News for hackers. Stuff that matters.
Re:Square buys $170M worth of bitcoins by endisneigh on Tuesday February 23, 2021 @05:03PM (#26243503)
If it ends up that 40 years from now quantum computing can beat all present-day encryption methods, would the bitcoin price instantly just go to zero? Is there anyway to retroactively make it more secure with tomorrow's cryptography?

EDIT: I only ask this because if the world becomes more and more dependent on the blockchain and quantum computing becomes an existential threat, the existence of blockchain-breaking computation could potentially result in a global catastrophe. I compare quantum computing in this scenario to a device that can just magically take money away from you without your knowledge. Given that this obviously hasn't happened yet, I'm curious what mitigation strategies exist now. Thanks everyone for your replies.

disclaimer: I believe it's just a matter of time before all current (2021) encryption is broken. whether that happens in 2060 or 3000 obviously is TBD, but it's a matter of when not if IMHO.

Re:Square buys $170M worth of bitcoins by Erlich_Bachman on Tuesday February 23, 2021 @05:19PM (#26243699)
They would just switch to another algorithm. Sad for all the miners who bought ASICs, doesn't change a thing for all users and investors. It would just be a hard-fork. It would not be the first one, and it would very likely be quite a consensual one, since no user wants their transactions to be unsecured.

I really wish people would research a bit more before posting these questions here. The quantum computing argument has been asked a thousands times before, it was first seen shortly after the bitcoin paper was published, it has been answered and re-answered and re-answered so many times, everywhere. And yet here we are filling the internet with more and more copies...

Re:Square buys $170M worth of bitcoins by Retric on Tuesday February 23, 2021 @05:30PM (#26243845)
The issue is the private keys become vulnerable not just the hashing algorithm. If 3/4th of Bitcoins are in lost wallets in 1,000 years those instantly flood the market once someone can discover them from the old public keys. You can’t distinguish between someone that’s kept a wallet inactive for all this time and someone discovering a private key.

Which means you need to fork early and abandon any wallets that don’t do a transaction with the new system. That’s not the kind of transition that works with just the miners involved.

Re:Square buys $170M worth of bitcoins by Erlich_Bachman on Tuesday February 23, 2021 @05:35PM (#26243893)
You can trivially distinguish those transactions from real ones. In fact so trivial that you could do it in one if clause

    #define LATEST_BLOCK_BEFORE_EMERGENCE_OF_QC 76283747
      # fill in number when that happens

    if (block_height>LATEST_BLOCK_BEFORE_EMERGENCE_OF_QC) {
       # Treat transaction as invalid if it still uses SHA256 
    }
You know transactions are timestamped through the medium of being included in a block, right?
Re:Square buys $170M worth of bitcoins by Retric on Tuesday February 23, 2021 @05:40PM (#26243952)
You don’t understand, this has nothing to do with SHA256 for mining after crypto. You need to sign a transaction with a private key. Quantum crypto makes all current private keys public.

This means you need to pick some date where everyone with a private key needs to create a new key under a new system. Then before quantum crypto is available they need to make this transaction from their old wallet using an old key to a new wallet with a new secure private key. And they need to do this before quantum crypto is available.

