CATASTROPHIC COLDCARD BUG

WEBVTT
NOTE
Transcription provided by Podhome.fm
Created: 8/3/2026 5:06:15 PM
Duration: 2252.096
Channels: 1
1
00:00:02.159 --> 00:00:03.040
If you're
2
00:00:03.520 --> 00:00:07.520
currently unaware, which I imagine many of you are not, but just in case,
3
00:00:07.839 --> 00:00:08.799
this is
4
00:00:09.200 --> 00:00:14.320
an emergency, rabbit hole recap because there is an emergency vulnerability disclosure
5
00:00:14.934 --> 00:00:15.655
from
6
00:00:15.974 --> 00:00:20.775
coin card pertaining to their cold card hardware wallets going back to version
7
00:00:21.335 --> 00:00:22.935
four point zero point
8
00:00:23.494 --> 00:00:26.454
one or four after four point zero point zero.
9
00:00:27.095 --> 00:00:28.535
The random number generator
10
00:00:28.800 --> 00:00:31.839
was used to create the entropy to create your private seed
11
00:00:33.040 --> 00:00:37.120
was not working properly. And so any private key that you produced
12
00:00:37.360 --> 00:00:42.640
with an m k three after that version or any version for that matter, any model
13
00:00:43.440 --> 00:00:44.320
is
14
00:00:44.605 --> 00:00:52.365
not safe. So move your Bitcoin off of cold card if you're securing it there. I know many of you are already aware of this, but just in case you aren't,
15
00:00:53.725 --> 00:00:54.845
get on it quickly.
16
00:00:57.725 --> 00:00:58.685
Yeah. I I Yeah.
17
00:01:00.180 --> 00:01:02.580
Just real quick just real quick there.
18
00:01:04.180 --> 00:01:05.860
On the on on this
19
00:01:06.260 --> 00:01:07.780
absolutely catastrophic
20
00:01:07.780 --> 00:01:08.580
bug,
21
00:01:11.460 --> 00:01:12.980
the bug was introduced
22
00:01:13.540 --> 00:01:14.420
March
23
00:01:14.420 --> 00:01:15.300
2021,
24
00:01:15.300 --> 00:01:15.700
I believe.
25
00:01:17.524 --> 00:01:20.085
Marty and I are running on very little sleep right now.
26
00:01:20.485 --> 00:01:21.365
It got
27
00:01:22.725 --> 00:01:23.524
discovered
28
00:01:25.125 --> 00:01:28.244
late Thursday night after last rabbit hole recap,
29
00:01:29.260 --> 00:01:31.820
and it's been straight chaos basically since
30
00:01:32.540 --> 00:01:33.580
that happened.
31
00:01:35.340 --> 00:01:46.205
So if you generated a seed on any cold card after March 2021 and you didn't add sufficient dice roll entropy, so dice roll randomness, over a 100 dice rolls,
32
00:01:47.965 --> 00:01:52.284
you have a compromised seed. Now if you have a passphrase
33
00:01:53.244 --> 00:01:55.884
that has probably protected you up until this point,
34
00:01:57.010 --> 00:01:58.369
M k fours
35
00:01:58.689 --> 00:02:05.490
also really fucked, but slightly less fucked might have protected you up until this point. If you did multisig,
36
00:02:05.490 --> 00:02:07.890
might have protected you. If you did multisig with
37
00:02:09.465 --> 00:02:13.225
two cold cards, like the threshold of cold cards, you're vulnerable.
38
00:02:13.465 --> 00:02:17.305
If you did one, I you you're not you're technically not vulnerable.
39
00:02:19.465 --> 00:02:20.905
But I I I
40
00:02:21.145 --> 00:02:23.385
think there shouldn't be really much nuance here.
41
00:02:24.200 --> 00:02:26.840
If you're using a cold card and you've never
42
00:02:27.080 --> 00:02:32.680
and you you you still have funds on the cold card, whether it's in a multisig, whether it's a pass raise or dice rolls,
43
00:02:33.000 --> 00:02:34.920
straw like, you should you should
44
00:02:35.319 --> 00:02:36.599
take a deep breath
45
00:02:38.905 --> 00:02:46.825
and move your funds as soon as possible. And as far as I'm concerned, never use cold cards again. Just never use cold cards again. Move your funds.
46
00:02:47.305 --> 00:02:50.265
This is a fucking disaster. This is really, really bad.
