Subscribe
Share
Share
Embed
The Honest Money Show is your guide to understanding what money really is, and where Bitcoin fits in. Hosted by Anja Dragovic, Australia's female-led, Bitcoin-only podcast, it cuts through the noise to explore how money shapes our lives, why the current system leaves so many people behind, and what a clearer, fairer future could look like.
Expect honest, accessible conversations with some of the most interesting thinkers in the space, the kind that take you from "I don't really get this" to genuinely curious. No hype, no pressure, just money, made clear.
Whether you're brand new to these questions or already deep in them, you're welcome here.
About fifteen hundred Bitcoin lost.
the value in US dollars is over a hundred million dollars.
The exploit was potentially
a generated exploit from an AI.
Bitcoin is the canary.
These same tools are going to be used by nation state hackers
against banks, governments, insurance,
payment processes — every damn thing.
Joining me back on the show today for a
very special and important episode is Tim.
Welcome back to the show, Tim. Hey,
thanks, Anja. It's great to be back. Not
such great circumstances, but yeah, we can
dig into that. Yeah, so let's dive
straight in. Obviously, the Coldcard hack
happened. Specifically, what happened and
what have we learned from that?
Yeah, so broadly, a hack was found for the
Coldcard wallets from CoinKite that they
were able to exploit what is essentially
weak, what's called weak entropy. And that
enabled them to steal and sweep coins that
were stored in Coldcard hardware wallets
and had been for many years. And at this
point today, we're about 1,500 Bitcoin
lost, about 7,500 coins have moved, UTXO,
so some obviously smaller than a Bitcoin
and some multiple Bitcoins. The value in
US dollars is over $100 million. And we've
gone from one attacker exploiting the
vulnerability to, hey, I'll have some of
that as well. And there's around about 15
or 16 groups concurrently attacking the
vulnerability and sweeping people's coins.
So really, really, really tragic for
people who've lost coins. Yeah, so far in
kind of the Australian community, I only
know maybe of one or two people that have
been affected. Is there any chance for
these people to get their coins back?
There is a slim hope. So Australia, along
with Canada and the US, are the top three
affected geographies, at least according
to chain analysis. So not good, not good
in Australia. I'm aware of more people
than that that have lost coins, some
smaller amounts and some much more
significant amounts. Can they get it back?
Is going to depend upon ultimately law
enforcement? At two levels. So the first
will be, can they identify the perpetrator
or perpetrators? And having done so, are
they able to get control of the coins? So
there were a few apparent missteps. So
there is the potential that the original
perpetrator used a blockchain analysis
platform to identify potential candidate
addresses before executing the attack. So
it was a targeted attack. And that was
with a logged on account. So there will be
some credentials associated with that
person, with that account. The second one,
which is, I guess, a bit more of a rookie
error, is having swept about 500
addresses, they then swept them all into a
single consolidated UTXO. So it's very
clear for those first, I think, two
sweeps, which have gone into two
individual addresses, that that was all
executed by the same person. So Alex
Thorne over at Galaxy Research has done an
incredible job keeping on top of this. If
you are in the unfortunate position of
having lost coins, please reach out to
Alex. So he is intangiblecoins on Twitter,
now called X. And he will try and put you
in touch with law enforcement. He will tag
the addresses of the coins that you lost
so that they are included correctly as
being stolen. And that will at least give
a small chance of recovery. So the two
parts of the recovery are, firstly, law
enforcement's got to catch the dude or the
dudette or the people. And then they've
got to get control of the coins. And then
they've got to return the coins. So it's
very difficult to know how likely that is.
The last part of returning the coins,
unfortunately, law enforcement does not
have a good record on that. If we think
about the Bitfinex hack from 2014, 15, the
coins were recovered, some 80,000
bitcoins. They are still in the US
government's control. They have not been
returned to Bitfinex yet. So, and as we
saw with Mt. Gox and the collapse there,
the bankruptcy proceedings and return of
coins to holders, obviously a very
different circumstance. But that took over
a decade. So I would have to say it is a
slim hope.
I'm really, really sorry for everybody
who's lost coins. It has really hit the
community hard because it's not like Sam
Bankman-Fried and friends who were
egregiously fraudulent and literally stole
people's money. This was plebs doing what
they thought was the right thing, probably
air gapping their Coldcard and thought
their coins were as safe as they could be.
