Decentralised crime fighting using private set intersection protocols

[Monster Tent]

One less reason to use google wallet....

[1714768681]

1714768681

[1714768681]

1714768681

[1714768681]

1714768681

[Mike Hearn]

You know, if you want to argue that the financial system should be completely off limits to law enforcement and "follow the money" should be abandoned, definitely feel free to do that. Heck, maybe I'll be there arguing that along side you. Although it may not seem like it, I too appreciate the simplicity that comes with that approach.

The problem is, it's going to be a really tough argument to make, because law enforcement will have lots of examples of real cases where the technique worked. They will say, what about this case, or that case. How will we do our jobs if this tool is taken away from us?

The people listening to that debate will ultimately be politicians, who are trying to decide what to do with Bitcoin. They'll want to do whatever is most likely to get them re-elected. You know, they're faced with this system, how do they respond to it? So what happens if you can't manage to convince people (the voters) that "follow the money" is a fundamental infringement on peoples freedoms?

Well, you can propose a compromise solution in which "follow the money" is still possible, but designed such that it doesn't allow global surveillance or abuses of power by governments. Politicians are often open to such compromise solutions. Look at the DMCA. It balanced the needs of the content and tech industries via the safe harbor provisions: the whole thing is a giant compromise. So there is precedent for it.


By the way, I don't work on Google Wallet.

[aric]

This is one of the first posts that makes me think twice about being on the bitcoin bandwagon to the extent I've been. The thought of cashing out some bitcoins even crossed my mind. It crossed my mind. That's all I'm saying. I continue to be skeptically optimistic and mindful.

It's destructive to promote an obvious, recurring concept that makes Bitcoin (big B) far more susceptible to collectives that set restraints and often have ulterior motives that employ force. It destroys the fungibility of bitcoins. Destroying (i.e. supplanting) a relatively freely-trading thing with a system that makes those things less free and depletes their intrinsic worth?

No, sir, I don't like it.

It would mean starting fresh. People who care about bitcoin's original intrinsic value would be forced to design a system that's substantially less susceptible to the central hierarchy of a small development crew, "feature" creep, mass pooling, and provider dependency. That would not be an easy task. ...To say the least. If, however, bitcoin ever fell to that point, it would be both a shame and an honorable endeavor.

[ldrgn]

Wow, this system has the exact same flaws as capitalism, democracy and Bitcoin's consensus model!  What a terrible idea!  Burn the witch!

[Serith]

It is obvious for a government to force merchants to use it's blacklists, just pay to a merchant with tainted coins for what seems like a regular purchase and then see if he is going to report it, use fines and so on if merchant didn't report it. And before you are going to resort to the usual argument that democracy will sort it out, Bitcoin was created in the first place because democracy doesn't work in the matter of regulating financial system, majority of the people can't make an informed decisions about the matter, otherwise we wouldn't have Ben Bernanke printing money or almost legalized theft from Cyprus bank accounts. Bitcoin can bring free market into financial system, in free market smart individuals who get it right take all the benefits even if majority think they are wrong, they will be able to invent useful services and continue their work if there won't be majority of uninformed people standing in their way.

[hazek]

You know, if you want to argue that the financial system should be completely off limits to law enforcement and "follow the money" should be abandoned, definitely feel free to do that. Heck, maybe I'll be there arguing that along side you. Although it may not seem like it, I too appreciate the simplicity that comes with that approach.

The problem is, it's going to be a really tough argument to make, because law enforcement will have lots of examples of real cases where the technique worked. They will say, what about this case, or that case. How will we do our jobs if this tool is taken away from us?

The people listening to that debate will ultimately be politicians, who are trying to decide what to do with Bitcoin. They'll want to do whatever is most likely to get them re-elected. You know, they're faced with this system, how do they respond to it? So what happens if you can't manage to convince people (the voters) that "follow the money" is a fundamental infringement on peoples freedoms?

Well, you can propose a compromise solution in which "follow the money" is still possible, but designed such that it doesn't allow global surveillance or abuses of power by governments. Politicians are often open to such compromise solutions. Look at the DMCA. It balanced the needs of the content and tech industries via the safe harbor provisions: the whole thing is a giant compromise. So there is precedent for it.