47
00:02:52.150 --> 00:02:53.590
Yeah. I said it yesterday.
48
00:02:55.670 --> 00:03:02.790
The best way to describe the feeling that I've had since Thursday afternoon is is horror, pure horror horror. This is a disaster.
49
00:03:03.525 --> 00:03:04.165
And
50
00:03:05.205 --> 00:03:10.885
I think a lot of what this episode is gonna focus on is accountability. I mean, we talk about it a lot on the show. We
51
00:03:11.285 --> 00:03:15.445
try to demand it of others and would be hypocritical of ourselves not to
52
00:03:16.090 --> 00:03:18.410
hold ourselves to account as
53
00:03:18.810 --> 00:03:20.410
two individuals who have
54
00:03:20.970 --> 00:03:23.930
advocated and recommended endorsed cold card
55
00:03:24.650 --> 00:03:25.770
for many years.
56
00:03:27.530 --> 00:03:29.530
And as
57
00:03:29.075 --> 00:03:30.675
you can imagine, this is
58
00:03:31.075 --> 00:03:33.875
this is as bad as it gets,
59
00:03:34.115 --> 00:03:34.755
literally
60
00:03:35.475 --> 00:03:40.515
fucking up the most important part of creating a private key to secure your Bitcoin.
61
00:03:42.730 --> 00:03:43.130
So
62
00:03:43.530 --> 00:03:44.090
I'm
63
00:03:44.890 --> 00:03:45.530
yeah.
64
00:03:46.330 --> 00:03:49.690
I was gonna say we dished out a lot of shit. We've we've
65
00:03:50.010 --> 00:03:55.610
defended cold card many times in the past, and it is very obvious now in hindsight that
66
00:03:56.225 --> 00:03:57.105
that was
67
00:03:58.225 --> 00:04:01.585
that defense and the the endorsement was
68
00:04:02.225 --> 00:04:04.145
was on ill gotten confidence.
69
00:04:05.985 --> 00:04:09.505
To be clear, this is happening on August 3.
70
00:04:10.300 --> 00:04:11.260
The current time
71
00:04:11.740 --> 00:04:14.060
the current block height since we
72
00:04:15.500 --> 00:04:21.020
I see a comment in the chat, which is why I'm bringing it up. The current block height is
73
00:04:23.055 --> 00:04:26.255
nine six zero eight eight five. Like I said, the vulnerability,
74
00:04:26.575 --> 00:04:27.535
the bug
75
00:04:28.815 --> 00:04:32.255
came to our attention after the last RHR.
76
00:04:32.655 --> 00:04:33.295
And
77
00:04:33.775 --> 00:04:40.180
and and the reason why the podcast is now instead of Friday is because
78
00:04:40.819 --> 00:04:42.340
as far as I'm concerned,
79
00:04:42.580 --> 00:04:49.620
getting this information to as many people as possible was most important, and I don't think the podcast actually is the way to do that.
80
00:04:50.500 --> 00:04:54.885
But, also, at the same time, we're not gonna sit here and, you know, wait till Thursday
81
00:04:55.205 --> 00:04:57.925
to discuss this. Like, we need to be on air now.
82
00:04:59.125 --> 00:05:01.125
And and so Marty and I
83
00:05:01.525 --> 00:05:04.085
have been content contacting people nonstop,
84
00:05:04.325 --> 00:05:04.965
tweeting out
85
00:05:05.240 --> 00:05:06.600
Noster posts,
86
00:05:06.920 --> 00:05:08.360
sending out emails,
87
00:05:08.440 --> 00:05:11.240
trying to notify as many people as possible to move.
88
00:05:13.720 --> 00:05:14.840
And and
89
00:05:15.800 --> 00:05:18.920
look. Okay. So that's that's why the podcast is going on right now.
90
00:05:21.074 --> 00:05:24.435
I am so fucking sorry for ever recommending gold cards.
91
00:05:24.595 --> 00:05:29.555
I honestly, until Thursday night, thought they were the best, most secure hardware on the market.
92
00:05:30.435 --> 00:05:34.194
I like I've I've if if you've been a freak for a long time, you know
93
00:05:36.080 --> 00:05:39.200
that that I've like, recommending
94
00:05:39.280 --> 00:05:47.200
recommending how you how you cold store your Bitcoin is is a very sacred recommendation to me. Okay. Since 2019
95
00:05:48.075 --> 00:05:51.915
since 2019 when we did our first cold card workshop
96
00:05:53.835 --> 00:05:56.635
in New York at Chaincode Labs,
97
00:05:58.235 --> 00:06:04.300
and we did it with Evan Kaludis and Beefsteak Josh and a bunch of other people. James O'Byrne was there.
