Quantum computing has been promised as the next great leap in computing for decades. In the past two years, money has finally started to follow the promise. Investors poured billions into quantum start-ups in 2025, and the industry’s revenue crossed $1 billion for the first time. The harder question is whether the technology is ready to justify it.

What a quantum computer is

An ordinary computer stores information in bits, each of which is either 0 or 1. A quantum computer uses qubits, which exploit the physics of very small particles to hold combinations of 0 and 1 at once, and to link their states together. For certain kinds of problem, this lets a quantum computer explore possibilities in a way an ordinary computer cannot.

The problems that could benefit are specific: simulating molecules and materials for chemistry and drug discovery, some optimisation problems in logistics and finance, and breaking some of the encryption that protects internet traffic today. For most everyday tasks, quantum computers offer no advantage at all.

The money

McKinsey’s Quantum Technology Monitor 2026, published in April, gives the clearest picture of the business.

Two features of the investment stand out. It is concentrated, with most of the money going to a few companies in very large rounds. And it has moved from public to private. A year earlier governments, sovereign wealth funds and universities supplied a third of the money. In 2025 private investors supplied almost all of it.

The revenue figure needs context. $1 billion is small next to the investment flowing in, and the industry is spending far more than it earns.

Investors put $12.6 billion into quantum start-ups in 2025. The industry’s revenue was just over $1 billion.”

Investment in quantum technology start-upsUS dollars, billions
Investment in quantum technology start-ups
ItemUS dollars, billions
2024About $2.0bn
2025$12.6bn

The 2024 figure is derived from McKinsey’s statement that 2025 was 6.3 times higher.Source: McKinsey Quantum Technology Monitor 2026. Chart: The Sidelight.

The technical problem: errors

Qubits are fragile. Heat, vibration and stray electromagnetic signals disturb them, and when that happens they make errors. Today’s machines make so many errors that they can only run short calculations before the result becomes noise.

The answer is error correction: grouping many physical qubits so they act together as one more reliable logical qubit. For a long time a basic question was unanswered: as you add more qubits, does the error rate fall, or do the extra qubits bring extra errors that cancel out the gain?

In December 2024, Google reported a result on that question with its 105-qubit Willow chip. As it grew its error-correcting groups of qubits from a 3-by-3 grid to 5-by-5 to 7-by-7, the error rate halved each time. Google described this as getting “below threshold”, the point where adding qubits makes the system better, not worse.

Google also said Willow completed a benchmark calculation in under five minutes that would take one of today’s fastest supercomputers an estimated 10 septillion years. But it was careful to say that this benchmark, random circuit sampling, has no known practical commercial use. The goal of a useful calculation beyond the reach of ordinary computers was still ahead.

The roadmap

IBM has set out a date. In June 2025 it announced plans to deliver a machine called Starling in 2029, to be built in Poughkeepsie, New York. It is designed to run 100 million operations on 200 logical qubits, which IBM says is about 20,000 times more operations than today’s quantum computers can manage.

That is a company target, not an achievement. But it sets out what the industry thinks a first useful, error-corrected machine looks like, and roughly when one might arrive.

The UK’s position

The UK has backed quantum technology for more than a decade. Its National Quantum Strategy sets out a ten-year, £2.5 billion research and innovation programme running from 2024 to 2034, and the government expects the private sector to add £1 billion.

What the evidence says

The case for quantum computing as a business rests on three things that are now visible:

  • Real money. Private investors are committing billions, and companies are paying for access and expertise.
  • Real technical progress. Error correction has been shown to work in principle at small scale.
  • Credible roadmaps. Large, well-funded companies have committed to dates for error-corrected machines.

What is missing is the thing the whole industry is built on: a commercially useful task that a quantum computer does better than an ordinary one. Until that is shown, most of the value McKinsey projects remains a forecast. Its estimate of up to $2.7 trillion by 2035 depends on the technology working as hoped.

What would change the conclusion

The industry would move from promise to business if a quantum computer solved a real problem, in chemistry, materials or optimisation, that the best classical methods could not, and if error-corrected machines arrived close to the dates promised. If the 2029 targets slip by years, as the schedule of ITER, the international fusion project, has, investors who funded the 2025 boom may grow less patient. The concentration of money in a few companies also means a setback at one of them would be felt across the sector.