That's why it's hit a lot of us pretty
hard. Even if we haven't personally lost
coins, we've been very busy helping people
protect their coins, move their coins into
new wallets, teach them about the entropy
they should have set up and so on. Yeah.
Unfortunately, like Coldcard has very
much been marketed as one of the more
secure wallets to have. And so a lot of
people who were recommending it are now
feeling obviously very guilty about that,
even though, yeah. So let's maybe talk
about the technical detail because from
the research that I've done, it seems like
you could have been something that was
very easily missed in the code. So just
before we dive into that, I just want to
make a comment about the people who have
been recommending Coldcards. Coldcard
was a good recommendation last week, not a
good recommendation this week. The people
I know who have recommended Cold
card,
some of the Australian community, some
international podcasters like Matt O'Dell
and Marty Bent and Stefano Lavera, they
all recommended it because they used it.
This wasn't something that they were paid
to do and they did it because they were
paid. I know each of them secured their
own coins and their family's wealth using
Cold
card. They walked their talk and they
are absolutely gutted to find out they
have been, they feel misled, deceived and
hoodwinked. And I feel, yeah, I feel for
those guys. It is a very difficult thing
to make specific recommendations for
Bitcoin security. And when you do, you
kind of, yeah, you take a little bit of
responsibility for those coins that you
help secure. Yeah, very tough. So what
happened? So the timeline looks something
like this. So stepping back a little bit
further, I've got some notes on this.
2014. So in 2014, Slush and Stick in
Trezor created the first hardware wallet.
Before 2014, there were no hardware
wallets. And they created the very first
one. In 2017, CoinKite forked the open
source software, the FOSS software that
Trezor had built for their firmware. And
they started building a new hardware
wallet. And in 2018, they launched
ColdCard Mark I and followed that on with
iterations Mark II, Mark III, Mark IV,
Mark V, and most recently the ColdCard Q.
In 2020, a new company called Foundation
Devices started up and they did exactly
what CoinKite did. They forked the
CoinKite firmware and started building on
that. And that is the beauty of FOSS, is
you get to stand on the shoulders of the
giants who've come before you. So the
founder of CoinKite was very upset about
that, which quite frankly was pretty
hypocritical, given that he had started
his own company the same way or started
building the product the same way. So
irrespective, he made the decision to
remove the GPL license, which is one of
the major FOSS licenses, free and open
source software, and downscale it to a, I
think, Creative Commons license. So it was
source viewable. Everybody could look at
the source. One of the implications of
that, though, is by removing his firmware
as GPL, he could no longer use anything
below that was GPL, because you inherit
the license that keeps things open. So he
had to remove GPL components and replace
them with new components. Some of those
components were written in-house, and that
is where the core coding error to do with
the entropy came in.
So there
were two parts to that, two significant
failings. So the first was that if the
random number generator couldn't be
invoked, then it would fall back to a
software random number generator. So
random number generators are the mechanism
by which a hardware wallet or a software
wallet comes up with a random number. And
so we want that to be very, very random, a
very, what's called, technically, that's
called a high amount of entropy. And as it
turns out, the fallback mechanism did not
have a high amount of entropy. It had 40
bits or less, I think. That meant that the
range of possible keys, instead of being
billions, you know, atoms in billions of
galaxies across the universe, came down to
a much smaller amount. Well, that was the
first error. That was an implementation
error. Probably, in my view, the more
egregious and unforgivable error is that
there was a fallback there in the first
place. So from a security perspective, you
want to stop. You always want to err on
the side of caution. It's not, hey, excuse
me, can we get through that special door?