98
00:06:07.100 --> 00:06:11.900
I think that was December 2019, and we did it with m k twos and m k threes. We
99
00:06:12.220 --> 00:06:14.940
walked people through dice rolls and passphrases.
100
00:06:17.315 --> 00:06:20.515
And at that point, it was with Wasabi. I think it was pre Sparrow.
101
00:06:20.915 --> 00:06:25.235
Since then, I've been convinced that it is the best hardware on the market.
102
00:06:27.715 --> 00:06:28.675
And
103
00:06:30.010 --> 00:06:36.330
that's been seven years. Seven years of of me and Marty and many others recommending this thing.
104
00:06:38.010 --> 00:06:42.650
And I looked at it, and I always just like, I would look at the cold card and be like, this is,
105
00:06:43.345 --> 00:06:44.465
you know, our
106
00:06:46.305 --> 00:06:47.825
fortress of solitude.
107
00:06:48.385 --> 00:06:52.865
And it was it was a piece of shit. The thing was completely insecure.
108
00:06:53.505 --> 00:07:00.400
And I know we preach about don't trust, verify all the time, about self custody. I still think self custody is extremely important.
109
00:07:00.800 --> 00:07:02.639
I think it's more important than ever.
110
00:07:06.080 --> 00:07:14.215
And that was one of the reasons why I loved cold cards was because you could add dice rolls. Right? You couldn't add dice rolls to other most other hardware wallets.
111
00:07:16.055 --> 00:07:21.255
That why we we talked about pass raises all the time, why we talked about multisig.
112
00:07:22.055 --> 00:07:25.655
But over the years, there was a level of complacency that happened
113
00:07:26.850 --> 00:07:31.330
My with my myself, I'll speak for myself. There was a level of complacency that happened.
114
00:07:33.970 --> 00:07:38.450
And I I I believed that someone just generating with a cold card
115
00:07:39.315 --> 00:07:49.315
without adding dice rolls, without using a pass raise, without using multisay because that was the defaults of treasurer and ledger was better than nothing, and it was a good first step. And I was completely fucking wrong.
116
00:07:49.715 --> 00:07:51.074
And we talked about don't trust, verify,
117
00:07:53.120 --> 00:07:59.600
but I I I trust it. I put way too much trust in NVK and Peter and CoinKite,
118
00:07:59.919 --> 00:08:03.840
more trust than I put in pretty much anyone except for family members,
119
00:08:04.400 --> 00:08:06.159
and they completely betrayed that trust.
120
00:08:07.085 --> 00:08:08.685
I'm I'm I'm
121
00:08:09.805 --> 00:08:12.845
I'm angry. I'm angry at myself. I'm angry at them.
122
00:08:14.045 --> 00:08:15.485
I'm I'm I'm
123
00:08:16.765 --> 00:08:22.445
I I have a bit of a crisis of confidence. Like, all the all all these people are asking me where do they move to next. I don't, you know,
124
00:08:24.090 --> 00:08:26.570
I don't want to recommend something bad,
125
00:08:27.850 --> 00:08:29.290
but you need to move.
126
00:08:29.770 --> 00:08:30.970
You need to move.
127
00:08:32.970 --> 00:08:34.410
Never use cold cards again.
128
00:08:35.355 --> 00:08:36.714
I I it's it's
129
00:08:38.635 --> 00:08:42.955
this this bug has been there for five years. There's been a ticking time bomb for five years
130
00:08:43.834 --> 00:08:47.915
that at any point, all of our funds could get removed if you didn't have a passphrase,
131
00:08:48.490 --> 00:08:51.050
dice rolls, multisig, any of those.
132
00:08:52.410 --> 00:08:59.930
And like I said, even if you know, maybe you did maybe like, Coldcard has this feature where it's, okay, you can add dice rolls on top of their randomness.
133
00:09:00.894 --> 00:09:17.340
Right? They they're supposed to generate secure randomness themselves, which they weren't. That's the bug. And then you could add dice rolls on top. So some people just did, like, 12 dice rolls on top, 14 dice rolls on top. That's not enough to protect you long term, but it might be enough to protect you right now. You might still have funds on there. You don't remember
134
00:09:17.820 --> 00:09:19.740
how many dice rolls you did. Move it.
