zlacker

[parent] [thread] 36 comments
1. loceng+(OP)[view] [source] 2022-02-08 17:17:44
I ask this every so often during threads regarding stolen cryptocurrency:

Is there any solution yet to preventing stolen cryptocurrency funds from being spent? Isn't the only solution to have a central database and require laws to require every transaction to be pre-checked to see if it's stolen funds or not?

And not only that, the centralized system will have to be constantly keeping track of wallet mixing to see where funds are being redirected to, attempted to being washed to?

replies(8): >>throwa+g1 >>optima+x3 >>pcthro+94 >>rlpb+t4 >>runeks+v4 >>mdoms+76 >>Alexan+I7 >>Alexan+x8
2. throwa+g1[view] [source] 2022-02-08 17:22:08
>>loceng+(OP)
Most cryptocurrencies are designed such that the concept of "returning" stolen funds is not really meaningful.

The best analogue is cash. If you want to return the cash you need to physically find it and move it back.

replies(3): >>lupire+b2 >>knorke+23 >>loceng+aE
◧◩
3. lupire+b2[view] [source] [discussion] 2022-02-08 17:25:38
>>throwa+g1
Sovereign law isn't so easily waved away, unless you are seasteading.
replies(1): >>throwa+T2
◧◩◪
4. throwa+T2[view] [source] [discussion] 2022-02-08 17:28:07
>>lupire+b2
I'm not waving it away, this isn't a discussion about the legal system.

If someone steals steal a fiver from your back pocket then there's no magical wand that the police can wave that teleports the cash back into your hand. They need to come and get it from the kitchen table or wherever the thief has put it assuming they haven't spent it.

Most cryptocurrencies are explicitly designed to act as digital cash in this way. The system is structured such that a coin is fully under the control of the owner of the private key, there is no third party involved to effect some sort of return like a bank can.

If the coins are sitting in an exchange or some other custodian i.e. not exclusively under the control of the owner of a private key then you can effect this change by leaning on the exchange (in a legal sense).

The exchange is the bank, the coins are cash.

replies(1): >>knorke+Y3
◧◩
5. knorke+23[view] [source] [discussion] 2022-02-08 17:28:40
>>throwa+g1
> Most cryptocurrencies are designed such that the concept of "returning" stolen funds is not really meaningful.

Which is why these things are not features, but bugs, in cryptocurrencies. The core design principles of cryptocurrencies are actually bugs, if you think about it.

replies(2): >>throwa+e4 >>gruez+C4
6. optima+x3[view] [source] 2022-02-08 17:30:14
>>loceng+(OP)
As I understand it, the Poly Network hacker found it impossible to transfer his stolen Tether due to those coins being frozen at Tether HQ.
◧◩◪◨
7. knorke+Y3[view] [source] [discussion] 2022-02-08 17:31:36
>>throwa+T2
Cash is not a good comparison to cryptocurrency at all.

Stealing $70M, or trying to launder $5B in cash, is absolutely not the same thing as doing the same with cryptocurrencies.

$70M or $5B is a serious logistical problem to steal, hide, and launder.

$70M is 700kg in $100s.

> under the control of the owner of the private key,

Not owner, no. Temporary viewer is enough. And that's a huge difference.

replies(1): >>throwa+q5
8. pcthro+94[view] [source] 2022-02-08 17:32:19
>>loceng+(OP)
With some things, such as the stablecoin USDC, the issuer can prevent specific addresses from transferring them
replies(1): >>runeks+U4
◧◩◪
9. throwa+e4[view] [source] [discussion] 2022-02-08 17:32:27
>>knorke+23
It's not a bug, it's an explicit design decision with trade-offs.

Is it a bug that my fork can't cut like a knife? Different tools, different purposes.

10. rlpb+t4[view] [source] 2022-02-08 17:33:16
>>loceng+(OP)
What does "stolen cryptocurrency" actually mean? For example, if one person says it was stolen from them, and the holder says they legitimately acquired it, then how is any solution supposed to decide who is correct?

For fiat currency, there's usually a court system that can be used to determine ownership. Though often they explicitly exclude cash from that - if somebody legitimately acquires bank notes that were previously stolen, they can keep them and they are valid as legal tender.

