Egregoros

@darkuncle

But what we have created is not even capable of doing what we're trying to achieve. Microsoft's latest can only hold a qubit state for 20 seconds. I don't care if someone comes out tomorrow with a quantum cpu that has a billion qubits -- it's meaningless unless they actually can old state indefinitely when actively powered.

This is my favorite quote from last year because it sums up the situation so well:

> 2^64 work that is non-paralellizable isn't a threat. 64 bits of classical security is insufficient because computers can do thousands of operations in parallel, and you can combine the effort of millions of computers. Grover's algorithm gives you a sequential complexity of 2^64, so if you have a quantum comptuer with a clock speed of 20GHZ (current quantum computers are in the khz to low mhz range), and you pretend that the quantum computer can process 14 rounds of AES per clock cycle (in reality it would be hundreds of cycles), it will take a quantum computer running for 30 years continuously to crack a single key (and if the temperature ever rises 1 millionth of a degree or the computer loses power for a nanosecond, you have to start over).

Alright so let's pretend for a second that these problems are mostly solved. We have other hurdles:

- no room temperature superconductors, so they need massive cooling

- This means we need He-3 which we have an alarming shortage of
Also need to find a way to ensure power stays as stable as stated above

- Needs to be paired with a classical CPU, remember. They don't work on their own, so for situations where we're crunching large amounts of data with them we are limited by the speed of our current busses to feed them

- Also we are in a massive hardware crunch so there's no chance anyone is going to build this soon

- We're suffering grid capacity issues, big doubt a datacenter with quantum computers will use less power than an AI datacenter so those hurdles remain

Assuming I live another 50 years, I still don't think they'll be possible by then.

@feld there are about eight hardware modalities for quantum computers right now; only some of them require superconductivity. There are also a great many real-world problems to be solved with quantum computers that have nothing to do with factoring large integers; breaking RSA or ECC is probably not even in the first five major breakthroughs that we will see. I think it’s important to keep in mind that the reality of quantum computing as a paradigm is very much distinct from whether or not someone can spoof a DSA signature or break your TLS 1.2 traffic in something approaching real time.

The problem is, we have been seeing for the last 10 years a series of exponential growth inflection points when it comes to quantum computing capability, and we usually cannot see that those were inflection points until they are in the rearview. Almost all of them were also not in areas that experts had previously predicted we would see breakthroughs.

The other problem is that transitioning the entire installed base of cryptography on the Internet to new standards is something that will take many years of effort. If we wait until we are reasonably certain that Shor’s algorithm at scale is around the corner, we have waited far too long to mitigate that business risk.

I would refer you here to Scott Aaronson’s blog post from a few weeks back; he is both well respected and not prone to hyperbole: https://scottaaronson.blog/?p=9718