By the way, I don't work on Google Wallet.

The fundamental problem with combating crime is the definition of crime.

Who gets to decide what is a crime and who gets to decide who is guilty of it. If you're proposal is opt in and just some layer on top of bitcoin that people may opt in and thereby agree, CONTRACTUALLY with 100% explicit consent, to be governed by certain rules about what crime is and by certain people who get to decide whether they are guilt of it or not, then I'm perfectly fine with your proposal.

But correct me if I'm wrong, I just assumed you wanted this to be mandatory for all Bitcoin users, right? I would never ever agree to that. Ever. No matter how many children would need to die, I would not.

[eb3full]

The fundamental problem with combating crime is the definition of crime.

Who gets to decide what is a crime and who gets to decide who is guilty of it. If you're proposal is opt in and just some layer on top of bitcoin that people may opt in and thereby agree, CONTRACTUALLY with 100% explicit consent, to be governed by certain rules about what crime is and by certain people who get to decide whether they are guilt of it or not, then I'm perfectly fine with your proposal.

But correct me if I'm wrong, I just assumed you wanted this to be mandatory for all Bitcoin users, right? I would never ever agree to that. Ever. No matter how many children would need to die, I would not.

lol, are you being sarcastic? Almost everything in this comment is addressed in his original post. In fact, it's described very explicitly as a voluntary system.

[Mike Hearn]

hazek, please read the first post in this thread all the way through. It has an entire section on the definition of crime. And it should be clear that it's an optional layer on top. Actually I thought you would quite like it, it's a rather libertarian way to handle (the proceeds of) crime as the whole thing is more or less market based and government is not a special player.

[d'aniel]

Destroying (i.e. supplanting) a relatively freely-trading thing with a system that makes those things less free and depletes their intrinsic worth?

But clearly appeasing the authorities to a reasonable extent can also lead to Bitcoin being more freely traded.  It's pretty hard to freely trade an outlaw currency.  Let's face reality like adults.

No, sir, I don't like it.

Haha, I know where that's from

It would mean starting fresh. People who care about bitcoin's original intrinsic value would be forced to design a system that's substantially less susceptible to the central hierarchy of a small development crew, "feature" creep, mass pooling, and provider dependency. That would not be an easy task. ...To say the least. If, however, bitcoin ever fell to that point, it would be both a shame and an honorable endeavor.

Not starting fresh at all though.  Cryptocurrency would have already made a big name for itself.  The UTXO set (current spendable coins) can be forked, so use is incentivized, and there's no speculative risk of being "on the wrong chain".  The weaknesses and proper countermeasures would be much clearer.  I'm sure there's more.

Thanks for the thoughtful post.

Edit: Though I hope it wouldn't ever come to having to create a separate underground chain, it's nice to have a contingency plan in mind.

[hazek]

hazek, please read the first post in this thread all the way through. It has an entire section on the definition of crime. And it should be clear that it's an optional layer on top. Actually I thought you would quite like it, it's a rather libertarian way to handle (the proceeds of) crime as the whole thing is more or less market based and government is not a special player.

Ok, I'll read all of it.

[Carlton Banks]

In fairness, this proposal does represent a well thought through attempt at a compromise; if I had significant numbers of coins stolen from me then perhaps I'd want a taint depth equivalent to Elvis' copyright extension (till they caught the thieving bastards!).

But the fungibility issue could have more macroscopic effects: multiple blacklists, each with differing reporting/fee hiking/acceptance across the network, all operating with a single USD price? Sounds unlikely. Not only could it justify more exchange rate volatility in general, it could also cause a more significant differing in the exchange rate in a given currency depending on the way each exchange handles flags.

[Mike Hearn]

Yes indeed, increased complexity of valuing tainted coins is a problem. Probably the easiest way to solve it is just clear the taint. Exchanges already have to go through all the ID verification and AML stuff, file suspicious activity reports, etc. If you send some "hint listed" coins to an exchange, they could just file a report and then treat the coins as normal. Because they're a nexus lots of other people will accept their removal of the taint/hint.