For cryptocurrency, which jursidiction's justice system is going to determine whether something has a "stolen" marker or not? What if that's not recognised by a different jurisdiction, or someone else comes to the opposite conclusion?

replies(1): >>jcranm+Vp
11. runeks+v4[view] [source] 2022-02-08 17:33:22
>>loceng+(OP)
There is no solution because you would end up punishing innocent people. E.g. if a thief buys a car with their proceeds then it's not fair to punish the car dealership by confiscating the originally-stolen coins. This was decided in Scotland in 1749, cf. https://en.wikipedia.org/wiki/Crawfurd_v_The_Royal_Bank
replies(2): >>loceng+YD >>swarsh+c71
◧◩◪
12. gruez+C4[view] [source] [discussion] 2022-02-08 17:33:45
>>knorke+23
>The core design principles of cryptocurrencies are actually bugs, if you think about it.

It's a system designed around a different set of trade-offs. Calling a bug doesn't really make sense. For instance, using full disk encryption means that you lose all your data if you forget your keys. That's not an issue if you use icloud (which presumably has an account recovery process). Based on this, can you say that the "core design principles of full disk encryption are actually bugs"?

replies(1): >>knorke+rc
◧◩
13. runeks+U4[view] [source] [discussion] 2022-02-08 17:34:33
>>pcthro+94
Which is only possible because it's centralized
replies(1): >>knownj+n8
◧◩◪◨⬒
14. throwa+q5[view] [source] [discussion] 2022-02-08 17:36:01
>>knorke+Y3
Possession is probably a better word than ownership, sure.

In the same way that if someone takes your cash into their possession, they might not have legal ownership, but now they have to somehow be involved in its' future transfer (even if that's like, handcuffing them and forcing them to hand it over).

In a cryptographic system you need the key in order to do things. Whether you think it's good or bad to apply that principle to the concept of money is orthogonal to the ground reality of how it actually works.

The original Bitcoin whitepaper explicitly refers to itself as a peer to peer electronic cash system (https://bitcoin.org/bitcoin.pdf). It's the 7th word in. It's designed to operate in a cash-like manner as opposed to a referential (credit-like? not sure what the term is for this) as in a bank ledger or similar.

replies(1): >>knorke+3a
15. mdoms+76[view] [source] 2022-02-08 17:39:31
>>loceng+(OP)
Some organisations already have such systems in place. Take for example this Redditor who lost hundreds of thousands of dollars because someone some time in the past pushed his coins through a mixer.

https://www.reddit.com/r/blockfi/comments/skxiei/blockfi_hor...

16. Alexan+I7[view] [source] 2022-02-08 17:45:33
>>loceng+(OP)
Is there a need for an algorithm to efficiently check if a transaction is in a particular list?

I have a fairly good idea on how to make a very efficient algorithm for this. If there is a need for it, I would love to help in any way I can.

◧◩◪
17. knownj+n8[view] [source] [discussion] 2022-02-08 17:48:34
>>runeks+U4
A DAO could just as easily decide to do the same, but it'd need to be built into the smart contracts.
replies(1): >>pcthro+8p
18. Alexan+x8[view] [source] 2022-02-08 17:49:11
>>loceng+(OP)
what do you think about tornado.cash? It's apparently a very effective mixer. I wonder what law enforcement can do if someone ends up using tornado_cash.

There is absolutely no way of knowing if the money is good or bad. If you consider every mixed_cash as bad, you would be forced to assume that the entire cryptocurrency is bad bec of how the money flows.

replies(2): >>Mauran+vh >>loceng+RE
◧◩◪◨⬒⬓
19. knorke+3a[view] [source] [discussion] 2022-02-08 17:54:10
>>throwa+q5
> Possession is probably a better word than ownership, sure.

I want to stress that I don't consider this a minor difference.

> In the same way that if someone takes your cash into their possession, they might not have legal ownership, but now they have to somehow be involved in its' future transfer

Yes. Cash can be stolen by a pickpocket. But two things make this not a difference in degree, but in kind:

1. You can't pickpocket $70M 2. A pickpocket can quickly hand the $100 in your pocket to an accomplice, but not to an accomplice in Bolivia.

If you want to move millions or billions in cash then you have to fill out paperwork exactly because that's how money laundering happens. Cash isn't actually easy to move, nor anonymous, at scale.

$70M is 700kg in $100s. And any legit business you show up with $1M in cash will report it, because they have to and/or because they don't want to be tried as an accomplice to money laundering.

I've had friends receive huge sums in cash, and they have reported it exactly for this reason. Enforcement against financial crime is actually built in.

I think the comparison to cash therefore is completely inappropriate, to the point where I question if it's even said in good faith.

> In a cryptographic system you need the key in order to do things. Whether you think it's good or bad to apply that principle to the concept of money is orthogonal to the ground reality of how it actually works.

In my opinion it's not "money" that's being replaced with math, but "intentions". It's not about replacing fiscal policy with math so much as replacing laws against theft and money laundering.