135
00:09:20.220 --> 00:09:22.060
Regardless, I would move it.
136
00:09:25.980 --> 00:09:27.660
Yeah. I I don't know. I'm
137
00:09:28.345 --> 00:09:36.505
I'm sad. I'm fucking broken. I've I've this fucking sucks. I'm I'm sorry, freaks. This really, really
138
00:09:37.385 --> 00:09:38.665
fucking sucks.
139
00:09:40.025 --> 00:09:40.425
Yeah.
140
00:09:43.870 --> 00:09:44.589
And
141
00:09:46.190 --> 00:09:49.230
just reflecting over the weekend is a
142
00:09:51.709 --> 00:09:53.630
lot of that. And I think,
143
00:09:54.670 --> 00:10:00.324
again, holding ourselves to account, like, the don't trust verify definitely lapsed on that. And
144
00:10:01.045 --> 00:10:05.524
there's so many social psychological aspects of this to group think
145
00:10:07.524 --> 00:10:08.485
in groups
146
00:10:08.644 --> 00:10:09.685
sort of catering.
147
00:10:11.060 --> 00:10:13.860
And it's obvious if we're being objective to that.
148
00:10:14.339 --> 00:10:18.339
That those types of actions, that psychology was certainly a play too.
149
00:10:21.459 --> 00:10:22.500
I'm I'm sorry
150
00:10:23.060 --> 00:10:25.620
for playing into that too. It's
151
00:10:28.205 --> 00:10:31.645
a it's a hard game. And to your point too, like, I I don't think I'll ever
152
00:10:32.205 --> 00:10:33.005
recommend
153
00:10:34.285 --> 00:10:38.285
the custody solution directly for people. I mean, layout options.
154
00:10:38.365 --> 00:10:39.485
So do your own research,
155
00:10:40.360 --> 00:10:42.200
figure out what works best for you.
156
00:10:47.240 --> 00:10:48.840
Yeah. It's a
157
00:10:50.040 --> 00:10:56.645
it's a sad day in there, and many good people have lost their loss. That's their life savings. That's the
158
00:10:57.205 --> 00:10:58.805
the biggest gut punch.
159
00:11:03.445 --> 00:11:06.885
And it just feels okay. Again, horror is the only
160
00:11:07.620 --> 00:11:10.660
word I can use to describe it. I don't think I could feel any worse.
161
00:11:12.740 --> 00:11:15.460
And to Matt's point, like, if you
162
00:11:16.260 --> 00:11:18.420
have a cold card, any version,
163
00:11:18.900 --> 00:11:20.340
any model after
164
00:11:21.245 --> 00:11:25.165
March 2021. Don't risk it. And that's that's the other thing
165
00:11:25.885 --> 00:11:28.285
that I learned over the weekend, trying
166
00:11:28.765 --> 00:11:37.700
to help people migrate as many people don't don't have the confidence of what they did originally. They don't know. They said, I rolled dice, I remember, but I don't know how many. And
167
00:11:38.580 --> 00:11:42.580
if if you're in that that boat, just move. Figure out ways
168
00:11:43.220 --> 00:11:43.940
to
169
00:11:44.580 --> 00:11:47.700
to move fast. It's a ticking time bomb.
170
00:11:48.500 --> 00:11:50.660
If you're just hearing this for the first time,
171
00:11:52.435 --> 00:11:56.595
basically, the the alarm bell has gone off, and now people who understand
172
00:11:57.075 --> 00:12:00.755
the vulnerability are are spinning up GPUs to to grind through
173
00:12:01.235 --> 00:12:03.715
seed phrases and pass phrases.
174
00:12:04.435 --> 00:12:07.315
And it's only a matter of time before
175
00:12:08.850 --> 00:12:13.730
yours gets exposed if if you don't have sufficient entropy from additional
176
00:12:14.130 --> 00:12:15.970
dice rolls and and passphrase.
177
00:12:36.130 --> 00:12:37.089
I think
178
00:12:38.450 --> 00:12:39.089
one
179
00:12:40.769 --> 00:12:45.170
one phrase that's been in my mind since Thursday night is is pride comes before the fall.
