I don’t know exactly how mixers work but my understanding is that that I can identify even fractions of a coin as having originated from some parent coin. All those fractions are stolen goods.
As a simple example of how this is possible: imagine you and I both have 0.1 BTC. We could simply swap wallets which would exchange tokens but that wouldn't be represented on the blockchain at all.
How would you prove to me that you don’t still have the key for your old wallet?
The only challenge is "filling" the wallet, but doable: one party starts sending small amount to the multisig wallet, then the other party evens out. Increase amount by the amount the other party paid in total, until the preagreed amount is in.
Let’s say we send 1 clean BTC to address A and 1 dirty BTC also to address A.
Are both those coins now tainted?
If we then send 1 BTC from address A to address B, and 1 BTC from address A to address C, who has the dirty coin? Is it B, C, both, neither?
So while there are technical differences between the "account" and the "UTXO" models, I doubt it would make any sense in an enforcement context.
Intent probably matters; mixing your money through a service that has been widely known to be used to facilitate and obfuscate financial crime is probably not the best look if you end up with tainted coins and their legitimate owner comes knocking on your door.
On the other hand, I don't even know how this works in detail in the traditional banking system, i.e. whether stolen money can be clawed back even if it's already moved across several accounts/transfers, the last of which completely oblivious to its source.
It is possible to say this $100 bill was once stolen so we are going to seize it from you even though you are perfectly unrelated to the crime and just happen to have ended up with it. The law treats money specially and prevents that from happening, but otherwise it could. It can happen with something like jewelry or a car.
With electronic money you don’t even have a practical option even if there was one legally. There’s no specific hundred dollar bill to seize. Banks don’t track with that kind of granularity.
The coins are fungible, but they don’t get mixed up until they share a transaction.
Do you understand the amount of computation and data analysis you would have to do to determine if a coin is "Tainted" or not.
But if there is a trusted mixer, the operation is simple: Every participant in a round of mixing tells the mixer (out of band) where they will be sending coins from, and where they want output coins to be sent to. Everybody then sends all of their coins to the mixer, who then takes all of these coins as inputs, and the out-of-band communicated addresses as outputs.
Now if 10 out of 100 input coins are tainted, the question is: Which of the 100 resulting coins are tainted? All of them? None of them? Or exactly 10 of them (but which ones)?
"All of them" would effectively amount to a ban of mixers (which I consider to be quite likely, if hard to enforce).
"None of them" means legitimizing mixers, and I doubt that US and European regulators have a lot of appetite for that at the moment.
You're absolutely not. Intentions and outcomes matter, and courts won't care if you achieved the outcome "money laundering" via on-chain code or good old face-to-face meetings and moneys in sports bags.
Note that "writing code that can also be used to facilitate money laundering", "writing code that can reasonably only be used to facilitate money laundering", "writing code that facilitates money laundering and operating it yourself (or selling it for money or a licensing fee to somebody that does)" are all different points on a spectrum.
> "All of them" would effectively amount to a ban of mixers.
Again, centralized mixers will be banned. You can plug all the exits to your centralized currency, but you cannot stop someone from trading in the cryptosphere utilizing various "Decentralized mixers" with proper ZK proofs, and headless wallet ledgers.