Re:Square buys $170M worth of bitcoins by Erlich_Bachman on Tuesday February 23, 2021 @05:43PM (#26243995)
Yes? That's exactly what they'll do.
Re:Square buys $170M worth of bitcoins by Retric on Tuesday February 23, 2021 @05:49PM (#26244056)
How and when? People are working on quantum crypto right now and as far as anyone know someone could literally break Bitcoin tomorrow February 24th 2020. It seems unlikely, but the same could be said for every day until it happens.
Re:Square buys $170M worth of bitcoins by dvh on Wednesday February 24, 2021 @04:19AM (#26248156)
Current record for quantum factorisation using Shor is 21, done by IBM in 2019.
Re:Square buys $170M worth of bitcoins by Retric on Wednesday February 24, 2021 @07:32AM (#26249191)
Current public record, NSA etc don’t publish their research.
Re:Square buys $170M worth of bitcoins by athriren on Tuesday February 23, 2021 @08:05PM (#26245311)
hopefully tomorrow is in 2021, regardless of time zone or day and month.
Re:Square buys $170M worth of bitcoins by Retric on Tuesday February 23, 2021 @08:26PM (#26245478)
Common misconception, real programmers use zero indexed years. None of this 1BC/BCE to 1AD propaganda people keep sprouting. ;-)
Re:Square buys $170M worth of bitcoins by xiphias2 on Tuesday February 23, 2021 @07:03PM (#26244778)
Shor's algorithm can be used for solving the discrete logarithm problem significantly faster (digital signing), but the miners are safe, so only a soft fork is needed.

Work is on the way to standardize a quantum-safe digital signature, but it takes years to get the best solution, and right now there's no time pressure.

Re:Square buys $170M worth of bitcoins by reidjs on Tuesday February 23, 2021 @05:03PM (#26243511)
If that happens I’m pretty sure we have much larger problems than someone stealing your bitcoins
Re:Square buys $170M worth of bitcoins by dzader on Tuesday February 23, 2021 @05:15PM (#26243641)
when*
Re:Square buys $170M worth of bitcoins by akvadrako on Tuesday February 23, 2021 @06:03PM (#26244210)
No, if. Scalable quantum computers might be physically impossible or close to it. That would explain why the world, which is really quantum, always appears classical.

Quantum computers have something similar to moore's law which is that adding each bit is as difficult as all the previous bits. i.e. the required noise floor scales exponentially, unless a certain threshold is reached.

Re:Square buys $170M worth of bitcoins by RIMR on Tuesday February 23, 2021 @05:14PM (#26243634)
Strong encryption turning into weak encryption is a severe problem, but it's a manageable and resolvable one also. The ease of exploiting quantum encryption-breaking might not be comparable to exploiting a public ledger of trust like the blockchain.

If BTC has a multi-trillion-dollar market cap, and quantum exploitation occurs, the only thing that backs the currency (trust) vanishes, and the values quickly drops to zero.

If we are reliant on BTC for global finances in the future, as many are proposing, this could result in a global depression.

This might be the single biggest problem quantum computing could bring us, short of the "talking to alternate universes" theoretical use cases some people are talking about (that probably won't happen).

Encrypted data can be re-encrypted to be quantum-secure. The vulnerability is temporary

A compromised blockchain dies forever. A brand new quantum-secure blockchain would need to be invented.

Re:Square buys $170M worth of bitcoins by Erlich_Bachman on Tuesday February 23, 2021 @05:20PM (#26243709)
Look up the term "fork" in the context of cryptocurrencies. It will answer all your questions and you will save yourself a lot of brain cycles.
Re:Square buys $170M worth of bitcoins by ASalazarMX on Wednesday February 24, 2021 @01:09PM (#26253342)
"Guys, we are going to start quantum exploiting your blockchain for profit in a couple of months, please fork before that XOXO" - No criminal ever

More likely: "We'll skim from dormant wallets that won't be immediately noticed, and when it becomes too obvious, we move all of Satoshi's bitcoins to force an emergency fork that launders our profits".

Re:Square buys $170M worth of bitcoins by RIMR on Wednesday February 24, 2021 @10:39AM (#26251024)
How do you migrate the old keys to a quantum fork without allowing the exploit to still exist retroactively?

I could exploit the old blockchain, then migrate my stolen keys to the quantum blockchain.

Forking doesn't work here. Traditional computing and Quantum computing are different enough that "porting" isn't just something you can do.

Re:Square buys $170M worth of bitcoins by echelon on Tuesday February 23, 2021 @05:28PM (#26243821)
If crypto is broken, any fork could be broken too.

It's a question of how much time a new algorithm buys.

