Thread by @queenofquanta

#QuantumSupremiority | The #Google Quantum Experiment is technologically impressive. However, the claims on Quantum Supremiority, verifiability, and the used terminology are problematic. Below 3 news articles: 1 by @electroweak, 2 by @elpais with myself & Porras, 3 with Cirac.

1. @Electroweak writes the best article I have seen on Quantum Supremiority in years. A necessary guideline to navigate the hype, and a must read before going to other news outlets.
medium.com/@emma.m.mckay/…

In short: "supremacy" is an awful term, the experiment still has flaws, Quantum Computing cant be separated from Ethics, Hype is bad for Science, and current "Quantum Supremacy" looks more like "Quantum Non-Uselessness".

https://twitter.com/electroweak/status/1187394137735270401

2. Article by @jordipc en @ElPais_tec. Interview with Google and 2 scientists, including @QueenOfQuanta. I have a brief mention, which I expand here: the Google experiment shows the advance of quantum technologies but will take years to rigorously verify.
elpais.com/tecnologia/201…

Why? Well, quantum scientists spent 33+ years closing loopholes in Bell experiments with 2 qubits. Do you seriously expect a 53 qubit experiment is just going to fly flawlessly? Sorry to break it to you *but* Ruling Out Classical Explanations For Huge Quantum Experiments Is Tough

As presently, we don't know how far the Google device is from being classically simulable:
1) the experiment is still far from what theory requires
2) the method used to characterize the noise runs in exponential-time, and
3) it requires noise assumptions

4) The current IBM benchmarks suggest that simulating the the Google quantum chip is still within reach of classical supercomputers.

Work will be needed to develop better methods to verify the chip and understand what is doing. As usual in science, we will need independent groups to reproduce the experiments. Scientists should remain cautious & skeptical, mostly *because this is our job*.

What seems clear to everyone is that a fairly-big superconducting qubit chip has been realized, that it's much bigger than before, and it seems to be entering a hard to simulate regime (but we don't know yet).

Comparing the Google Experiment to the 1st Flight to the Wright brothers is ridiculous because *planes currently fly* but *quantum computers have not flown yet*. It could reach that status one day in the future, but this is Hypothesis.

Last, the Google experiment is not doing any useful computation. It was designed this way. It's a clever "quantum Turing test" to challenge the computational power classical supercomputers. But it is only a test. +

There is not yet evidence that quantum computers will help in the production of lighter batteries, fertilizers or medicines any time soon, where "any time soon" is the next 15 years, which is the interval of time where scientists can make sort-of meaningful predictions.

PS: my opinion is of course different than that of other people mentioned in the original article.

3. Another article in @ElPais by Ignacio Cirac. We agree on a few things, disagree on others.
elpais.com/tecnologia/201…

END. Follow the #QuantumSupremiority hashtag for upcoming "quantum supremacy" memes.