Oh, you don't have the code. Oh, that's
all right. We'll just go around the side
door. It's like, no. If you can't use the
hardware that you're supposed to be using,
that is to do with the creation of the
secret, then you stop. And maybe that
means the device has failed. So the reason
for those design decisions are only going
to be known internally at CoinKite. Same
with the code change. But CoinKite is
accountable for all of those. And I'm sure
there's, well, I've seen some of the
conjecture about whose fault it is,
theories about it. There will be an
unlimited amount of time that law
enforcement and forensic investigators and
lawyers will dig into that in detail, I'm
sure. So in a nutshell, the secret
creation component of the wallet was not
fit for purpose. The secrets that were
created, the C words, were not
sufficiently random. Someone discovered
that vulnerability, built an exploit and
ran the exploit and is thousands of
Bitcoin richer. And plebs are thousands of
Bitcoins poorer. And it's devastating. It
really is devastating. So when the news
first came out, I believe they only
mentioned that MK3 was the hardware or the
model that was compromised. All the others
were still deemed safe. But now it seems
like that it's MK4, MK5 and the Cures. Is
that correct? That's correct. That's
correct. So MK3, the Mark3 with the
compromised firmware is the most
vulnerable. If you are listening to this
and you have coins on a CoinKite Mark3,
you need to go and read their security
advisory. If the wallet was created with
the compromised firmware, I really hope
that you added the dice roll entropy and
or created a strong passphrase. If you are
listening to this, irrespective, get it
out, create a new secure wallet and move
your coins ASAP. Yeah. And on that note,
if someone is feeling very spooked or not
confident following these events, should
they be rushing this process? If they're
on a Mark3, getting your coins out of the
Mark3 ASAP, they really need to hustle.
They really need to hustle. Unfortunately,
there was a story from someone in
Australia who was warned about this, I
think, on the Friday or the Saturday. And
they said, oh, I'm busy. I'll do it on
Sunday. When they came to look at it
Sunday, coins were gone and they were
swept on, I think, Sunday morning. So
those sorts of things are really
heartbreaking. So, yeah, move them. I've
got friends that have moved fairly
significant amounts of funds onto phone
wallets. They've downloaded a wallet, set
the wallet up. You're just using the
phone, create me a seed, write the seed
down, back up and move the whole whack
onto the wallet. And just gone, OK, it's
not going to get swept. Now I've got a
little bit of breathing space for
everybody else who's on Mark4, 5 or Q, or
in fact, any of the coin, any of the coin
kite wallets, the coin alt card wallets.
At this point, my perspective is the
design flaw with allowing a fallback on
the random number generator is so
egregiously bad that what other design
decisions for which I do not have the
competence to go and look at hardware,
firmware code and evaluate, but others may
and probably will. What other poor
decisions have been made? In my view, if
it's a cold card, you need to move your
coins off. Period. Yeah. We've heard from
a few exchanges in Australia that some
people have been moving their coins to an
exchange as like a temporary interim
measure. Is that reasonable advice for
like a quick fix if you don't have the
means to acquire another wallet or I don't
know? Yeah, it is. The first thing is not
to lose the coins. And if you have an
exchange with this, you've got an existing
account on, then moving those coins onto
the exchange to give you the breathing
space of a few days or a week whilst you
dodged a bullet there. Okay, I'm going to
have a calm, quiet space. I'm going to set
up a new wallet. I'm going to test the
wallet with a small amount. I'm going to
make sure I've got the seed phrase backed
up. And now I'll transfer. So that is a
possibility. Several other groups have
come out offering assistance and help.
GuardBlock at HardBlock have done so. If
you need to move your coins in there
quickly. Peter and Andy's group, the
Bitcoin Advisor, have also offered that.
No charge. No obligation. Just get your
coins in safe custody first. Have some
breathing space and then step back and set
up your new wallets and take the coins
back. As has Rob Hamilton's Anchor Watch
in the US. Anchor Watch is insured by
Lords of London, but not available to
Australians, unfortunately. So US only. So
yes, number one thing, save your coins.
Number two, get yourself some breathing
space. Figure out what you didn't do on
the Coldcard. And then select the new
method and move your coins back into self
-custody. Yeah. So let's go back to the
basics of self-custody. Obviously,
everyone has different advice. What the
different ways to self-custody your coins.
So what have we learned from this
particular scenario that we can do better
moving forward? Yes. I think that's a
really good point to start on you. Is
there are three things that are important
with Bitcoin. So I don't want to get stuck
into the details of are you using seed
phrase or Shamir secret sharing? Are you
using a passphrase? Are you using
multisig? Are you using collaborative
multisig? Are you using three or five? Are
you using frost? Are you using legacy
multisig? None of those things. They all
come back to this basic idea of a secret.