Re:Square buys $170M worth of bitcoins by danaur on Wednesday February 24, 2021 @05:22AM (#26248499)
There are quantum resistant methods forks can use.
Re:Square buys $170M worth of bitcoins by RIMR on Wednesday February 24, 2021 @10:40AM (#26251035)
But my old keys, that hold my funds, are still traditionally encrypted, so how am I supposed to securely claim my same bitcoins on the other side of the fork?
Re:Square buys $170M worth of bitcoins by tomerico on Tuesday February 23, 2021 @05:28PM (#26243828)
Bitcoin is vulnerable in two ways:

1. The oldest (and largest) wallets directly stored their public key on the blockchain. This would mean that attackers could transfer money away from these wallets (many of them are dormant).

2. Relatively early, the blockchain shifted to only storing the hash of the public key, which seems less susceptible to a quantum computing attack. The risk in these wallets is that once you want to transfer bitcoins out, you need to temporarily publish your public key - this key is not stored on the blockchain, but might be stored by future attackers. In addition, at some point getting your private key will be so fast that the moment you publish a transaction request you'll many fake requests asking to transfer to another wallet.

The solution is likely a soft fork - new addresses that are quantum resistant (similar to how in the past Bitcoin moved away from public key address into hash). The trick is that this needs to be deployed before attacks exist, and people need to transfer their Bitcoin into the new wallets.

Re:Square buys $170M worth of bitcoins by Melting_Harps on Tuesday February 23, 2021 @05:37PM (#26243915)
> 1. The oldest (and largest) wallets directly stored their public key on the blockchain. This would mean that attackers could transfer money away from these wallets (many of them are dormant).

That doesn't mean the Public key has ever been exposed online and therefore not true. Look up cold storage best practices, I'm not going to line it out for you but what you said is entirely false and built on a flawed premise.

> 2. Relatively early, the blockchain shifted to only storing the hash of the public key, which seems less susceptible to a quantum computing attack. The risk in these wallets is that once you want to transfer bitcoins out, you need to temporarily publish your public key - this key is not stored on the blockchain, but might be stored by future attackers. In addition, at some point getting your private key will be so fast that the moment you publish a transaction request you'll many fake requests asking to transfer to another wallet.

You started it off with a valid observation, but then led into a non-sequitur, what does that have to do with Bitcoin's vulnerability if as we just said 1 is not true if done correctly.

> The trick is that this needs to be deployed before attacks exist, and people need to transfer their Bitcoin into the new wallets.

While I agree security on mainchain is an issue, we should dhave had taproot long ago, and mixing by default by now as tx fees get more and more expensive--all you're doing is mixing it with other adddresses to make the sum indistinguishable from it's source and that cost adds up as the netowrk gets more expensive to operate on.

But also your premise is at odds with the fact that most of the traffic will be done on LN moving forward, and the bitcoin (token) will locked out of the network's mainchain (Bitcoin) to operate on LN and thus does not require your 'solution.'

In short, these are non-concerns.

Note how I didn't downvote you, and instead I challenged your arguments in order for others to benefit from this conversation, instead of being a child and downvoting simply because I disagree with your points.

Edit: HAHA!

Re:Square buys $170M worth of bitcoins by mudlus on Tuesday February 23, 2021 @06:16PM (#26244357)
If that happens you're worried about Bitcoin? lol

Besides, the algorithm in question here is Shor's algorithm, it WILL be used to break some basic encryption. If Satoshi is dead, his coins will be hacked (they are old P2PK outputs). But there is no algorithm for breaking P2SH outputs (eg Native Segwit wallets). There could be, but Bitcoin has the ability to adapt if necessary. Bitcoin is not a static thing.

See also : https://www2.deloitte.com/nl/nl/pages/innovatie/artikelen/qu...

