You can never really know the future.

You can never really know the future.

The debate over whether or not quantum computers are a real threat to the Bitcoin network has been going on for more than a decade. It's a topic that was discussed in earnest 13 years ago, which is about the same time I first became aware of Bitcoin myself.

From those early days, when I was just a rookie trying to figure out what was going on up until today, there's been quite a bit of progress in both theory and engineering in the real world.

There are two important milestones that have occurred since then, which significantly affect the likelihood that a truly functional quantum computer will be created in the next decade, although this does not mean it will become widespread or accessible to those with large budgets.

However, there is a high probability that some functional machines will be produced in the near future.

Progress in the correction of errors

The first important step is error correction, to compensate for interference signals that occur naturally when working on very small scales such as this. Creating a logical qubit that actually works requires multiple physical qubits.

The most advanced method is surface codes, which are a way of grouping multiple physical qubits together into tables by using some as check qubs that "check" their neighbours in phases to ensure no errors occur within the stack. Gaps in each table need to be filled with checks.

This creates an additional burden that can reach almost 1,000 physical qubits per one logical qubit, and it gets worse as the size increases because check qubs are only checked by adjacent quarks so all groups must have detectors at equal distances.

Quantum low-density parity-check (qLDPC) codes help eliminate this bottleneck by allowing check qubits to be checked for other quantum particles over long distances across devices, resulting in a 10-fold reduction in the number of physical qubit required to create a reliable logical qubit.

That's not a small thing, even if it isn't the full-fledged machine that is getting more efficient, but rather a significant increase in efficiency in the engineering process behind producing fully functional quantum computers.

Progress in the demonstration of basic principles

The second involves a more fundamental question, namely the assertion that adding more physical qubits leads to a decrease in overall interference signals in the system rather than an increase. So far this remains only a theory and it must be remembered that until today there has never been a fully functional quantum computer which calculates what classical computers cannot do from start to finish.

Google conducted an experiment using a Sycamore (and later Willow) chip to test the effect of adding physical qubits. It should be made clear that this is not a representation of computation, but merely a depiction of storing data in non-volatile memory

They demonstrated, through the use of logical qubits made up of physical qubit groups 17, 49, and 101, that the error rate for logic - which is the frequency of data loss - decreases as the number of physical quarks increases. This test passed an important criterion whereby a logical quantum created from these physical qubs maintains coherence longer than any physical quantum composed of it.

Again, this is not a breakthrough to fully functional quantum computing which calculates what classical machines can't do but an appreciable advance in proving one of the fundamental assumptions behind quantum computers.

A to Z

This is not the only thing that we're finding better ways to solve this puzzle, artificial intelligence (AI) has become a key component in these systems. It's being used in the process of reading and decoding data from actual quantum computers, which is an important milestone for wider applications.

AI is also being used to develop new quantum algorithms tailored for these types of machines, and considering the past years in which it has helped solve important problems in mathematics (or even disrupted existing conventions), it's not unrealistic to expect that AI will lead to major breakthroughs.

AI is being used in the same way to design actual physical quantum circuits created by various architectures, and it's a really complicated problem: figuring out the most appropriate way of arranging quantum gates on physical space so that they minimize interference at the quantum level without creating too much empty space causing delays, inefficiencies, and other problems to be solved.

This is a factor that can accelerate progress in solving much-needed basic problems.

Looking to the future

In the end, in my opinion, it all comes down to one question: Is the assumption that adding physical qubits will reduce interference still valid when it comes to computation and quantum data handling rigorously?

If that hypothesis still holds and is not disproven by experiments in the near future, I think it's very plausible that a working quantum computer could be created within the next decade.

There are enormous resources devoted to this problem, there is significant progress (but not enough collaboration) in solving each individual problem and if basically there's a way of doing something people will often find it.

I'm not saying it's time to panic, but don't ignore this possibility.

This article is part of the latest print edition of Bitcoin Magazine, entitled The Quantum Issue. We share it here to give you an overview of ideas that have been explored throughout the issue

This article The Quantum Issue: You Never Really Know the Future was first published on Bitcoin Magazine by Shinobi