JBI wrote:I think it is being over-hyped. A lab demo like this is a long way from practical application. Our encryption keys are still safe (for now).

The paper is about how their quantum computer is better at simulating a quantum circuit than a classical computer.

Given that a quantum computer is

*composed* of quantum circuits, this is interesting, but not exactly terribly surprising: fundamentally, their quantum computer is "computing" "natively" the same thing the classical computer is emulating.

And since that emulation, for a classical computer, is cost-exponential, yeah, no, it doesn't exactly "compete" very well.

But this is basically a spread-sheet comparison: They didn't actually do this. They are extrapolating the cost for a classical computer, and since the cost scales DRAMATICALLY with

*accuracy* (remember, "quantum" essentially equates to "probabilistic"!), well chose your time/memory trade-off (because you run out finite memory REAL QUICK, bam, bounded, so

*automatically* it turns into time++++++ after that point) + "hey what is 'good enough' accuracy, anyway" = get whatever you like for a headline. (10,000 years! wait a minute, that's awfully

convenient SCIENTIFIC!!!)

And that's the thing on the other side too:

*accuracy*. How "accurate" is the quantum computer? Remember,

*you have to measure it*. Box? Dead Cat? Yup, it's a dozy, and they're basically hand-waving that whole thing:

https://www.nature.com/articles/s41586-019-1666-5Google's Nature Paper wrote:Our model assumes that entangling larger and larger systems does not introduce additional error sources beyond the errors we measure at the single- and two-qubit level.

Ohhhhh....

Paper wrote:A key assumption underlying the theory of quantum error correction is that quantum state errors may be considered digitized and localized

Oh, ok then.

Paper wrote:the engineering of quantum error correction will need to become a focus of attention.

Are our assumptions justified? Yeah, I would agree that such a thing should be a focus of attention...

...but nah, let's get those numbers (10,000 years!) in the headlines first.

----

Look, I'm all for research. But here's the thing: If a lab has to play misleading PR games like this,

*is that really a good sign?*