More importantly, quantum computers actually allow for something better than encryption. If you are having a secret conversation over the quantum internet, not only can no one listen in, BUT you will KNOW if they are listening. I read about this in David Deutsch's "Fabric of Reality," which I HIGHLY recommend and is freely available on archive.org https://archive.org/details/TheFabricOfReality/page/n1/mode/...

https://en.wikipedia.org/wiki/Quantum_cryptography#Mistrustf...

Re:Square buys $170M worth of bitcoins by trident5000 on Tuesday February 23, 2021 @05:14PM (#26243632)
Vitalik Buterin himself said sha256 is likely not hackable by quantum computing, certainly not any time soon. If it is then consensus or a fork is needed to change the security.
Re:Square buys $170M worth of bitcoins by nickysielicki on Tuesday February 23, 2021 @05:22PM (#26243744)
Just because he developed ethereum doesn't mean he has any idea what is going on in quantum computing. Whatever we've seen out of the private sector in terms of quantum computing is a total joke in the context of black budget NSA programs.

Nobody knows except the people doing it.

Re:Square buys $170M worth of bitcoins by wmf on Tuesday February 23, 2021 @05:26PM (#26243793)
The whole field of quantum computer science would like a word. It's possible to predict what quantum computers will and won't be able to do.
Re:Square buys $170M worth of bitcoins by akvadrako on Tuesday February 23, 2021 @06:11PM (#26244300)
Not really. There are no proofs that SHA can't be broken by efficient classical means, let alone quantum means. Once the incentive is there people will start looking harder.
Re:Square buys $170M worth of bitcoins by trident5000 on Tuesday February 23, 2021 @05:27PM (#26243803)
Quantum computing is not incalculable. Solving sha256 by traditional means would need the energy of the sun. Quantum will only further the goal by so much. And Vitalik didnt just invent ethereum he was/is a cryptography expert who wrote many publications.
Re:Square buys $170M worth of bitcoins by svachalek on Tuesday February 23, 2021 @05:46PM (#26244018)
Quantum is not just faster, it changes the problem from being sub-exponential to logarithmic.

https://en.wikipedia.org/wiki/Shor%27s_algorithm

Re:Square buys $170M worth of bitcoins by pixelperfect on Tuesday February 23, 2021 @06:49PM (#26244655)
not sha256
Re:Square buys $170M worth of bitcoins by bottled_poe on Tuesday February 23, 2021 @05:22PM (#26243727)
Wow not a crypto founder himself. He wouldn’t have a conflict of interest would he?
Re:Square buys $170M worth of bitcoins by Erlich_Bachman on Tuesday February 23, 2021 @05:32PM (#26243866)
Vitalik is not only extremely smart, has many times demonstrated his abilities, but has also shown himself to be incredibly honest and straight-forward kind of guy when it comes to public utterances. Can you find even a couple of times when he has misled or lied to the public, even a bout a small thing? He comments on technical issues and engineering problems, things he knows a thing or two about.
Re:Square buys $170M worth of bitcoins by jude- on Tuesday February 23, 2021 @11:31PM (#26246586)
He once tried to sell investors on using a classical computer to simulate a quantum computer to mine Bitcoin faster [1]. Granted, this was before Ethereum.

> He comments on technical issues and engineering problems, things he knows a thing or two about.

He still doesn't seem to truly understand BFT [2].

[1] https://davidgerard.co.uk/blockchain/buterins-quantum-quest/

[2] https://news.ycombinator.com/item?id=25984251

Re:Square buys $170M worth of bitcoins by jackson1442 on Tuesday February 23, 2021 @05:04PM (#26243525)
Presumably there could be another fork like there was with BTC vs Bitcoin Cash (BCH).
Re:Square buys $170M worth of bitcoins by nickysielicki on Tuesday February 23, 2021 @05:17PM (#26243678)
If the discrete logarithm problem turns out to be not so difficult, the entire concept of proof of work is meaningless. There is no patch or fork that can possibly fix that.
Re:Square buys $170M worth of bitcoins by Erlich_Bachman on Tuesday February 23, 2021 @05:22PM (#26243730)
That's just not correct. There are plenty of crypto algorithms already, including hash functions, that are designed to be resistant/resilient against a functional quantum computer. A PoW scheme can use any (well, at the very least many) of them.

