Quantum computers are commercially available. Companies such as Quantinuum are selling products. IBM has developed the IBM Eagle, a 127 bit machine. China’s built Jiuzhang 2, which at the time was claimed as the fastest in the world. There are dozens of quantum languages based on a couple of main family lines. And the NSA (according to the Snowden leaks) has been developing quantum capability since at least 2014.
And the topics of “how to write a quantum algorithm” aren’t new: Shor’s Algorithm to factor large numbers was developed long before quantum computers were known to be viable. There’s a large body of work where people have already written algorithms for quantum computers that will be able to solve problems immediately once the hardware becomes available.
Any cryptographic algorithm claiming to be quantum resistant today does have a high bar to cross, as we’ve clearly only scratched the surface of what is possible with quantum computers. The algorithms presented are (mostly) trying to avoid basing their difficulty on mathematical problems that are known to have a quantum computing solution. (e.g. submit a solution that’s not based on factoring the product of two large primes.) So you’re partially correct in that we probably shouldn’t bind ourselves to a future world based on today’s promises of “quantum-proofness”.
Kinda sorta. But “less quantum-vulnerable” doesn’t trip off the tongue as easily.
Please pardon me while I geek out about this a bit…
Modern encryption depends entirely some math being easy but the reverse process being hard. For example, if I asked you to multiply 5023 x 7841, it would only take you a couple seconds with a hand calculator to discover the answer is 39385343.
But if instead I asked you: find the only two prime numbers that multiply together to produce 39385343, it’d take you several orders of magnitude longer.
Same for computers. If we want to make it take a long time for a computer, we can use bigger numbers that are harder to factor into their prime multipliers.
For years, calculations like this - easy one way, difficult in reverse - have been the basis for how the Internet communicates securely. It’s what allows a private communication to be scrambled without agreeing in advance what the secret code will be. It’s why online business exists at all; it’s why credit card numbers aren’t published more often.
We know some of the things a quantum computer will do well that a traditional computer can’t, and some of the things that are hard (like factoring a big number into a product of prime numbers) are actually pretty easy for a quantum computer. We also know some kinds of problems a quantum computer will probably offer no particular advantage over a traditional computer.
It’s true that we won’t know all the strengths of a quantum computer in advance. But where we do know they’ll be strong, there’s active research in how to come up with other ways of ensuring privacy. So “quantum-proof” might be overstating the case, but there’s a sane basis for doing the research before quantum computers enter production.