The definition for the features of cryptocurrencies tend to be the exact description of money laundering and tax evasion.

So if the goal is "I want to commit all the financial crimes" then yes, for those purposes cryptocurrencies have found their use cases.

replies(1): >>throwa+vb
◧◩◪◨⬒⬓⬔
20. throwa+vb[view] [source] [discussion] 2022-02-08 17:59:17
>>knorke+3a
You seem to have descended into some sort of monologue.

I hope that it's cathartic for you; I just wanted to discuss how possession works in cryptocurrencies.

My apologies for using the word ownership instead of possession, lazy language on my part.

◧◩◪◨
21. knorke+rc[view] [source] [discussion] 2022-02-08 18:02:50
>>gruez+C4
Good point.

It's a bug to the vast majority of people, but that doesn't make it a bug to ultralibertarians.

It's a bug if the goal is actually to have everyone adopt it.

◧◩
22. Mauran+vh[view] [source] [discussion] 2022-02-08 18:23:49
>>Alexan+x8
> If you consider every mixed_cash as bad, you would be forced to assume that the entire cryptocurrency is bad bec of how the money flows.

That's exactly what's happening, according to this page that was on the HN front page a few days ago: https://news.ycombinator.com/item?id=30224637

replies(1): >>Alexan+Hn
◧◩◪
23. Alexan+Hn[view] [source] [discussion] 2022-02-08 18:49:38
>>Mauran+vh
Thanks for the link.

Millions of innocent people use cryptocurrencies. Even if you assume a currency is bad, its impractical to think that millions of people are bad.

replies(1): >>elkos+ym1
◧◩◪◨
24. pcthro+8p[view] [source] [discussion] 2022-02-08 18:54:51
>>knownj+n8
This works with the smart contracts with USDC, it's just that there's an authority which has permission to update the state to deny transfers from a specific address.

This is because USDC is a centralized stablecoin (as is USDT). There are decentralized stablecoins such as UST and MIM (and I believe DAI as well).

replies(1): >>knownj+GG
◧◩
25. jcranm+Vp[view] [source] [discussion] 2022-02-08 18:58:25
>>rlpb+t4
> For cryptocurrency, which jursidiction's justice system is going to determine whether something has a "stolen" marker or not?

It's the same decision process as the normal justice system. Broadly speaking, you can analyze it as follows (for civil complaints):

1. Is there a clause in the contract that says "disputes follow XYZ jurisdiction"? Then that's the jurisdiction. (And adding such a clause is Contracts 101 material).

2. If not, then you can usually get jurisdiction based on where the offense actually happened, or where the defendant lives. The analysis can get complicated, but it's not going to meaningfully change for cryptocurrency.

3. There's also a potential for extraterritorial jurisdiction in some cases.

> What if that's not recognised by a different jurisdiction, or someone else comes to the opposite conclusion?

Well, jurisdiction really comes down to a) can you get a court to agree that it has jurisdiction, and b) can you get other people to agree to the court's orders for relief.

replies(1): >>rlpb+Ym1
◧◩
26. loceng+YD[view] [source] [discussion] 2022-02-08 19:55:46
>>runeks+v4
How exactly are you punishing innocent people? What's happening is you're rewarding thieves.

Your example doesn't fit what I am saying either. With a digital currency you can do a pre-sale trick, so you'll see the funds were stolen - and you then don't sell them the car in the first place.

We can't incentivize theft.

replies(1): >>runeks+Ei2
◧◩
27. loceng+aE[view] [source] [discussion] 2022-02-08 19:56:36
>>throwa+g1
So we pretend to treat it like cash and ignore that we can track and know exactly what digital currency is stolen? That's your proposed solution?
replies(1): >>throwa+Qe1
◧◩
28. loceng+RE[view] [source] [discussion] 2022-02-08 19:59:18
>>Alexan+x8
I don't understand all technicalities of cryptocurrencies, however you make it illegal to mix with known stolen cryptocurrencies - and then it won't get mixed; else yes, it's a problem if your ethics make you okay with rewarding criminal behaviour by allowing stolen money to be spent.
◧◩◪◨⬒
29. knownj+GG[view] [source] [discussion] 2022-02-08 20:06:37
>>pcthro+8p
Correct, though on second thought, given how slow many DAOs are to operate, the perpetrators would already be in some other uncontrollable currency by the time people voted to blacklist certain wallets.
replies(1): >>pcthro+QL
◧◩◪◨⬒⬓
30. pcthro+QL[view] [source] [discussion] 2022-02-08 20:28:56
>>knownj+GG
A DAO could have a privileged user (voted in by the DAO of course) who has the ability to blocklist specific addresses that aren't explicitly on an allow-list. Then the DAO vote could add accounts to the allow-list.