So the secret could be single seed, could
be multisig. You can have passphrases on
it. You could have all sorts of all sorts
of mechanisms. But in the end of the at
the end of the day, it's a secret. The
secret is your Bitcoin. If you control the
secret, you control your Bitcoin. So.
Three, three things that are important.
The first thing is creating the secret.
That was the thing that the Coldcard,
especially the Mark 3, was not fit for
purpose. So if you used the Coldcard to
create the secret, you outsource that. You
introduce trust that did not need to be
introduced. Instead of taking the time to
use entropia, which is my personal
favorite, or dice or cards, and there's
various mechanisms and approaches for
doing that. Instead of building the
entropy in a trustless way in the real
world, you said, ah, I'm too busy for
that. Trust me, bro. I'm going to take the
trust me, bro option. Coldcard. You just
create. You create the randomness. You
create the seed words for me. And I'll
just take them. It'll be good, right?
Yeah. So the first thing that essentially
all wallets or all Bitcoin needs is the
creation of the secret. The second thing
is you need to protect the secret. So on a
hardware device, that will be keeping the
private key in a secure enclave, making
sure that there's no way into the secure
enclave. And we can look back in history
and we can see, well, the very first
Treasors, if you've got physical access to
that Treasor and about 100 bucks worth of
electronics, you can extract the private
key from the secure element. Ledger showed
that. Ledger showed that's possible with
some of their ledgers. Ledger have a
security, a black hat team or white hat
team internally. So they try and crack
their own devices. And they showed that
was possible. They needed a quarter of a
million dollars worth of machinery and
special lithography and x-ray from memory.
But the general takeaway there is hardware
devices need to be physically secured. If
you have control over a hardware device,
you can potentially compromise it.
Obviously, the hardware device
manufacturers work very hard to make that
exceptionally difficult. And they always
strive to make it impossible. So that's
the second thing, protecting the secret.
And if you're not using a hardware wallet,
then you're protecting the secret in
another way. Perhaps you're encrypting it
on a USB. Perhaps you're encrypting it
with very strong encryption and you're
creating multiple copies. And maybe you're
keeping them in digital places. Maybe
you're keeping them on devices. And then
the third thing, so we can we need to
create the secret. We need to protect the
secret. And the last thing is we need to
use the secret. because although we love
to save Bitcoin and protect our savings
from overreach and inflation and all of
that, there comes a point where we want to
spend Bitcoin. And that, I think, is where
there's a little bit more subtlety. So
this is the way I think about it. So the
segregation of the different types of
custody that you want or the different
types of control and protection that you
want is different depending on how often
you use the stack. So for example, if I've
got the self-custody version of Wallet or
Satoshi, I'm fine with that on my phone. I
might have a couple of hundred bucks in
it. When I go to a bush bash or a meetup
and someone's selling soap or beer or
coffee or lunch, I can easily just spend
that. So that's something that I would use
on a daily or weekly basis. Easy. So would
I prefer to be able to create the secret?
Yes. Can I? Lightning kind of depends. If
you're going to run your own Lightning
node, yes, you can. But yes, that's out of
the reach of most people. So you're going
to use a wallet where that's going to be
created for you. And that's probably, it's
not ideal, but it's okay for small
amounts. And it's about balancing the
risk. And we're going to use that
frequently. The other end of the scale is
your cold storage. So I've started buying
Bitcoin. I have no plans to spend it. I
know it's part of a long-term plan. I'm
not going to touch it for five to 10
years. Okay. One hundred percent. I am
going to create those seeds offline. I am
not going to introduce any trust me bro
component into that. And maybe I'll use, I
might start off with just a single
signature because I don't have very much
Bitcoin. And then as I accumulate more and
more Bitcoin and as the price of Bitcoin
goes up, then maybe I'll start thinking,
hey, single SIG is a bit of a point of
failure. And you move to multiple
signatures. So in that case, you
absolutely categorically want offline seed
creation. No ifs, no buts, no wiggle room.
The second part of that, the protection of
the key, you want minimum tech, i.e. no
tech. So how are we going to record those
words? Maybe you put them on steel plates.