By the way, whilst looking for info on how existing AML controls influence law cases, I discovered FinCEN publish a magazine for the banking industry that summarizes cases where police reports have been used. Here's an old one.

http://www.fincen.gov/news_room/rp/files/sar_tti_07.pdf

I think it's a pretty mixed bag. A lot of these are what you might call reporting violations - the "crimes" are meta-crimes, that is, if the AML system didn't exist then those people wouldn't have been considered criminals either. So they aren't really convincing for the value of the system. On the other hand, there are also a few real cases of fraud and embezzlement that were solved thanks to "following the money". In that report there's a case of a guy who was stealing money from a health insurance fund. Pretty nasty to screw with peoples health care in that way.

[Zeilap]

You know, if you want to argue that the financial system should be completely off limits to law enforcement and "follow the money" should be abandoned, definitely feel free to do that. Heck, maybe I'll be there arguing that along side you. Although it may not seem like it, I too appreciate the simplicity that comes with that approach.

The problem is, it's going to be a really tough argument to make, because law enforcement will have lots of examples of real cases where the technique worked. They will say, what about this case, or that case. How will we do our jobs if this tool is taken away from us?

The people listening to that debate will ultimately be politicians, who are trying to decide what to do with Bitcoin. They'll want to do whatever is most likely to get them re-elected. You know, they're faced with this system, how do they respond to it? So what happens if you can't manage to convince people (the voters) that "follow the money" is a fundamental infringement on peoples freedoms?

Well, you can propose a compromise solution in which "follow the money" is still possible, but designed such that it doesn't allow global surveillance or abuses of power by governments. Politicians are often open to such compromise solutions. Look at the DMCA. It balanced the needs of the content and tech industries via the safe harbor provisions: the whole thing is a giant compromise. So there is precedent for it.


By the way, I don't work on Google Wallet.

I find this political view rather naive, as if this good faith act of providing a compromise will count for anything in the eyes of those who don't like Bitcoin. Instead it simply provides a platform and a foothold on which to build. You cite DMCA as a good example, but ignore SOPA 

This isn't really a compromise at all,
- the regulatory side are given a tool for free which may or may not be helpful, depending on the adoption of users (assuming no legislation to make it mandatory).
- the users of bitcoin get a more complicated system where some coins are less than others, in exchange for the hope that governments don't ask for more.

[Mike Hearn]

SOPA didn't pass, did it?

You're quite right in that law making, democracy, the systems we have today are all flawed in many ways. But let me quote Churchill. Democracy is the worst form of government known, except for all the others.

I think most politicians will come to these debates without much of an opinion on Bitcoin. It's a new technology. It has benefits and costs, like any other. On one hand, maybe law enforcement is complaining about it. On the other hand, their constituents hate banks and are clamouring for a real alternative to the existing financial systems. So they're going to be looking for some kind of middle road that tries to make everyone happy, or at least, not too unhappy.

That's why it's worth thinking about reasonable proposals. Like I said, left to their own devices and if put under pressure lawmakers will come up with regulations that are completely un-Bitcoinish, because they will just automatically assume the state can be trusted, they'll impose lots of costly paperwork, etc. There won't be any real usage of advanced technology because they don't understand it.

A lot of law makers do understand this problem, by the way, they understand that the costs of regulations can be really high. It's quite common to see them propose that an industry self-regulates with actual, real laws only as a last resort. So if the Bitcoin world can show that it isn't just blowing off the whole issue of crime, that it's able to come up with interesting solutions to its own problems, that's a very strong argument for just leaving it alone.

[aric]

Destroying (i.e. supplanting) a relatively freely-trading thing with a system that makes those things less free and depletes their intrinsic worth?

But clearly appeasing the authorities to a reasonable extent can also lead to Bitcoin being more freely traded.  It's pretty hard to freely trade an outlaw currency.  Let's face reality like adults.

I'd have to respectfully disagree with that reasoning then. If the thought is that appeasing authorities is ever a route one should take just to make bitcoin more traded, it's usually inconsistent with bitcoin. I would rather spread solutions that are less dependent on central authorities and utilize more aspects of information transmission and helping others to make it more traded. It's easy to trade bitcoin already. It will get much easier and secure with more hardware wallets. The most adult way to face this, from my perspective, is to both preserve the intrinsic properties of bitcoin and to support a bitcoin-based economy -- away from the constraints of outside forces. Tainting and limiting at the majority level is one of the greatest threats to an independent bitcoin economy.

