The Quantum Corner: How the World's Most Impossible Computer Started to Work

Quantum computing has spent decades as the most tantalising promise in science: a machine that could solve problems no ordinary computer ever could, from designing new medicines to modelling the chemistry of the universe. It also spent those decades as a beautiful failure, because the very thing that gives a quantum computer its power also makes it maddeningly fragile. In the past two years, that stubborn wall finally began to crack.

A conventional computer stores information as bits that are either a one or a zero. A quantum computer uses qubits, which can exist in a blur of both states at once, letting the machine explore an enormous number of possibilities in parallel. The catch is that a qubit is exquisitely delicate. The faintest vibration, heat or stray particle knocks it out of its quantum state and corrupts the calculation. For years, adding more qubits simply added more errors, and the dream stalled.

The magic of a quantum computer and its greatest weakness are the same thing: a qubit that falls apart if you so much as look at it.

The Milestone Chased Since the 1990s

The breakthrough that changed the mood came from Google. In December 2024 the company unveiled a quantum chip called Willow, with 105 qubits, and used it to demonstrate something researchers had pursued since the 1990s: error correction that works below the critical threshold. In plain terms, when the team added more qubits to protect the information, the error rate went down instead of up. That single reversal is the whole ballgame, because it means a quantum computer can, in principle, be scaled up without drowning in its own mistakes.

For a field defined by the fragility of its building blocks, crossing the below-threshold line was the equivalent of proving a leaky boat could be made watertight by making it bigger. The result, published in a leading scientific journal, turned a decades-old theoretical goal into a demonstrated fact.

A Head Start You Can Actually Check

Google followed that with a second, headline-grabbing result. In a 2025 experiment nicknamed Quantum Echoes, Willow ran a particular calculation about 13,000 times faster than the world's most powerful classical supercomputer. What made it matter more than earlier speed claims was that the result was verifiable: independent researchers could check that the quantum machine had genuinely done something a normal computer could not practically match, rather than winning a contrived race designed to flatter it.

A Global Race Heats Up

Google is not alone. IBM is pushing toward ever larger processors and has laid out a roadmap aimed at a genuinely fault-tolerant machine later this decade, while Microsoft is chasing a different design based on so-called topological qubits, which are meant to be far more resistant to errors from the start. Governments and corporations are pouring billions into the field, because whoever builds the first practical quantum computer will hold an advantage in everything from science to security. After years of quiet lab work, quantum computing has become one of the most fiercely contested frontiers in technology.

What It Could Unlock

The prize is enormous. A working quantum computer could simulate molecules and materials directly, something classical machines struggle to do, potentially accelerating the discovery of new drugs, better batteries and cleaner industrial chemistry. It could also, eventually, break some of the encryption that protects the modern internet, which is why security experts are already racing to deploy new post-quantum codes long before such a machine exists. The same technology that could cure diseases could also crack the world's secrets, and both possibilities are driving the investment.

The Honest Timeline

A note of realism is essential. Today's quantum computers are milestones, not products. They cannot yet run the world-changing applications people imagine, and a large, fully fault-tolerant machine is still widely expected to be years away, much like commercial fusion power. The difference is that the central obstacle, uncontrollable errors, is no longer a theoretical roadblock but an engineering challenge with a proven path forward. That is a profound shift in what counts as possible.

The Progress in Brief

The recent leaps are easy to summarise:

• Google unveiled its 105-qubit Willow chip in December 2024.
• Willow achieved below-threshold error correction, a goal pursued since the 1990s: more qubits meant fewer errors.
• In the 2025 Quantum Echoes experiment, Willow ran a task about 13,000 times faster than the best classical supercomputer, verifiably.
• IBM and Microsoft are racing along different designs toward fault-tolerant machines.
• A large, practical quantum computer is still expected to be years away, but the core barrier now looks solvable.

From Impossible to Inevitable

For most of its history, quantum computing lived in the same uncomfortable place as fusion energy: forever promising, forever just out of reach. What changed in 2024 and beyond is not that the machine arrived, but that the reason it might never arrive was quietly removed. Errors, the flaw that doomed every previous attempt, now look like a problem engineers can beat rather than a law of nature they cannot. When the impossible starts to look merely difficult, history tends to accelerate, and quantum computing has just taken its first real step from one to the other.