Quantum computing's threat to Bitcoin: Assessing progress and possibilities

The question of whether quantum computers could ever pose a real danger to the Bitcoin network has been discussed for more than ten years. When I first came across Bitcoin over 13 years ago, this was already a serious topic of debate.

A great deal has changed since those early days when I was still a clueless beginner struggling to understand the basics. Both theoretical advances and practical engineering improvements have occurred.

Two key developments have since emerged that significantly shift the odds of a working quantum computer appearing within the next decade or so. This does not automatically mean such machines will become widespread or even easily accessible to those with deep pockets.

Still, there is a real chance that several functional machines will be built in the coming years.

Error Correction Improvements

The first major advance lies in error correction. To compensate for the natural noise that comes with manipulating objects at such a tiny scale, creating a single useful logical qubit requires multiple redundant physical qubits.

Previously, the best method was surface codes — arranging many physical qubits in a grid and using some as check qubits that regularly verify their neighbors for errors in the superposition without destroying it. Every empty spot in the grid had to be filled with a check qubit.

This need for check qubits added extra overhead that could approach 1,000 physical qubits per logical qubit. The problem grew worse at larger scales because a check qubit can only monitor its immediate neighbors, so every group of qubits required equally spaced checkers.

Quantum low-density parity-check (qLDPC) codes remove this constraint. They let check qubits monitor other qubits across large distances — either through traces that connect different chip sections or by physically shifting atoms using the neutral atom design. As a result, the number of physical qubits needed for a reliable logical qubit has fallen by a factor of 10.

That is a meaningful improvement. It does not yet represent a fully operational machine, but it marks real efficiency gains in the engineering that will underpin a working quantum computer.

Progress In Proving Fundamentals

The second development addresses a deeper question: whether adding more physical qubits actually reduces overall system noise rather than increasing it. This remains theoretical for now, and it is important to remember that no fully functional quantum computer has yet performed an end-to-end computation that a classical computer cannot match.

Google ran an experiment with its Sycamore (and later Willow) chips to test the effect of adding more physical qubits. To be perfectly clear, this was not a demonstration of computation — it simply showed that information could be stored in memory without decaying.

The researchers used logical qubits made from bundles of 17, 49, and 101 physical qubits. They showed that the logical error rate — how often data becomes corrupted — dropped as the number of physical qubits increased. The test crossed a critical threshold: the logical qubit built from independent physical qubits held coherence longer than any single physical qubit it contained.

Again, this is not a leap to a fully working quantum computer that outperforms classical machines. But it is real progress in validating one of the basic assumptions that quantum computing relies on.

AI

These are not the only areas where better solutions are emerging. Artificial intelligence now plays a major role in these systems. It is used in reading and decoding information from a quantum computer — a key bottleneck for using such machines at scale.

AI also helps develop new quantum algorithms tailored to these devices. Given how AI has recently helped solve — or even disprove — major mathematical conjectures, it is not far-fetched to imagine AI enabling breakthroughs in this field as well.

The same techniques are applied to designing the physical quantum circuits built with different architectures. Finding the optimal layout for quantum gates in physical space to minimize noise at the quantum level, without creating so much empty space that it introduces latency or inefficiency, is a very complex challenge.

This factor could dramatically accelerate progress on the fundamental problems that remain.

Outlook Ahead

Ultimately, in my view, it comes down to one question: does the assumption that adding more physical qubits reduces noise hold true for actual computation and active manipulation of quantum information?

If that assumption holds, and is not experimentally disproven in the near future, I believe there is a realistic chance that a viable quantum computer will be produced within the next ten years.

Enormous resources are being poured into this problem. There is significant — though not overwhelming — progress on individual pieces of the puzzle. And if something is fundamentally possible, humans generally find a way to achieve it.

I am not saying it is time to panic. But do not dismiss the possibility.

This piece is featured in the latest Print edition of Bitcoin Magazine, The Quantum Issue.