180
00:12:46.505 --> 00:12:47.865
And I think if you know
181
00:12:49.705 --> 00:12:52.585
how Coldcard has been antagonistic in the market,
182
00:12:53.144 --> 00:12:54.425
very prideful,
183
00:12:55.305 --> 00:12:59.305
exuding a ton of confidence in their technical prowess, it was all
184
00:13:00.220 --> 00:13:01.180
hot air.
185
00:13:02.460 --> 00:13:04.140
And incredibly fucked up.
186
00:13:04.460 --> 00:13:05.180
Yeah.
187
00:13:12.940 --> 00:13:15.180
Yeah. I mean, look. I
188
00:13:16.875 --> 00:13:21.355
I linked in the show notes. I like I said, especially now, but,
189
00:13:25.035 --> 00:13:27.515
you know, I don't I I
190
00:13:27.515 --> 00:13:32.395
don't know, like, where it's safe to move. I don't feel comfortable
191
00:13:31.410 --> 00:13:33.810
telling people how to move, where to move.
192
00:13:35.010 --> 00:13:37.089
The one positive note is the
193
00:13:38.610 --> 00:13:48.325
the Bitcoin community has really rallied together over this past weekend. People have been up all night talking to people, moving people through, being active on acts and noster with guides,
194
00:13:49.605 --> 00:13:51.525
and that's been really amazing to see.
195
00:13:52.325 --> 00:13:56.885
So I linked to a very extensive guide by Rob Hamilton on
196
00:13:58.190 --> 00:14:00.190
who's vulnerable, how you're vulnerable,
197
00:14:00.510 --> 00:14:02.430
what are your options, where to move.
198
00:14:03.870 --> 00:14:04.910
So definitely
199
00:14:04.990 --> 00:14:10.510
go to that link. And if you're listening to this live and, you know, we won't probably have it up on our podcast feeds for,
200
00:14:11.754 --> 00:14:13.915
you know, like, another hour or so.
201
00:14:14.954 --> 00:14:18.714
Just go to Rob one ham on x, and you can see the
202
00:14:19.274 --> 00:14:20.714
I think he has it pinned.
203
00:14:22.795 --> 00:14:23.435
But
204
00:14:24.680 --> 00:14:30.920
that's really good to see. And, also, on the topic of Rob Hamilton, him and, like, a bunch of crack devs are currently
205
00:14:32.760 --> 00:14:35.400
running frontier open source models,
206
00:14:35.560 --> 00:14:35.960
k three.
207
00:14:36.965 --> 00:14:40.725
And, also, it seems like they've they got access to OpenAI's,
208
00:14:40.725 --> 00:14:42.085
you know, walled garden,
209
00:14:42.565 --> 00:14:44.805
Cypher program, and they've just been
210
00:14:45.525 --> 00:14:46.965
just burning through
211
00:14:48.005 --> 00:14:49.845
AI tokens to
212
00:14:51.170 --> 00:14:55.010
hit every major Bitcoin project. I think it's a 100 and plus
213
00:14:55.410 --> 00:14:57.570
open source Bitcoin projects right now.
214
00:14:58.210 --> 00:15:00.130
OpenSats will be funding that,
215
00:15:01.330 --> 00:15:03.890
but up until this point, it's been funding it out of pocket.
216
00:15:05.665 --> 00:15:07.425
Pretty amazing to see
217
00:15:08.704 --> 00:15:10.705
the community come together like that.
218
00:15:11.185 --> 00:15:11.985
I
219
00:15:13.584 --> 00:15:17.825
don't I don't think self custody is dead. I think the idea of, like, a
220
00:15:18.689 --> 00:15:20.449
magic bullet for beginners,
221
00:15:21.009 --> 00:15:27.089
for high value self custody is is definitely dead. Like, I think people should be way more paranoid.
222
00:15:27.410 --> 00:15:28.290
Think a lot of
223
00:15:30.050 --> 00:15:33.730
and that that's the key lesson here. You know? Multi vendor, multisig,
224
00:15:35.145 --> 00:15:36.185
passphrases,
225
00:15:36.185 --> 00:15:41.705
a lot of things that could cause foot guns. If you can't handle that, then maybe self custody is not for you.
226
00:15:43.065 --> 00:15:44.185
I just don't
227
00:15:46.520 --> 00:15:53.640
I question. I need time to reflect. I question if I should be involved in that process of of helping people self custody anymore. Like, I feel
228
00:15:55.320 --> 00:15:58.440
I'm I'm I'm grateful that we've recommended
229
00:15:59.285 --> 00:16:04.725
many times in the past dice rolls and pass raises and multisig and trusted third parties or security holes.