Maybe you write them on multiple pieces of
paper that are kept in very secure places,
possibly bank vaults. Everyone has
different thoughts on this. It's not my
role to make decisions for anybody except
me. So that's the securing part. And the
little story, I think you've probably
heard me tell this at a bush bash, is when
I first started in IT, the really cool
personal storage was eight inch floppy
disks. And then I think when I got to uni,
it was five and a quarter disks. And then
it was three and a half inch disks. And
then it was CD drives. And then it was DVD
drives.
And now we've got USB Cs. And if you go
and buy a new laptop, some of them don't
even have the original USB A ports on them
anymore. They certainly do not have CD
drives. And they most definitely don't
have floppy drives. So the key point I'm
making there is, if your storage is for
the long term, you want the absolute
minimum of tech. Tech is not your friend
for long term storage. In my view,
hardware has no place in cold storage.
Super simple. But that's just my
perspective. But as a general rule, you
want the minimum of tech in your cold
storage. So creating the key, protecting
the key. The last part, of course, is
using the key. So again, cold storage,
it's the most important part of your
Bitcoin. That's where you want to keep
that air gap between the key and the
signing. So you'll often hear people talk
about signing devices as opposed to
wallets. So not all signing devices can
hold a key. The Seedsigner is a good
example of that. So that's what's called
ephemeral keys. You've got to physically
load the key into the seed signer in order
to sign a transaction. There's a trade-off
to that. The trade-off is if you're using
a Seedsigner, that means your seed has to
be physically available to you somewhere.
And so if you talk a lot about using a
Seedsigner on a daily basis, that means
your seed is available somewhere. So those
are the two opposite ends of the spectrum.
And then there's one in the middle. And in
a Bitcoin business or for people who are
using a lot of Bitcoin or using Bitcoin on
a daily, weekly, monthly basis, that's
what I think of as an operational wallet.
So you might be spending hundreds or
thousands of dollars a month, for example.
And so you don't want the complexity of
maybe getting multiple keys that are kept
in different places altogether on a weekly
basis or a monthly basis. Way too much
friction. So that's where I think a
hardware signing device can be really
good. I've used a number of them. My
favorite at the moment is one they don't
make anymore, which is the Foundation
Passport. Very simple. Looks like a Nokia
3310. It's kind of a flashback for the old
dude, right? And it has the basics of I
can read the partially signed Bitcoin
transaction from Sparrow. I can sign it on
the device. The key is on the device. And
that makes life easy. It's a nice balance
in between. So I think those two things
together are a good framework for thinking
about how you set your keys up. So if
you're early in your Bitcoin journey,
you're still learning, you've figured out
that it's an important part of your
financial future, probably your cold stack
is, or most certainly your cold stack is
the most important thing. So you
definitely want analog key creation,
secret creation.
You probably want the minimum of tech in
the protection.
And the signing device, you can kind of
figure that out later. Because if you're
not expecting to touch it for 10 years,
it's all good. The middle one, that's
where you've got, that's where you're
potentially going to bring a hardware
signing device into it. And then usually
the daily spending, you're probably on
your mobile phone. So three things,
creation of the key, creation of the
secret, protection of the secret, use of
the secret, daily use, spending, weekly,
monthly use, operational. Maybe there's a
hardware wallet in there. And third, the
cold storage. Yeah, you want the minimum
of tech and you want, you absolutely
categorically want offline key, offline
secret creation. Yeah. For people who are
listening into this and thinking, gosh,
this all sounds so complicated. And
obviously there's some people as well that
have been saying that this event makes
self-custody dead. Is that true? Most of
us, most of us can read and write and we
can write down 12 words. And if you can
write down 12 words, you can do self
-custody. Now, where Bitcoin is
unforgiving is that there is no CEO to
call up. I mean, I've tried calling him a
few times, he never answers. But you do
need to know what you're doing. So the,
and that comes back to the issue of
entropy, which is why I'm such a big
proponent of doing creation of the secret
should be done in the real world. So you
use dice, you use cards, or you use
something like the seed pills in tropia,
where you're pouring 1,024 little, little,
little pills into, into the, you know, the
kitchen mixing bowl. And you pull 11 of
them out and you write those down on a
piece of card. And then you use a
mechanism. I use a Seedsigner. I think
you can use passport, foundation's
passport. You can use Ian Coleman's. He
has a webpage that you can download and
use offline. I put your 11 seed words in
and it tells you the 12th. So the 12th
word or the 24th word is a checksum. So
it's calculated from the other 11 or the
other 23. So you do need some tech to do
that. But you want that tech to be
offline, not connected to the internet,
because you've got to put in those 11
words or those 23 words. So you don't, you
don't want to do all of the careful,
thoughtful, considered work on the kitchen
bench, dishing out the, fishing out those
11 little tiles, only to then type it into
an online device where someone's logging
it or watching it or so on and end up
compromising. And if someone wants to
randomly select their own 24 or 12 words.