It would be like scientists someday discovering a technique to implant gold with molecular signatures or additives and then heads of centralized gold exchanges promoting this "feature", where people give more power to other collectives (e.g. uncontrollable states and others) through gold tainting, merely to be in "compliance," merely so that gold itself can be traded more by the masses. That's somewhat circular reasoning too. That in and of itself would decrease the intrinsic property of gold: a substance essentially not prone to tampering and duplication at its fundamental level. Smelting/mixing pools would form. Then to combat mixers a collective could declare that too much taint in the mix needs to be further regulated, limited, or seized -- effectively fully controlling [people and their trade] completely when it comes to gold. It's just an analogy and it applies to anything.

That betrays the original aspect of bitcoins. If people really cared about fighting crime, they'd stop funding the largest entities' (various governments, shall we say) actions and sub-entities that directly perpetrate mass, global violence and bankroll war contractors and the prison industrial complex via draconian lawmaking. That's why a lot of people were originally choosing to use bitcoins in the first place: to ethically object to violence and promote peaceful trade. Funding "criminals" is a matter of perception that's contingent entirely upon choice and consent. For instance, I would think most people here support ending the drug war. Thus, people who support bitcoin as a mechanism of free trade for those reasons (ending the drug war) are conscientiously acting.

Bitcoin itself must -- ironically, in a sense, to some -- remain free/fungible for any entity that wishes to use and exploit it.

Crime is an artifact of perception; democracy (otherwise known as mob rule) is too. It requires significant proof and authority to even begin to support systems of law, judging, and jury -- which are often wrong or morally wrong -- let alone to allow Bitcoin as a so-called-independent medium to become subject to every whim of authoritative entities.

That's why it's special: ...like gold. ...like the internet. ...like free speech. ...so long as it's freer.

No, sir, I don't like it.

Haha, I know where that's from

I was hoping someone would.  

OK, time to make a meme pic for good measure:

http://i.qkme.me/3tidx4.jpg

Not starting fresh at all though.  Cryptocurrency would have already made a big name for itself.  The UTXO set (current spendable coins) can be forked, so use is incentivized, and there's no speculative risk of being "on the wrong chain".  The weaknesses and proper countermeasures would be much clearer.  I'm sure there's more.

Thanks for the thoughtful post.

So true! That's a very good point as it wouldn't start fresh per se. There's far more understanding now of why bitcoin is/could be so valuable. There's a great movement behind anything that would preserve it or something better. Thank you.

[Carlton Banks]

Yes indeed, increased complexity of valuing tainted coins is a problem. Probably the easiest way to solve it is just clear the taint.

I'm not sure I expressed quite what I intended: I didn't mean that differently tainted outputs could be valued differently on an individual exchange. Your system seems to only reveal the tainted outputs to those that transact with them, the taint is therefore private data to these transacters and the blacklister while the taint is still being enforced. I was suggesting that this property of limited knowledge of the identity of tainted outputs in itself creates price uncertainty, as depending on the point in time that coins become listed, a given user could easily have accepted the coins before the taint was applied, and due to transaction finality, denied the opportunity to accept/hike/report/reject. This uncertainty about whether already accepted funds can have taint retrospectively applied damages Bitcoin's value storage property, and could justify wilder price swings.

[hazek]

Well... where to begin. So I've read the whole thing and here is why this is idea is unworkable (I didn't say bad.):

First, I support prevention of the violation of ownership of private property through tracking. Getting your car voluntarily protected by an agency which will monitor it's whereabouts and return it in case it's stolen or just make it so damn hard to steal and get away with it no one would even try? Yeah I'm down with that.

