Quantum Computers Are a Million Times Too Small To Hack Bitcoin (newscientist.com)
- Reference: 0158030873
- News link: https://hardware.slashdot.org/story/22/01/25/2243231/quantum-computers-are-a-million-times-too-small-to-hack-bitcoin
- Source link: https://www.newscientist.com/article/2305646-quantum-computers-are-a-million-times-too-small-to-hack-bitcoin/
> Quantum computers would need to become around one million times larger than they are today [2]in order to break the SHA-256 algorithm that secures bitcoin , which would put the cryptocurrency at risk from hackers. Breaking this impenetrable code is essentially impossible for ordinary computers, but quantum computers, which can exploit the properties of quantum physics to speed up some calculations, could theoretically crack it open.
>
> [Mark Webber at the University of Sussex, UK, and his colleagues] calculated that breaking bitcoin's encryption in this 10 minute window would require a quantum computer with 1.9 billion qubits, while cracking it in an hour would require a machine with 317 million qubits. Even allowing for a whole day, this figure only drops to 13 million qubits. This is reassuring news for bitcoin owners because current machines have only a tiny fraction of this -- IBM's record-breaking superconducting quantum computer has only 127 qubits, so devices would need to become a million times larger to threaten the cryptocurrency, something Webber says is unlikely to happen for a decade.
The study has been [3]published in the journal AVS Quantum Science .
[1] https://slashdot.org/~MattSparkes
[2] https://www.newscientist.com/article/2305646-quantum-computers-are-a-million-times-too-small-to-hack-bitcoin/
[3] https://aip.scitation.org/doi/10.1116/5.0073075
Now do secp256k1... (Score:3)
Hashes like SHA-256 and symmetric ciphers like AES are relatively robust to quantum computers. Traditional asymmetric cryptography is much easier for quantum computers to break, and would allow attackers to forge new transactions that move bitcoins between arbitrary wallets -- for example, the huge number of coins that Satoshi mined early on. The
Re: (Score:2)
Easier, yes. But still orders of magnitude beyond any existing quantum computer. And keep in mind, difficulty maintaining coherance scales exponentially with size.
Submitter didn't understand bitcoin or the paper (Score:5, Informative)
The paper is talking about the 256 bit ECC keys. SHA-256 doesn't secure bitcoin, it is the current proof of work method for bitcoin. Many coins change their proof of work method, it doesn't really affect the transactions. The paper talk about a real threat to bitcoin in that a large enough quantum computer could crack the 256 bit ECC keys used to spend the contents of a bitcoin wallet. Math Trigger Warning! A bitcoin wallet address is the hash of the public key. If I give you a public key and a signed transaction for the wallet with the address that is the hash of the public key, you can hash the public key to verify the wallet address and then use the public key to validate the transaction. If you have an ECC public key and a really big quantum computer you can find the private key needed to sign transactions and thus steal the contents of the wallet. Except in bitcoin you only see the public key the first time money is spent from a wallet. Assuming people in the future only use wallets once this means you only have at most 10 minutes from the time the transaction is published to the network to find the private key, create your own transaction and get some miner to mine it into the block chain before the legitimate spend is recorded in the block chain. The paper calculates a quantum computer would need 317 million cubits to do this calculation in the 10 minute window.
Re: (Score:2)
Thanks it was helpful.
Next Week: (Score:1)
Hacker cracks bitcoin with HP 35C.
Interesting, but... (Score:1)
Interesting, but the question everyone is wondering is... can we do quantum mining! /s
Correct me if I'm wrong (Score:2)
Wouldn't the cracking of Bitcoin be a relatively minor story in the shitstorm if SHA256 was broken?
Just wait until the Webb telescope finds... (Score:2)
An exoplanet made of qubits.
Break Bitcoin or Mine it? (Score:2)
If you had a computer that powerful that it approaches the power to break Bitcoin, it would definitely be more profitable to mine Bitcoin with it instead.
Even if.... (Score:2)
a quantum hacker you crack bitcoin, the majority of miners would probably agree to fix it?
Quantum Computers Are a Million Times Too Small.. (Score:1)
Bitcoin is completely secure, nothing to worry about.
something Webber says is unlikely to happen for a decade
oh. so all my bitcoins might become worthless in ten years?
Re: (Score:2)
or ten weeks.
Re: (Score:1)
Or next week. They don't need to be hacked to become worthless.
Re: (Score:2)
As nearly as I can tell from here, you may/may not have bitcoins to worry about. Want me to open the box and see?
Doublings (Score:2)
If the size of quantum computers continues to double annually as it has been, that's log2(13 million / 127) = 17 years, or 2039.
If, on the other hand, the pace drops to doubling biennially, that'd be 34 years, or 2056.
This stuff is counted (Score:2)
by order of magnitude. So, that means that SHA-256 is only “6” away from being broken.
you just need 51% of the mining to control bitcoin (Score:2)
you just need 51% of the mining to control bitcoin
moore's law (Score:1)
13 million qubits in about 32 years [1]https://www.wolframalpha.com/i... [wolframalpha.com]
[1] https://www.wolframalpha.com/input/?i=127*2%5E(X%2F2)+%3D+13000000
really pathetic article (Score:2)
Firstly yes Quantum computing that is a danger is not happening anytime in the near future, way to many problems still to overcome for any sort of viability. Regardless it doesn't need to be a million times stronger, if a computer could break encryption in a year that would be devasting, so based on the maths in the summary that is around a 280 times increase in size required just to do in a year.
So, about a decade then? (Score:2)
One of my major misgivings with crypto as a holding medium (transfer medium is somewhat of a different case) is the inevitability of "today" crypto being busted by "tomorrow" hardware and code. Even without an implementation flaw in the code or key sharing or something.
Good thing there's no precident (Score:2)
of computers getting more complex and capable by several orders of magnitude. Now if you'll excuse me I need to do some programming on my IBM 700 series mainframe.
It's a matter of time (Score:1)
I don't find this reassuring at all. We will have that kind of power someday so Bitcoin will essentially become futile in the future. Or am I missing something?
MD5 required more than heatdeath of the universe (Score:2)
MD5 required more than heatdeath of the universe. Then it was cracked, and eventually easy.
This I don't believe.
Re: (Score:2)
> MD5 required more than heatdeath of the universe.
Still does - if you're bruteforcing the full keyspace.
But you're right- that's always, and will always be the rub- cryptographic hashes (seem) to always be vulnerable to some kind of smaller-than-bruteforce keyspace search.
Re: (Score:2)
One of interests is to someday make an image that contains its own MD5 checksum. Should be doable by the average cryptographer eventually on modern hardware.