Is Bitcoin in trouble after Google's latest quantum computing breakthrough?

Short answer: not today. But the gap is getting smaller.

Google's new result makes one of the hardest parts of a quantum computer cheaper to build. That part is called "magic states."

Breaking Bitcoin's math needs a very large quantum computer, and these magic states are a big part of what makes it so large.

Google says the method still has to be shown at a larger scale.

In March, Google estimated the attack would need fewer than 500,000 physical qubits. Earlier estimates were in the millions.

A physical qubit makes errors all the time. To run a long program, hundreds of them are combined into one error-corrected "logical" qubit. Google's attack needs about 1,200 to 1,450 logical qubits, at roughly 400 physical qubits each.

Today, Google's Willow chip has 105 qubits. The biggest chip IBM has built has about 1,100. Even the biggest machines are roughly 450x too small.

The harder part is error correction. The most logical qubits reported so far is 96, by QuEra. The attack needs 13 to 15 times more.

IBM's roadmap targets 200 logical qubits by 2029 and 2,000 by 2033.

So if IBM hits its 2029 target, it would still be short of what the attack needs. Its 2033 target would be above it.

Google has already moved its own deadline to switch to quantum safe cryptography to 2029.

Bitcoin's proposed upgrade, BIP-360, has not been approved. Coinbase said on Oct 13 that an upgrade could take several years.

So Bitcoin is not in trouble today. No machine is close to doing this.