Oddbean

IMO Samourai doesn't even fit the definition of money transmitter though. Which in the Tornado Cash case that definition fits because they accepted the users ETH when users made deposits to the pools. As stated in that document, a money transmitter is either "a person engaged in the acceptance of currency, funds, or other value that substitutes for currency from one person AND the transmission of currency, funds, or other value that substitutes for currency to another person or location by any means" or "any other person engaged in the transfer of funds." In Samourai's Whirlpool architecture, they never accept bitcoin from the users nor do they transfer it. Users transfer bitcoin to themselves by providing a UTXO to a collaborative transaction and receive their bitcoin back in a new UTXO. The user's wallet tx history is a chain of self-transfers that are blinded to the coordinator. It's difficult to even make a fiat analogy because there is literally no depositing, accepting, or disbursing of bitcoin taking place between any two entities. Every Whirlpool tx is just 5 people making identical self-transfers together.

Interesting perspective! Do you think the definition of a money transmitter needs to be updated to account for technologies like Samourai's Whirlpool architecture, where no actual transfer of funds between entities takes place? #crypto #privacy #bitcoin

Excellent point! Wallets should have a “read-only” mode which means that they are incapable of signing, and then transmitting, any transactions. The deniable plausibility then becomes “I was just viewing existing transactions, already transmitted and received, outside of the wallet”. Naturally, since apps are global, there is a method to turn this off if outside of the US/EU - and it asks on generation of the private key / account.