to figure out who that owner is. Linking
Bitcoin accounts to real-world identities
is possible because information tends
to leak out. Regulated cryptocurrency
exchanges—generally those in the US or
Europe—must follow know-your-customer
and anti-money-laundering rules, which
require people to hand over identification
before using their services. Some people
are even so careless as to post their supposedly private Bitcoin addresses in online
forums. “What people forget is that the
blockchain is just one half of the equation,” says Knottenbelt.
Chainalysis and Elliptic now use
machine learning to help cluster addresses.
Soon it might even be possible for an AI to
police blockchains in real time.
The wall-size data visualization at
Imperial College is a step toward that.
The blue-and-yellow tangle that caught
Knottenbelt’s eye was a coin tumbling network, a sequence of transactions deliberately designed to make it harder to track
individual coins. It’s like dropping money
into a jar, shaking it about, and then taking
it out again: the amount doesn’t change,
but it’s hard to tell which coin was which.
The effect is much the same as if you move
money through a bank in a place like the
Cayman Islands, where there are strict
secrecy laws around banking.
Staying one step ahead
HOWEVER, TUMBLERS AREN’T NECESsarily a sign of criminal activity. “Some
people just do it for privacy reasons,” says
Knottenbelt. And in any case, there are better ways for criminals to cover their tracks.
As the limits to Bitcoin’s privacy become
more apparent, people are moving to new
cryptocurrencies, like Zcash and Monero,
that reveal almost nothing about the transactions recorded on their blockchains.
Zcash uses a so-called zero-knowledge
proof to verify transactions. This is a math-
ematical way to confirm that a transaction
took place without revealing any informa-
tion about who was involved or how much
was transferred. Zcash also lets you hand
back coins and have fresh ones mined, the
equivalent of trading your marked bills in
for clean ones at the bank.
Monero, meanwhile, is effectively a
big tumbling network. When you want to
transfer coins, your address is mixed in
with a bunch of others so that no one can
tell which one was spending the money.
Zcash and Monero certainly take privacy to the next level. But that doesn’t
mean they’ll never give up their secrets.
Meiklejohn points out that sloppy user
behavior, such as posting your private
address in forums, will again leave behind
clear trails, just as with Bitcoin.
What’s more, Monero gives users the
option to carry out transactions with no
obfuscating coins mixed in. This removes
the privacy for that particular transaction
and adds a way for researchers to disentangle, through a process of elimination,
any mixers that subsequently include those
coins. Malte Möser at Princeton University
and colleagues estimate that 62 percent of
inputs to Monero transactions are vulnerable to this analysis. When users of Zcash
and Monero start to hemorrhage clues, the
likes of Meiklejohn and Möser will be ready.
Perhaps the biggest problem for law
enforcement, though, is the large number
of unregulated exchanges, where criminals
can wipe away the traces of their theft by
laundering the stolen cryptocurrency into
other forms of wealth. Many exchanges
defy regulation out of principle: the likes
of B TC-e and the conversion service Shapeshift, for instance, sell themselves on the
promise of asking for no identification
from their users. Shapeshift founder Erik
Voorhees is especially outspoken about the
political implications of regulation.
Security and cryptocurrency researcher
Ross Anderson at the University of Cambridge, UK, argues that these exchanges
thrive in part because laws are ineffective.
“The problem with anti-money-laundering
generally is that nobody wants it done
right,” he says. “If you’re a city bank, you
don’t want to know that John Gotti is a cus-
tomer, and so banks would never tolerate a
law that said whoever banks the mafia will
go to jail.” If that’s how the world works,
why should crypto exchanges be different?
Anderson’s cynicism about the author-
ities’ willingness to act has led him to
formulate a plan to take down the cryp-
tocrime system himself. He is creating
what he calls a taintchain—a public list
of bitcoins with clear links to criminal
activity. “What I’m going to do is publish
a list of all the stolen Bitcoin and the soft-
ware you need to generate it so that every-
body can check it for themselves,” he says.
Exchanges would then think twice about
handling stolen coins.
Even if regulation were stricter, how-
ever, it’s not clear that it would make a dif-
ference. “I don’t think outlawing anything
is going to help anyone,” says Knottenbelt.
Driving the tech underground, he argues,
will simply mean that transactions will be
hidden rather than broadcast openly on
the internet, making it even harder for
researchers like Meiklejohn to analyze the
money flows and find the thieves.
Surprisingly, Meiklejohn herself turns
out not to worry too much about regulation—or lack of it. “Once you’ve isolated
the problem to bad exchanges operating
outside of typical jurisdictions, then you’ve
kind of won,” she says. Take BTC-e, an
exchange based in Russia that was known
to have taken a lot of criminal money. A
lot of ransomware operators appeared to
be using BTC-e almost exclusively. It was
also where the missing Mt. Gox funds were
last seen before the trail vanished.
But in July 2017 it was closed down.
US authorities arrested staff and seized
computers at one of the exchange’s data
centers, and Alexander Vinnik, its suspected operator, was arrested. “They
clearly were not going to respond to subpoenas,” says Meiklejohn. “On the other