230
00:16:05.204 --> 00:16:16.790
But, yes, also mistakes were made in terms of thinking that CoinKite was producing a secure project. Like, I trust I I trusted them that they were making a very secure hardware,
231
00:16:17.030 --> 00:16:18.710
and and they they weren't.
232
00:16:22.550 --> 00:16:23.910
And, yeah, so I just
233
00:16:24.790 --> 00:16:29.270
seven years, man, and the bug has been sitting there for five years. Like, I just
234
00:16:29.654 --> 00:16:31.175
I really need to
235
00:16:32.375 --> 00:16:34.295
I I really need to
236
00:16:36.454 --> 00:16:37.815
reflect on on
237
00:16:38.295 --> 00:16:39.334
on that.
238
00:16:39.654 --> 00:16:42.774
And I think we all do as an as an industry in a big way too.
239
00:16:43.529 --> 00:16:44.170
Yeah.
240
00:16:44.810 --> 00:16:47.850
And this is by no means a way to
241
00:16:49.370 --> 00:16:51.210
cover for Coincite.
242
00:16:52.170 --> 00:16:54.330
No. Fluff that. Mistake was
243
00:16:55.610 --> 00:16:57.290
so, like, mistake was all theirs.
244
00:16:59.644 --> 00:17:03.324
But it it is undeniable, like, with the age of AI.
245
00:17:04.365 --> 00:17:06.764
You need to prepare for this if you're running
246
00:17:07.885 --> 00:17:11.245
a company that is securing people's Bitcoin. And so that's
247
00:17:11.645 --> 00:17:15.450
that has been very encouraging to see over the weekend.
248
00:17:16.490 --> 00:17:19.050
I mean, again, I don't want it to come off as AI
249
00:17:19.530 --> 00:17:26.570
discovery. This bug was gonna be found regardless at some point. It was a fucking ticking time bomb, but it does seem like AI might have found it.
250
00:17:26.970 --> 00:17:27.370
I mean,
251
00:17:29.245 --> 00:17:32.924
NVK rightfully deserves a lot of shit for his license change.
252
00:17:36.365 --> 00:17:38.445
And, I mean, I know there's
253
00:17:38.445 --> 00:17:41.164
thousands of hours of us talking about things, so
254
00:17:41.565 --> 00:17:43.085
people are finding clips everywhere.
255
00:17:43.940 --> 00:17:46.820
There are also clips of me giving him shit about it publicly,
256
00:17:47.860 --> 00:17:49.620
but the source was viewable.
257
00:17:50.260 --> 00:17:54.179
And so there was a lot of people that could have looked at it if they wanted to.
258
00:17:55.140 --> 00:17:57.380
Obviously, it was way above our fucking pay grade.
259
00:18:00.125 --> 00:18:04.685
And we don't know exactly when it got discovered. It does seem like it might have gotten discovered this week with
260
00:18:05.645 --> 00:18:10.205
like, when the attack when the attacker really discovered it with AI with the help of AI.
261
00:18:12.160 --> 00:18:15.760
But, I mean, I remember, like, Ledger attacking it with fucking
262
00:18:15.840 --> 00:18:17.440
space lasers and shit,
263
00:18:17.760 --> 00:18:19.600
and they were just looking in the wrong place.
264
00:18:19.920 --> 00:18:20.800
You know? If you're
265
00:18:21.760 --> 00:18:23.520
I don't know. I just,
266
00:18:27.485 --> 00:18:28.445
yeah. I just
267
00:18:30.845 --> 00:18:33.005
it's it's it's brutal.
268
00:18:34.205 --> 00:18:36.685
It's fucking brutal, and it was literally and
269
00:18:37.640 --> 00:18:41.880
this one of the scariest things here, by the way, is, like, if a good guy had found it
270
00:18:42.440 --> 00:18:44.440
first before the attacker,
271
00:18:44.840 --> 00:18:46.360
it would have been really hard,
272
00:18:47.640 --> 00:18:48.840
if not impossible,
273
00:18:49.080 --> 00:18:50.040
to fix it, I think,
274
00:18:50.855 --> 00:18:57.015
because people needed to regenerate seeds. So it would have been a race regardless. It would have been strictly better. Don't get me wrong.
275
00:18:58.535 --> 00:19:00.135
But it would have been a race
276
00:19:00.375 --> 00:19:01.735
race altogether.