[1] https://en.wikipedia.org/wiki/Post-quantum_cryptography

Re:Square buys $170M worth of bitcoins by darawk on Tuesday February 23, 2021 @05:19PM (#26243697)
Proof of work does not depend on the DLP. It depends on SHA.
Re:Square buys $170M worth of bitcoins by nickysielicki on Tuesday February 23, 2021 @05:55PM (#26244126)
PoW is just a filter to prevent denial of service attacks, to ensure that if you're a bad actor, you'll waste electricity to send bogus messages.

But if the DLP is easy, then anyone can fabricate proper ECDSA sigs, and you're not wasting electricity to send bogus messages, you're wasting electricity to send valid messages.

The point I was trying to make is that there is no migration path from ECDSA. You can't fork bitcoin if the DLP is easy. It's just game over. Proof of work becomes irrelevant.

Re:Square buys $170M worth of bitcoins by arcticfox on Tuesday February 23, 2021 @05:05PM (#26243545)
In short, yes it's possible to secure, but it'd be a project. Like securing everything else against QC

https://en.bitcoin.it/wiki/Quantum_computing_and_Bitcoin

Re:Square buys $170M worth of bitcoins by jude- on Tuesday February 23, 2021 @05:23PM (#26243750)
Yes -- the network could soft-fork to require post-quantum transaction signatures. Naively, you could make it so that each transaction would put the hash of its parties' post-quantum signatures in an OP_RETURN output, and distribute both the transaction and the post-quantum signatures to the network. Miners would package the transactions into blocks, and separately validate them with the associated post-quantum signatures and distribute them as ancillary block data (e.g. the coinbase transaction could contain a Merkle tree root that authenticates all the block's extra post-quantum signature data).
Re:Square buys $170M worth of bitcoins by Vecr on Tuesday February 23, 2021 @05:05PM (#26243534)
No, only the signatures would be at risk, and the fix would be quite simple. Create a new address type that's quantum-secure, then have everyone move their coin over. After practical breaks start happening, disable the old signature type.
Re:Square buys $170M worth of bitcoins by kube-system on Tuesday February 23, 2021 @05:12PM (#26243608)
I would hate to be a scientist working on quantum computing when there is so much wealth to be threatened by their success. Sounds like a quick way to meet the mafia.
Re:Square buys $170M worth of bitcoins by missedthecue on Tuesday February 23, 2021 @05:20PM (#26243707)
I think the threat would be less from billionaires and big companies. Their fortunes get disrupted, accumulated and lost all the time. If such a threat to those researchers exists, it is likely from intelligence agencies around the world.
Re:Square buys $170M worth of bitcoins by chemmail on Tuesday February 23, 2021 @05:18PM (#26243689)
Quantum computing gives near infinite parallelization. It does not solve/break serialized equations/problems.
Re:Square buys $170M worth of bitcoins by dylkil on Tuesday February 23, 2021 @05:10PM (#26243587)
yes, you can fork the network to use a new hashing algorithm
Re:Square buys $170M worth of bitcoins by RIMR on Tuesday February 23, 2021 @05:09PM (#26243572)
If Bitcoin doesn't destroy itself somehow, quantum computing will be the destruction of the modern blockchain. Quantum Bitcoins might be possible, but developing a quantum blockchain before traditional Bitcoin can be exploited by the same tech is highly unlikely, so any Bitcoins today will almost certainly never become quantum Bitcoins.

I don't think there is any feasible way you could ever convert the Bitcoin blockchain into a quantum variant, even if you set a hard sunset date before any exploitation could occur. You would have to start from the ground up and completely reinvent the wheel.