out one after the other, but all “simultaneously”. This means that solutions can be found much faster.

Small brothers of the big quantum computers are even available for home use. SpinQ offers small versions the size of a conventional desktop computer for sale. These currently only have 2 or 3 qubits, but it is a first step.

4. Quantum computers will destroy the internet

The internet as we know it relies largely on encryption. Every call to a website, every transfer in online banking or even the sending of chat messages uses encryption. Current encryption standards are based on a simple fact: it is easy to multiply two numbers, but very, very difficult to decompose a large number into its prime factors.

An example: imagine the prime numbers 3 and 5, which give the result 15. 15 can easily be converted back into 3 and 5 in your mind. Now try 143, and the prime factors are 11 and 13. Numbers with 256 bits are normally used in encryption, which is a number with 78 digits. This is not even difficult for a human, but also for today’s computers.

However, Prof. Peter Shor has developed an algorithm⁷ with which a number can be broken down into two factors, which classical computers cannot do very well, but which quantum computers may be able to do in the future. Nevertheless, there are two reasons why there is probably not much danger at present:

1. Shor’s algorithm would require thousands, possibly even millions of qubits, depending on the encryption algorithm. The largest quantum computer to date has 127 qubits. While IBM has presented a roadmap envisioning a Quantum Processing Unit (QPU) with over 4000 qubits by the end of 2025⁸, experts do not consider even this a threat.
2. The US National Institute of Standards – NIST – has launched a Post Quantum Cryptography (PQC) competition to find cryptographic algorithms that are secure against quantum computers.⁹ PQC is supposed to replace the encryption methods used so far step by step and thus keep communication secure. This sounds good, but it will probably take many years to select a finalist for the competition and to implement new algorithms in current programmes and operating systems.

So it is a race against time. What will happen first? Will there be a quantum computer big enough to break current encryption, or will a new algorithm be available that is safe from quantum computers?

Here, estimates of when current algorithms will become obsolete vary widely. The timing is referred to as Y2Q, short for Year-to-Quantum (borrowed from Y2K, the fear of the year turning to 2000). The Cloud Security Alliance shows a Y2Q clock¹⁰ that suggests this point in time for 2030, but one should not take it so precisely, because real progress makes an exact estimate very difficult.

5. Quantum computers are the only field of application of quantum technology

Quantum technology is the basis of quantum computers, but quantum computers are not the only field of application of this technology. Currently, it is the topic with the greatest interest, but it is not the only field of application. Quantum technology can contribute to great advances in many areas.

Example: Random number generation

Random numbers are used in many areas of information. For example, in encryption on the internet, in the lottery, in training an AI or also for calculations that require random numbers (e.g. in the Monte Carlo simulation).

Classical computers, however, are deterministic and therefore quite bad at generating “real” random numbers. They often use pseudo-random number generators (PRNG), which only produce random numbers at first glance.

Quantum technology is very good at generating random numbers because the randomness is virtually built in. Different methods are used and they can deliver actual random numbers in sufficient quantity and quality. Quantum-based random number generators (QRNG) are already widely available. For example, you can use the random numbers from the University of Canberra, Australia¹¹ for experiments or also hardware from ID Quantique¹², which offers a generator for installation.

Example: Quantum communication

Quantum communication makes use of fundamental properties of quanta: The state of a quantum can be read once, after which the quantum is “destroyed”. If a quantum is transmitted and someone reads out this quantum on the way to the destination, then it can no longer be read by the originally intended receiver. Consequently, it is possible to detect whether the transmission was intercepted or not. This possibility is not offered by “conventional” communication via the internet.

Quantum key distribution (QKD), for example, takes advantage of this. This procedure uses the technical principles of quantum communication for the secure transmission of keys (e.g. from a QRNG), which are then used for subsequent encryption. In other words, the keys that are used for the actual transmission are already transmitted in a secure manner. This, together with new algorithms, creates the basis for a secure internet.

Example: Quantum sensor technology

Quantum sensor technology describes the use of a quantum phenomenon to measure a physical quantity. Here, a wide variety of properties are used in which the smallest change in a certain measurand (e.g. gravity) can be detected.

In March 2022, scientists presented a prototype¹³ that can measure the smallest changes in the Earth’s gravitational pull. The changes can be caused by tunnels, pipelines or ruins ,and the device helps find voids without having to tear up a road to do so.

The accuracy of quantum sensing can also help when it comes to the human body. Quantum sensing could massively improve neuroimaging – the measurement of a person’s nervous system – or perhaps even become a brain-computer interface in the distant future. This could be used, for example, to control prostheses better or more precisely than by measurements using electrodes, as Neuralink¹⁴ has demonstrated. In the future, the whole thing might even work without medical intervention.

Conclusion

Google claimed years ago that quantum computers have already arrived when it announced quantum superiority with a 57-qubit machine.¹⁵

However, this has proven to be “wrong” or exaggerated because it was a specific use case that can also be computed on a classical computer. Many industry experts are still optimistic that quantum computers will come soon, some say in 5 years, but some only in 10 years or some even say never.

But even if we don’t reach the necessary stability and performance of quantum computers that quickly, the knowledge gained with current quantum computers and research is useful (e.g. quantum-inspired algorithms or concrete optimisations by quantum computers) and applications outside of quantum computing such as quantum communication, quantum key distribution or even quantum sensing are already in commercial use.

So it is certainly not a mistake to acquire the most basic knowledge. At least this way you can distinguish between hype and reality and clarify any misunderstandings 😉

Notes:

If you are interested in more information and perspectives on cloud technologies, blockchain or quantum computing, then it is certainly worth taking a look at Gottfried Szing’s blog: https://kjoo.be.

[1] Quantencomputer rechnen schneller als jeder Computer zuvor
[2] So wird Quantencomputing die Zukunft der Finanzbranche verändern
[3] Experte warnt: Quantencomputer können “jedes Blockchain-Sicherheitssystem untergraben”
[4] Common misconceptions about quantum technology and computers
[5] Problem des Handlungsreisenden
[6] Rucksackproblem
[7] The Story of Shor’s Algorithm, Straight From the Source
[8] IBM Quantum Roadmap
[9] Post-Quantum Cryptography
[10] Countdown to Y2Q
[11] ANU QRNG
[12] Random Number Generation
[13] New Quantum Sensor Sees Beneath the Beneath
[14] Neuralink: Interfacing with the Brain
[15] Quantum Supremacy, https://www.nature.com/articles/s41586-019-1666-5

If you like the post or want to discuss it, feel free to share it with your network.

Gottfried Szing has published another post in the t2informatik Blog:

t2informatik develops and modernises software as a service provider. Click here to learn more.