Security Notes: Quantum Unknown: A Cautionary Tale

When automotive vehicles came on the road, they were described as “fast horses.” When the transistor replaced the vacuum tube, it was described as a “smaller switch.” And now that quantum computers are in the offing, they are being described as “faster computing machines.”

We know very well that the reality that unfolded from automotive vehicles was much more than “fast horses,” and the revolution unleashed by integrated circuitry is much more than a smaller vacuum tube. Similarly for quantum computers. They represent a bold new way to harness the enormous computing power of nature for the benefit of mankind.

When banks set up a “quantum-safe” digital-money system, or when payment companies build a “quantum-resistant” payment system, what they say is that their solution is safe against the published quantum papers. They claim resilience against algorithms that were devised in the last century.

No sooner did it become clear that quantum is so powerful, than its leading edge became veiled and hidden from public view. What is most concerning is the fact that we know very little about how effective quantum computers have become against the security of public/private keys.

Central banks around the world are moving to digital currencies. Most of them, impressed with the sustained success of Bitcoin, are opting for a Bitcoin variant that is free from the most disturbing Bitcoin attributes. These newly cast national currencies, in most cases, hold to the intellectually intriguing idea whereby a financial account is no longer a statement written in a well-protected database, but rather is a string of alphanumeric characters known to the public at large.

This public key has a matching private key, which, so the theory goes, is known only to the account holder. And this private key is all that is needed to drain the account. You don’t need a name, or picture, or fingerprint.

The builders of these currencies believe that quantum computers will not advance to threaten their design, and also that that no mathematician will be innovative to the point of discerning a deductive algorithm for this task.

Unfortunately, many central bankers, while being very smart financially, are not very well educated on cryptography. They rely on the expressed confidence of the currency designers, and are enticed to migrate the wealth of humanity out of protected databases and into public account keys.

One thing is for sure: Global terrorists are smiling beneath their hoodies. No longer do they need to assemble half a dozen atomic bombs to render the globe into ashes. All they need to do is to find this hidden algorithm. Then, overnight, they could drain all the money trusted to public keys.

The solution that BitMint and a few others promote is to achieve the desired capabilities of digital money, but with the assumption that, over the life of the system, the state of the art in computing will advance dramatically. Specifically, we say: Since the public-key account identifier does commit to its matching private key, then, over the life of the system, someone will find out how to deduce the latter from the former.

This notion applies to regular encryption, too. All the prevailing ciphertexts commit to the plaintext that generated them. So we must assume that, soon enough, someone will be able to extract the plaintext from the ciphertext. The solution is conceptually simple: Use ciphertexts that don’t commit to their generating plaintext.

A payment system, and more so a digital-money system, is an enduring and expensive project. It should be designed with a cautious attitude towards the not-yet-known capability of quantum computing.

— Gideon Samid, gideon@bitmint.com