Mysterious export controls on quantum computers (opens in new tab)

(newscientist.com)

27 pointsdraazon1y ago18 comments

18 comments

Could be related to the Russian announcement, a few weeks ago, that they have built a 30 qubit computer (which may or may not be reasonably true – in the field, everybody has a different definition of "qubit", "computer" and even "have").

fsh1y ago

There's no mystery here. One country came up with some arbitrary criteria, and the other countries copied them.

HillRat1y ago

Yeah, I suspect. As the article notes, it might be related to quantum simulation limits (e.g., QISKIT can't IIRC simulate more than 32 qubits), or taken from the literature on using Grover's alg to attack S-AES, but it feels pretty arbitrary to me (though I haven't been current on quantum computing for several years now).

spacecadet1y ago

Whats mysterious? When I led a team working on some government funded encryption stuff a few years ago, everyone in gov was terrified of post-quantum cryptography. 10x a day I had to answer questions about PQKD.

Maybe some very well funded quantum projects have made certain implementations broken- but it never really mattered, because why have PQKD when you have XKCD. lol

Id still employ social engineering, deepfakes, and violence over the cost of building a machine.

By the way, we all know the Cloudflare lava lamps? I built a laser diode/beam splitter random number generator at home, fun toy.

Yoric1y ago

I've seen a recent paper that claims that they have successfully executed one (single) instance of Grover's algorithm using existing commercial quantum hardware, with lots of hypotheses and lots of manual intervention.

We'll get there, but I don't think that anybody has reasonably/reproducibly broken RSA using a quantum computer just yet.

asdff1y ago

When you think of the significance of being able to break encryption like this, it stands to reason that tech that achieves these capabilities would be born secret.

2 more replies

gnabgib1y ago

Discussions

(15 points, 2 days ago, 6 comments) https://news.ycombinator.com/item?id=40865218

(20 points, 17 hours ago, 8 comments) https://news.ycombinator.com/item?id=40879417

jzemeocala1y ago

Hmmm.... I wonder if someone has finally cracked RSA or its friends.

I also remember a conspiracy theory that Bitcoin was actually made as a litmus test to know if\when someone somewhere achieves quantum supremacy (because then they would be able to crack the block....or something like that

QuantumAlex1y ago

Best estimate I know of for cracking an interesting RSA (2048 bit) is 8 hr and 20,000,000 superconducting (physical) qubits with error rates ~ 0.1% [1].

This seems super premature.

[1] https://quantum-journal.org/papers/q-2021-04-15-433/

Hizonner1y ago

I would put a fairly large wager on it just being bureaucratic dumbassitosity, and give you odds.

sandworm1011y ago

The NSA and their ilk would not have waited for AES to be broken. They would move to ban these things in response to theoretical albeit confident assessment of the risk.

refulgentis1y ago

Hmm, that doesn't sound right to my ear.

They've loudly assumed it is possible.

c.f. Their focus has been on incentivizing private actors to do post-quantum algorithms, yesterday.

c.f. most recently, https://blog.chromium.org/2024/05/advancing-our-amazing-bet-...

Do you have more info on why they'd ban import of it? Seems like an obviously wrong strategy to combat it.

1 more reply

davidgerard1y ago

there was some crazy around bitcoin and quantum, led by Vitalik Buterin when he was young and foolish https://davidgerard.co.uk/blockchain/buterins-quantum-quest/

j / k navigate · click thread line to collapse

18 comments

Yoric1y ago

fsh1y ago

There's no mystery here. One country came up with some arbitrary criteria, and the other countries copied them.

HillRat1y ago

spacecadet1y ago

Maybe some very well funded quantum projects have made certain implementations broken- but it never really mattered, because why have PQKD when you have XKCD. lol

Id still employ social engineering, deepfakes, and violence over the cost of building a machine.

By the way, we all know the Cloudflare lava lamps? I built a laser diode/beam splitter random number generator at home, fun toy.

Yoric1y ago

We'll get there, but I don't think that anybody has reasonably/reproducibly broken RSA using a quantum computer just yet.

asdff1y ago

When you think of the significance of being able to break encryption like this, it stands to reason that tech that achieves these capabilities would be born secret.

2 more replies

gnabgib1y ago

Discussions

(15 points, 2 days ago, 6 comments) https://news.ycombinator.com/item?id=40865218

(20 points, 17 hours ago, 8 comments) https://news.ycombinator.com/item?id=40879417

jzemeocala1y ago

Hmmm.... I wonder if someone has finally cracked RSA or its friends.

QuantumAlex1y ago

Best estimate I know of for cracking an interesting RSA (2048 bit) is 8 hr and 20,000,000 superconducting (physical) qubits with error rates ~ 0.1% [1].

This seems super premature.

[1] https://quantum-journal.org/papers/q-2021-04-15-433/

Hizonner1y ago

I would put a fairly large wager on it just being bureaucratic dumbassitosity, and give you odds.

sandworm1011y ago

The NSA and their ilk would not have waited for AES to be broken. They would move to ban these things in response to theoretical albeit confident assessment of the risk.

refulgentis1y ago

Hmm, that doesn't sound right to my ear.

They've loudly assumed it is possible.

c.f. Their focus has been on incentivizing private actors to do post-quantum algorithms, yesterday.

c.f. most recently, https://blog.chromium.org/2024/05/advancing-our-amazing-bet-...

Do you have more info on why they'd ban import of it? Seems like an obviously wrong strategy to combat it.

1 more reply

davidgerard1y ago

there was some crazy around bitcoin and quantum, led by Vitalik Buterin when he was young and foolish https://davidgerard.co.uk/blockchain/buterins-quantum-quest/

j / k navigate · click thread line to collapse