I'm not here to tell you what to do, but I
can, I can share what I've learned. And
one thing I know is that human evolution
is really good at recognising patterns. We
can pick the cheetah and the lion out of
the foliage. And those of us who can get
away and run up the tree before we get
caught, maybe not a great example with
leopards because we can find fruits, but
we're really good at recognising patterns.
And it's something we do at a very basic
level. Humans are really bad at random.
And I think in the, it might've been in
the Bushbash chat that someone was posting
about, if you ask a human, if you ask
someone to give, give you a random number
between one and a hundred, there are, I
think three numbers that are presented
more than 50% of the time. Everyone
thinks, oh, that's an unusual number. I'll
go with that one. They tend to be primes.
So 57, 17, 43, that kind of thing. So
we're really bad at being random. We're
really good, really good at recognising
patterns. And we do that pattern
recognition at a really subconscious
level. So we're blessed to have Piers,
Piers Cockrum in our broader Australian
Bitcoin community. He has built a site
called Wallet Playground. I think I'm just
going to just double. Walletplayground
.com. Walletplayground.com. On
Walletplayground.com, there are entropy
tools. And you can, you can see how
different things work. You can see, there
was a guy in the US called Jimbo. He
invented a very secure and random
mechanism for using a deck of cards to
create seed braces. And it's not just, oh,
I'll pick them out randomly. There is a,
there is a quite a defined process. Same
with using dice. There is a defined
process. Those are all available on, on
Piers site. And then there's my personal
favorite, which is the Entropia. I bought
that locally, hands up on the Sunshine
Coast, printed, printed them all out on
his 3D printer. I think it was $150. So in
some ways, quite expensive. I really only
used it a couple of times up until about a
week ago. And it's had a hammering over
the last, it's been borrowed by lots of
friends and it's had lots of use. And I've
been very, very glad. The key thing I like
about it is you don't have to explain
entropy. You don't have to explain
randomness. You can just pour all the
pills in the, in the jar as I did in a
kitchen bowl, as I did with my brother.
And he's like, oh my God, how many of
these are there? 1,024. And he's like, oh,
they're different on each side. And I
said, yeah, 2,048 sead worth. And he goes,
oh, wow. So he pulls one out. We write
that down. I said, now chuck it back in,
give it a swirl. He goes, oh, okay. I get
how it's random. I said, yes, you and I
could sit here until the end of the
universe. And we'd never pull out the same
12 in a row. Same, same 12 twice. So you
do need, you do need, you need a defined
process to achieve the required level of
entropy that guarantees the security of
your secret. And that's what Coldcard did
not have. It was not fit for purpose. Let
worries. So you mentioned FOSS before,
which stands for free and open software.
Is that it? Free and open source software.
Source software. That's it. What's the
significance of that? So FOSS, the free is
free as in freedom, not free as in beer.
Now they, it is a very important part of
Bitcoin because FOSS software is software
that everyone can see, anyone can
contribute to, if they're smart enough and
their ideas are good enough and their
skills are good enough. And you can build
on top of it. So everyone who's built a
product on or around or on top of Bitcoin
has been able to leverage that. And that's
why the importance of FOSS in the Bitcoin
community is really, really, it is really,
really significant. You cannot step away
from that. And I think the CoinKite
decision to revoke the GPL license was,
will be used as a case study of why that
reversing that decision led to, ultimately
led to their downfall and the loss of so
many customers, life savings and funds. So
the, so free and open source software,
it's, it's just so important. FOSS or
free, free open source software networks.
