Welp, how much longer till someone building some magic to crack any modern encryption in blink of an eye.

...

tl;dr Google claimed it has managed to cut calculation time to 200 seconds from what it says would take a traditional computer 10,000 years to complete.

https://nydailynews.com/news/...

A lot longer, but we are already working on quantum resistant cryptography

Quantum safe crypto.... here we come, what a world we live in.

https://en.wikipedia.org/wiki/...

@tekashi hopefully at the same pace right

@yorda i am legally not allowed to answer that question

@tekashi and by making that statement you answer the question... ;-)

This is just cool. Yeah people will use it for bad things, but people will also use it for good things. Like calculating voxel worlds and rendering 3D faster. Not sure if those are quantum computeable though.

At the current speed we are building quantum computers, we have still a few decades time for establishing post-quantum cryptography.

I would not worry at all, and symmetric encryption will remain secure if reasonable key length are used (e.g. a 256 bit cipher would be attackable with a quantum computer lime a 128 bit cipher is now).

This press statement of Google is just to show that some problems are better to solve specific algorithms on a quantum computers (the often quoted "quantum supremacy"), but has little to no relevance to (current) cryptography.

It is just quantum advantage and not supremacy
IBM proved that it van be done in a few days on a classical computer

I thought I read an article saying Google expected to have quantum computers capable of breaking SHA256 by 2025.

@sbiewald any sources to back up that claim?

@Demolishun no, if i answer to THIS, you will know the answer so i wont

@tekashi What claim specifically?

Obviously, if there is a sudden increase of qbits in quantum computing, attacks against Public Key Crypto with current keylengths and current design would become possible with Shor's algorithm.
Fortunately the current amount of qbits in one is to low for this attack (53 qbits available of log(2**2048) = 1400 qbits required in one unit).

Attacks against symmetric crypto: They can be attacked with Grover's algorithm, making a brute force attack possible in O(sqrt(N)) instead of O(N) attack (N is the amount of possible keys, e.g. 2^bitlength), effectively reducing the keylength by half.
Of course, a high enough qbit is of course required (note: I'm not sure how high this is), but 256-Bit encryption remains secure.

As @gitlog said. IBM proved their statement to be wrong.

@lasseontherant Not entirely wrong, but over specific, as Google compared the quantum algorithm to a not optimized version for regular computers. Google's quantum computers would have been still faster, but not the the extend Google claimed.
IBM also admitted that with little advance in quantum computing in the foreseeable future, Google's quantum computers would essentially be a lot more faster in a bit more complex variant of the algorithm discussed.

@sbiewald yeah, but quantum computers dont follow moore's law afaik, theres nothing saying we dont suddenly get a 100x increase in computational power by the end of this week.

I mean at this point its all just speculation but being prepared never hurt anyone

@sbiewald yup and thus the term Quantum advantage and not Quantum supremacy

@gitlog But if we collectively "think" it has supremacy will that change the outcome? I am hoping as a side effect they learn how to create neural interfaces with quantum tech that is non invasive.

@tekashi Of course, being prepared never hurts.

@tekashi Yes moore law dosent apply to quantum but you still have to build both of them in "similar" way
You have bits there and qbits there but there is a problem with qbits - they require low temps and stable enviroment. It is very difficult to scale them in comparison to silicon.
The increase will still be exponencial but not of the base 2, i think the base is going to be smaller than 2.

@tekashi @yellow-dog big oof. Where do I sign my NDA and employment contract?

@yorda how good are you at advanced maths and cryptography