Why Quantum Computing Will Not Revolutionize Technology (At Least Not Yet)

Google made an impressive announcement with Willow, its new quantum computer. The company says it solved in five minutes a problem that would take a classical supercomputer 10 septillion years. It is a real milestone. It is also great for the stock price and for keeping research funded.

But there is a detail that matters. Comparing quantum to classical computing is like comparing an airplane to a submarine. They are different worlds with different jobs. A supersonic jet wins in the air. Put it in the water and see what happens.

The test was built for the test

Willow ran a benchmark called Random Circuit Sampling, designed specifically to measure quantum performance. It is Google performing in a championship tailored for Google. It tracks progress, which is genuinely useful, but it has no real-world application yet. Quantum is amazing at very narrow problems like prime factorization, and that is exactly where its biggest risk lives too: the potential to break the encryption we rely on.

The numbers do not lie

We are at around 107 qubits. A minimally functional application needs roughly 2,048. Something genuinely useful needs millions. The first quantum computer appeared in 1998 with 2 qubits. By 2018 Intel had 49 and Google had 72. It took almost 20 years to go from 2 to 49, and every additional qubit gets exponentially harder.

We love to assume technology grows exponentially forever. It does not always. In 2000, Intel expected 10 GHz processors by 2010. More than 20 years later we are nowhere close, stopped by heat and the limits of physics.

The physical walls

• Cryogenics. Qubits need temperatures near absolute zero, around minus 273 degrees Celsius. The cooling systems cost millions and are hard to operate.
• Stability. Qubits lose their state from heat, radiation, or vibration. They hold coherence for a very short time, so you cannot run a long calculation.
• Error correction. Qubits are fragile and error-prone. Protecting the data takes many extra qubits, multiplying the resources you need.
• Scalability. Real problems need thousands or millions of qubits working reliably together, which is an enormous engineering challenge.
• Architecture. Almost every computer today runs on x86, based on a chip from 1978. To make quantum useful for daily life, we would have to rewrite essentially all of our software for a new architecture.
• Networks. Even with thousands of quantum machines, the switches, routers, and storage feeding them would not be quantum. We would still be bottlenecked by today's infrastructure.

So what is it good for

Quantum computing is the future. It is inevitable, and it will produce incredible breakthroughs on extremely specific, high-impact problems, like helping cure disease. But do not expect it to run your games or your photo editor faster. Like nuclear fusion and civilian supersonic flight, it is coming. We just do not have a clear timeline for when it arrives in everyday life.