That's a terrible straw man. My counterargument to the "indeterminate potential for future abuse that makes it important" is that the same logic applies to all human transactions, so it's a moot premise.
Increased security standards on metadata for human transactions is actually the right response to the existence of bad actors, but increased security != forcing bans on the use of metadata.