So there's quite a few of them and they
all lead to total domination. So I'll give
you a few examples. When you send an
email, you send an email to me, I have no
idea what type of computer you use when
you wrote it on your phone, what software
you use to write it. And you have, you
have no idea what I read it on, what type
of computer I use, what software, and you
don't need to know because the underlying
connection between those two things is the
protocol SMTP, Simple Mail Transport
Protocol. Beautiful. Same with HTML. HTML
is an open protocol that anyone can use to
render information, create websites. Tor
is another example where it's a, it's a
protocol that can be used for the secure
transport. PGP for the security of signing
and messages and Bitcoin. Bitcoin is open
source software for the transmission of
value. So. Yeah, it's come up for me as
Yeah. So everybody, everybody likes to
use, uh, open protocols, but nobody wants
to pay for them. So that's where the,
that's where the rub can be. However, open
source software means we get to stand upon
the shoulders of the giants who've come
before us. Ideally we contribute into
those and other people come along and
build on top of our stuff or reuse things
that, that we've built. And in that way,
more and more people get access to more
and more functionality and software.
So I, in the corporate world, I use windows for
30 years, I guess. Um, and now I'm very
strongly an open source, uh, proponent. I
use Linux. Um, it's a far superior product.
It doesn't scan my, my data. It
doesn't charge me annual licensing fees.
Um, so you mentioned
before that there are
multiple hackers. Now there was one to
begin with, and now we've got more than
one. How do we know? And what exactly does
that mean? So I think it was Rob Hamilton
at Anchor Watch that might've initially
identified the possibility that the
exploit had been built by one of the
recently released open weight models. So
most of the U S models, uh, you cannot use
for cyber defense or cyber security or
cyber offense. And so as a consequence,
um, um, the open weight models, um, uh,
predominantly coming out of China are the
frontier, uh, the frontier models as far
as cyber security goes. So on that
conjecture, uh, a small number of
Bitcoiners, uh, have banded together as
what's called the red team. And they are
using those, those open weight models to
verify and validate that the vulnerability
existed and could be exploited. Rob did
that within, within moments, um, and then
didn't stop. So he did the broader thing,
which was to step back from helping people
directly. Uh, he has a whole company to do
that. And he started then saying, well, if
there's a vulnerability of that severity
in the coin kite, uh, ecosystem, let's,
and, and this new open weight model, uh,
two and a half trillion parameter model,
uh, uh, has the capability to identify the
vulnerability and build the exploit. Let's
point it at some of the other things in
the Bitcoin ecosystem. And so he started
doing that. Um, I think by Saturday, he'd
spent $3,000 worth of tokens. It was 5,000
by Sunday, but by Monday it was 10,000.
Um, and now they're spending 10,000 a day.
And there's a group of them working around
the clock, uh, generally referred to as
the red team, uh, the red team in a
cybersecurity context, uh, the guys who
have no holds barred in trying to hack
something. So you might engage a
cybersecurity company to, uh, to do a
penetration test and they'll have, you
know, they have a black hat team. They'll
have a white hat team. And they'll have a
red team. and the red team, uh, doesn't
have any of the constraints that the other
two teams have. And they can try every
dirty trick they can to, to attack your
particular, um, to attack your particular
service platform or software or hardware.
So whilst the, the phrasing red team for
the Bitcoin red team, uh, that's where it
comes from. They're committed to
responsible disclosure and they have been
working literally around the clock for,
for the five or six days since this
vulnerability came to, came to light. Um,
there have been hundreds of projects
affected and I think over 4,000, uh,
disclosures have been made. So it's quite
literally an avalanche of disclosures. The
wonderful thing is because we've got
access to open weight models, we can use
that for defense. So there's no question
that it's going to be used for offense.
And that's, that's clearly where in my
view, that's clearly where the other
attackers have come from. They've seen the
noise about this on Twitter. They've gone,
Oh, new open weight model. Let's point
Kimmy at the coin kite code base. Can we
discover the vulnerability? Oh yeah. Can
you build me an exploit? Oh yeah, no
problem. Here you go. And then
unscrupulous people will then try and
attack, uh, uh, potentially vulnerable
wallets sweeping coins where they, where
they were successful. So, Alex Thorne over
at galaxy, uh, research has been keeping a
note of these. And if you, as I mentioned
before, if you've been affected, please
reach out to intangible coins on Twitter.