277
00:19:02.615 --> 00:19:03.495
And
278
00:19:04.130 --> 00:19:14.610
I don't know. This just fucking sucks. Like, I was on the phone with friends and family. I recommended Coldheart to so many fucking people. People lost every some people lost everything.
279
00:19:17.855 --> 00:19:25.455
Fortune like, the a lot more people could have lost money if it wasn't for the dice rolls and the multisig and the pass raises, but please move.
280
00:19:26.335 --> 00:19:30.815
Like, you might not be safe. You might just have bought yourself time.
281
00:19:34.410 --> 00:19:36.970
And it yeah. It's ongoing. This isn't
282
00:19:38.010 --> 00:19:39.929
this isn't going to end until
283
00:19:40.010 --> 00:19:40.730
every
284
00:19:41.530 --> 00:19:43.530
public ash address associated
285
00:19:43.530 --> 00:19:46.890
with one of those private keys that was generated on the cold card has been
286
00:19:48.894 --> 00:19:54.014
attempted to be brute force. The private key of that public address is attempted to be brute force. And it's
287
00:19:54.975 --> 00:19:56.654
hard to know exactly how many
288
00:19:57.934 --> 00:19:58.575
are out there.
289
00:20:01.090 --> 00:20:03.330
I'm seeing a comment in the
290
00:20:03.810 --> 00:20:07.810
in Noster from Aldoka. This is was the worst weekend of my life.
291
00:20:08.770 --> 00:20:09.890
I'm fucking
292
00:20:10.290 --> 00:20:11.250
destroyed, man.
293
00:20:14.445 --> 00:20:16.044
What a fucking mess.
294
00:20:16.525 --> 00:20:17.244
Yeah.
295
00:20:17.725 --> 00:20:18.524
I
296
00:20:24.684 --> 00:20:31.120
see a comment about MBK saying they did an LLM audit on their code and didn't find the bug. I don't you know?
297
00:20:31.520 --> 00:20:34.159
He's, the last person I trust right now. But
298
00:20:35.040 --> 00:20:44.480
I know from other people that they did as well, and and they didn't find anything up until this point. And there is definitely some piece of it, by the way, that's, like, once one finds it
299
00:20:44.955 --> 00:20:49.674
and people know where to look, then that's when they start finding it in, eight minutes or whatever.
300
00:20:50.315 --> 00:20:59.515
I saw the frost the frostnap team, like, seven weeks ago used Opus four eight, and it didn't find it. I don't think I found the exact bug, but it did highlight that
301
00:21:01.440 --> 00:21:03.200
entropy was insufficient.
302
00:21:04.880 --> 00:21:08.160
No. I don't even think it did. Right? He said he was close, but it wasn't.
303
00:21:17.165 --> 00:21:21.644
And, yes, technically, if you roll a 100 dice correctly
304
00:21:21.645 --> 00:21:24.205
on your own, you will have good entropy.
305
00:21:24.684 --> 00:21:25.245
I
306
00:21:25.680 --> 00:21:31.440
I mean, I saw people trying to call me calling me out of next because by by the way, my guides have always been open source,
307
00:21:31.600 --> 00:21:34.320
and I'm not trying to hide anything. I've left the GitHub
308
00:21:35.600 --> 00:21:45.904
the GitHubs are up on on on when I change things and and how they've how they've been modified over time. And in 2025, I removed my dice roll recommendation.
309
00:21:46.705 --> 00:21:52.705
And it was because my understanding was people were footgunning themselves by rolling too few dice, and I felt horrible by people losing money.
310
00:21:53.530 --> 00:21:59.450
In hindsight, I feel like a fucking idiot. I mean, I feel like a fucking idiot forever recommending cold card. I can't believe I thought this thing
311
00:21:59.770 --> 00:22:02.889
I can't believe I thought this thing was the most secure hardware on the market.
312
00:22:04.650 --> 00:22:06.010
I'm I'm I'm
313
00:22:07.290 --> 00:22:08.170
what the fuck?
314
00:22:10.635 --> 00:22:11.274
But
315
00:22:12.475 --> 00:22:13.514
here we are.
316
00:22:14.715 --> 00:22:15.355
No.
317
00:22:15.515 --> 00:22:17.835
We will get through this. We have to get through this.
318
00:22:25.440 --> 00:22:27.600
Yeah. I'm reflecting. Just trying to think of
319
00:22:30.160 --> 00:22:31.279
other things.