However Mike's idea is unworkable. For two reasons:

1) bitcoins are not unique like a physical car, they are all fungible, numbers in a ledger, you only think you can tell them apart by their trail of transactions but you can't really. It's just way too hard to know whether someone being on a blacklist actually committed that crime and it would simply punish way too many users through false positives and invade their privacy if you wanted to investigate every false positive not to mention this would cost a shit ton of bitcoins

2) it's unworkable for the same reason racism will bankrupt a business. If you are a business and you decide to discriminate against your customers, you'd need a really good monetary incentive to do so, otherwise someone else will open up the same shop and outcompete you simply because they by definition have a broader customer base. It's why banks like HSBC will launder money and it's why Bitcoin transactions can't be blocked. If the current providers of these services start to discriminate someone will open up shop and put them our of their business.


So, yeah. Instead of running after criminals, why don't we instead find ways to prevent crime? Why don't we figure out how to increase security, how to get each person personal protection, how to raise kids into better people etc. IMO that's a much more sensible route to take if one wants to reduce crime.

[Mike Hearn]

Ah, I see. Well, I suppose it depends on how often people actually end up with tainted coins. In practice, I doubt it'd be very common - you have to identify particular funds as the result of some crime and want to catch up with the owners, which is itself not that common, and then as soon as the 'bad guy' puts the money through a nexus/participant in the system they can take some action and clear the taint by whitelisting the outputs. So the actual amount of time the taint lasts should be very low unless they try and keep the coins circulating in the black economy.

How often do you think you receive the proceeds of an identified crime today? Probably never. And when you do, how often is it a crime you would personally care about? Even less common, I bet.

So the practical impact on volatility might be very small.

[aric]

SOPA didn't pass, did it?

You're quite right in that law making, democracy, the systems we have today are all flawed in many ways. But let me quote Churchill. Democracy is the worst form of government known, except for all the others.

I think most politicians will come to these debates without much of an opinion on Bitcoin. It's a new technology. It has benefits and costs, like any other. On one hand, maybe law enforcement is complaining about it. On the other hand, their constituents hate banks and are clamouring for a real alternative to the existing financial systems. So they're going to be looking for some kind of middle road that tries to make everyone happy, or at least, not too unhappy.

That's why it's worth thinking about reasonable proposals. Like I said, left to their own devices and if put under pressure lawmakers will come up with regulations that are completely un-Bitcoinish, because they will just automatically assume the state can be trusted, they'll impose lots of costly paperwork, etc. There won't be any real usage of advanced technology because they don't understand it.

A lot of law makers do understand this problem, by the way, they understand that the costs of regulations can be really high. It's quite common to see them propose that an industry self-regulates with actual, real laws only as a last resort. So if the Bitcoin world can show that it isn't just blowing off the whole issue of crime, that it's able to come up with interesting solutions to its own problems, that's a very strong argument for just leaving it alone.

The solution is to let them try.

...and to fight back with more instruments of decentralization, not less.
...and to most certainly not aid them.

SOPA didn't pass, did it?

I dare not even get started on this line of anecdotal reasoning (which, by the way, was temporarily halted specifically because people vocally resisted and rightfully shamed supporters). Yikes. Meanwhile, many of those same entities are determined to bring more convincing, slippery-tongued legislation that will eventually pass because the system is inherently dictatorial. Regarding the US, a small group of [535+1+etc] kings/queens -- [Congress + Executives] -- ruling over and deciding everyone's fate and abilities is by no definition a "democracy" either.

[stillfire]

I don't think the comparison with Spamhaus as an example of how privately run blacklists can aide a community is very accurate. Spamhaus has made many mistakes with their blacklist, but that's okay because if you ignore Spamhaus the worst that can happen is that you receive spam.

On the other hand, if you ignore a Bitcoin "hint list", the end result could be much worse. You'd be settled with the tainted coins. If the "hint list" you ignored was from your government, you might go to jail or be aggressively interrogated to track the money if you transmit that money onwards.

So it'll be in your best interest to pay attention to as many official sounding "hint lists" as possible to avoid buying tickets in the go to jail lottery.

Now an excellent point here, I think, is that prosecution of individuals based on taint is of course already possible to do today with no change to the Bitcoin protocol. Your favourite evil government could announce a list of tainted coins and say they'll line you up against the wall if you don't surrender and report them in when you receive them.

The real question is, if Mike Hearn's idea was implemented and it staved off worse coin tain systems, would that be an improvement? This question can be considered independently from the wider question of what we can do to defend against taint tracking in general at the protocol level.