[manholio]

> it is entirely up to you to either reveal your transactions to parties of your choice or keep them private forever. Keep in mind, that none of it is possible without the Tornado.cash Note and it is your responsibility to keep a record of it if you want to show the origin of funds later.

That's just a digital timestamp service, only a small part of what I'm talking about.

Mind you, I don't claim any novelty in the process I sketched above - just professing the need to have it implemented before we can treat crypto-assets as anything other than a money laundry tool.

[searchableguy]

It is more than a timestamp.

What you have proposed is technically infeasible to implement if you want to preserve properties of blockchain. It would be no different than a bank because KYC will add inherent centralization and fragmentation in the entire system.

So we need a compromise here because either way you cannot retroactively enforce KYC approved wallet to participate on-chain. If you do, it will be optional in that case the solution implemented by tornado is decent and does not require giving data to brokers who often sell it without permission.

Government can directly subpoena data from individuals when investigating them.

Any crypto to fiat ramp is already KYCed and businesses dealing with payments require KYC beyond a certain value transaction so in practice, the impact of KYC at on-boarding will be negligible. Not to mention, KYC can be easily faked by actors with significant stake in a globalised system.

Furthermore, the data does not support that money laundering is a bigger problem as percentage of transactions in crypto than traditional finance. The AML laws have been largely ineffective in practice in traditional finance with huge cost of compliance and degradation in experience for consumers & businesses.

So we need to think of a better solution to discourage money laundering than mass surveillance.

[manholio]

KYC is by definition a form of centralization - the government endows a well known private entity with the authority to run financial transactions on behalf of individuals, subject to certain constraints and information gathering. Since any form of practical KYC involves centralization, my proposal is to have an open system where brokers can join and compete amongst themselves, and where hopefully you might find one that does not sell your data.

There is no logical compromise between a KYC value transfer system and a non-KYC system, you either have auditable customer identity or you don't, in which case all is lost due to the Sybil attack: the money launderer can create a limitless number of pseudonymous identities (for example in Bitcoin, wallets) and transfer value both om chain, and off chain, by selling those funded identities (see for example the method used by Chipmixer, where the actual value transfer happens after the "clean" coins have been injected into the blockchain, sometimes months after).

That fiat converters are obliged to follow KYC is irrelevant to the issue being discussed here: blockchains as money laundering machines. The police will never ask a successful criminal for their tornado proof because police will most often be unable to follow the proceeds of crime through the pseudonymous value mixer that is the blockchain.

> the data does not support that money laundering is a bigger problem as percentage of transactions in crypto than traditional finance. The AML laws have been largely ineffective in practice in traditional finance

Anti-money laundry coverage is spotty in the traditional financial system. Where it is employed, it's very effective in exposing criminals, curbing their profits, thus motivation for crime, or pushing them into risky and noisy beheaviours (ex. truckloads of cash). Current blockchains are by design unable to comply with the most basic AML requirements and it's sheer lunacy to brush that problem aside or claim that crypt is in way comparable to physical cash; it's crack cocaine vs weed.