Doing so would mean the token could be transacted, except by users who are on the blocklist and not on the allow-list. And this would prevent a privileged user from abusing the power to add accounts to the block-list. Getting unblocked at the speed of DAO is less of a concern, as long as blocked account-holders can still vote with their tokens.

◧◩
31. swarsh+c71[view] [source] [discussion] 2022-02-08 22:06:29
>>runeks+v4
That Scottish decision, while still the basis for bona fide acquisition of money in the UK and US, does not yet appear to apply for cryptocurrencies.

Your Wikipedia link cites a 2019 paper published in the Georgetown Technical Law Review whose analysis (https://georgetownlawtechreview.org/wp-content/uploads/2019/...) on page 415-6 says that 2016 US v 50.44 Bitcoins (https://casetext.com/case/united-states-v-5044-bitcoins) determined "cryptocurrencies do not meet the UCC's definition of money" and thus bona fide acquisition is not sufficient to prevent the crypto from being legally seized from the possessor and returned to the original owner.

replies(1): >>runeks+Si2
◧◩◪
32. throwa+Qe1[view] [source] [discussion] 2022-02-08 22:49:09
>>loceng+aE
No, you're using words that I don't think are meaningful in the context of what a cryptocurrency is.

Assuming you can't physically track down a thief and seize control, the technical best case you can achieve with Bitcoin is to blacklist specific transaction outputs e.g. you can choose not to accept them. You can't prevent others from accepting them, but you could for example as a governmental body add them to a global blacklist of sorts and legally forbid exchanges from accepting transactions which have at some historical point interacted with those blacklisted transaction outputs.

With the use of Lightning or coinjoin or various other privacy preserving protocols you're going to end up in a situation in which you have to taint the entire coinbase (e.g. all coins) eventually; the ultimate endgame of doing that would be to "ban Bitcoin" on exchanges.

With something like Monero or ZCash there's no serial number to track in the first place so you have no ability to blacklist anything; your only option is to refuse to accept those currencies at all.

These are possible legal avenues you can go down. But _returning_ the funds is mathematically impossible without somehow gaining access to the private keys that control them.

The fact that there is no "solution" here is an explicit goal of most of the cryptocurrencies that I'm aware of. It's certainly the reason that I'm interested in the space; it's non-custodial, as cash is.

If someone steals your car, takes it abroad and you don't know where it is, it's gone. There is no solution. Goodbye car. So it goes. If I could add a mechanism that drove it back to me, I wouldn't want it for a host of reasons.

◧◩◪◨
33. elkos+ym1[view] [source] [discussion] 2022-02-08 23:37:38
>>Alexan+Hn
Do you have a estimate on people using cryptocurrencies on a monthly basis (even in orders of magnitude)
◧◩◪
34. rlpb+Ym1[view] [source] [discussion] 2022-02-08 23:40:52
>>jcranm+Vp
That decision process would lead to different people coming up with different results. For example, it's not possible for me to look for a clause in your contract with a third party. I don't even having any way of knowing if the contract you might show me is the one that applied to a cryptocurrency transaction. So if you claim that your cryptocurrency was stolen in that transaction, and your favored court agrees it has jurisdiction and concurs, I still have no way of verifying any of that.

Your decision process might work for you, but it's not meaningful for establishing consensus in a cryptocurrency.

◧◩◪
35. runeks+Ei2[view] [source] [discussion] 2022-02-09 07:44:05
>>loceng+YD
> With a digital currency you can do a pre-sale trick, so you'll see the funds were stolen - and you then don't sell them the car in the first place.

It's a race against time. As soon as the coins are sent to a new address you can't know whether goods or services were exchanged in this process and you are thus punishing a well-intentioned seller as opposed to the thief.

replies(1): >>loceng+JW2
◧◩◪
36. runeks+Si2[view] [source] [discussion] 2022-02-09 07:47:19
>>swarsh+c71
To be clear, I'm arguing that the same reasoning behind the Scottish decision is why cryptocurrencies don't have a built-in features that prevent stolen cryptocurrency funds from being spent: because it would make the currency non-fungible.
◧◩◪◨
37. loceng+JW2[view] [source] [discussion] 2022-02-09 13:46:52
>>runeks+Ei2
Isn't there a log to immediately know what wallet coins just came to to the new address? E.g. you could check the history of the new address, to see where the coins came from, before approving the transaction by first verifying it didn't come from known stolen funds?

Isn't that the prime value of blockchain - the immutable chain/record of transactions?

[go to top]