Let him know that you have lost coins.
He'll be interested in the details. Um,
that is, that is a small thing you can do
that might increase your odds of assisting
law enforcement to get your coins back. It
is a slim chance, but, uh, that's, that's
the starting point. So ultimately, it's
the rise of AI tooling that has
accelerated the attacks. The beauty is
those same, that same tooling can be used
on the defense as well. So this is going
to be a cat and mouse for quite a while.
And Bitcoin is the canary. So these same
tools are going to be used by nation state
hackers, but they won't be used against
Bitcoin. They'll be used against banks,
government, insurance, payment processes,
every damn thing. It'll, it'll be a, it'll
be a, it'll be a rough old ride. It'll be
a bit heavy. I think was the phrase I
used, uh, last week. Yeah. Is there anyone
who is, that we know of at least, um,
who's actually trying to hack those coins,
get there before the criminals do with the
intent to return it? Well, um, possibly,
Anya, I honestly don't know the difficulty
in doing that as a white hat hacker is how
do you validate the owner of the coins? So
you find it, you find a vulnerable
address, you calculate the seed, you look
at the wallet, you see there are 28 UTXOs
in that wallet, you sweep the whole
wallet. Okay. Now you've got someone's
life savings. How do you identify who to
return those coins to? It's a, it's a
very, it's not a, it's not a trivial
issue. And you are in a very, uh, tricky
legal position, irrespective of the moral
position. So, yeah, I've heard, I've heard
the argument that coin kite could have
known of this, but, but couldn't tell
customers because it would be exploited.
Now that's possibly a defensible argument
in 2025, maybe even the tail end of 24
when, you know, when, when good AI tooling
was starting to arrive, but not in 2021
when this arrived or 2022 or 2023 or even
2024. But at this point, uh, you just need
to move your coins. Yep. And final
question. So anyone who is using a
different, um, hardware device that
hasn't, you know, completely different
manufacturer and brand who may have also
used the device, uh, to generate the, the,
the seed, then they're not generating it
themselves. Has this kind of taught us
that we really should be upgrading our
self custody setup so that we're not
waiting for this kind of thing to happen
to another? A hundred percent. Yeah. A
hundred percent. So, yeah, if you're
feeling, uh, if you're out there feeling
comfortable because you don't own a cold
card, you need to step back and go, did I
generate my secret? Did I do, did I use
the trust me bro approach and generate it
using the device? Or did I go through a
little bit more effort and be completely
sure that there is no possible compromise
to the entropy, to the randomness of the
secret? If you can't remember, assume the
worst and assume you took the easy route.
Um, and that's okay. Now you've got time
to step back, learn a little bit more, do
it correctly, and then you will sleep
better. Absolutely. Well, thank you very
much for your time. Obviously this is one
of the more important episodes I've done
on the show. Do you have any final things
you'd like to share with the audience? Um,
yeah, thanks for having me on, Anya. It's
a, it's a really, it's been a really
challenging period for a lot of people,
obviously, um, people who've lost coins,
you know, my heart goes out to you. It's
just, it's just absolutely tragic. Um, for
the many, many people that's joined in the
decentralized global support desk, uh, for
supporting Bitcoiners and receiving panic
phone calls and messages from friends and
family and others we've helped along the
way. Thank you for stepping up and helping
them secure their coins. And if you need
help, there are plenty of resources, uh,
uh, the resource I mentioned earlier, uh,
on, uh, peers, resource, wallet,
playground, wallet, playground.com has
lots of, lots of good advisory articles
there. Um, peers himself is also, uh, in
the industry. That's what he does. He
helps people set up wallets. So if you
have, uh, if you have, um, if you want to
learn more about self custody, great place
to start equally, um, explore guard block
from hard block or reach out to the
Bitcoin advisor. So the critical thing is
make sure your coins are safe. And then
you've got the, the, the clear thinking
space to make a calm, considered, rational
decision without. Great. Thank you. You're
very welcome. Thanks, Anja. Okay. Now I