320
00:22:32.800 --> 00:22:35.840
Yeah. I mean, one one positive note, I guess,
321
00:22:36.455 --> 00:22:37.174
is
322
00:22:37.255 --> 00:22:43.815
a block. So I also linked the best write up, I think, is by the team at block, specifically, like, Bitkey team,
323
00:22:44.775 --> 00:22:49.095
which also who knows what kind of AI they have available to them, like like, top top,
324
00:22:49.960 --> 00:22:50.840
like,
325
00:22:50.840 --> 00:22:52.999
closed US models. But
326
00:22:53.720 --> 00:22:56.279
they did a great write up that's linked in the show notes.
327
00:22:56.759 --> 00:23:01.799
But they believe they figured out that the attacker was pinging a centralized
328
00:23:01.799 --> 00:23:06.174
API provider to look up addresses, and they've reported that to the authorities.
329
00:23:06.895 --> 00:23:09.294
So there is an investigation going on with thief.
330
00:23:11.615 --> 00:23:12.575
So that's happening.
331
00:23:19.590 --> 00:23:23.269
Yeah. And it's just getting messy too because now there's white hat hackers
332
00:23:24.230 --> 00:23:27.429
beginning to sweep coins. So it's hard to know at this point
333
00:23:28.390 --> 00:23:31.270
after the initial wave what what is what
334
00:23:31.270 --> 00:23:32.470
is actually being swept
335
00:23:32.674 --> 00:23:35.315
by nefarious or white hat hackers. And
336
00:23:36.355 --> 00:23:38.354
this is literally a worst case scenario,
337
00:23:38.835 --> 00:23:39.794
Chernobyl
338
00:23:41.475 --> 00:23:42.114
event
339
00:23:43.235 --> 00:23:44.115
within Bitcoin.
340
00:23:47.240 --> 00:23:49.159
And if you that that that was
341
00:23:50.360 --> 00:23:53.960
one of the most distressing things over the weekend, literally just going through
342
00:23:54.440 --> 00:24:00.039
my contact list and trying to think of anybody that I know that either recommended cold card to our nose in the Bitcoin.
343
00:24:00.440 --> 00:24:05.125
And they may have sent to the show and I'm not aware of. It's like, hey. And
344
00:24:05.365 --> 00:24:08.085
they're like, and the it's fucking
345
00:24:09.125 --> 00:24:11.684
first week of August, which is, like, peak vacation
346
00:24:12.245 --> 00:24:13.125
season.
347
00:24:13.125 --> 00:24:14.485
Like, I think
348
00:24:15.550 --> 00:24:22.910
at least three people that I was helping move stuff over the weekend were on the other side of the planet away from their private keys.
349
00:24:24.910 --> 00:24:27.230
Yeah. I mean, geographically
350
00:24:27.230 --> 00:24:28.750
distributed multisig
351
00:24:28.945 --> 00:24:31.025
really makes you start to think too. Right?
352
00:24:32.065 --> 00:24:35.345
I mean, we were scrambling to make sure open such funds were safe.
353
00:24:35.905 --> 00:24:36.945
They are.
354
00:24:39.905 --> 00:24:55.289
I see one question. Do you have brick devices? I'm seeing reports that people are their devices might be getting bricked by the new firmware that's getting released. I do not have confirmation on that. But just don't Don't update the firmware. Yeah. I've been consistent since this news broke on Thursday.
355
00:24:55.690 --> 00:25:00.904
Fuck this shit. Do not upgrade your firmware. Do not stay with CoinKite. Transfer your funds off.
356
00:25:01.625 --> 00:25:04.105
Just transfer your funds off to another wallet.
357
00:25:04.345 --> 00:25:07.865
And if you can't get hardware in time, Rob Hamilton has has
358
00:25:08.425 --> 00:25:12.505
his own advice that, like I said, we link to in the meantime,
359
00:25:13.070 --> 00:25:16.429
but just transfer off ahead of time. I I I
360
00:25:17.150 --> 00:25:18.350
I I
361
00:25:18.990 --> 00:25:25.309
cannot support this company at all. This is anything they're doing right now is I fuck all of that shit.
362
00:25:27.215 --> 00:25:30.655
Move your funds to a a secure place. I also see another
363
00:25:31.295 --> 00:25:32.335
comment here.
364
00:25:32.975 --> 00:25:35.855
Is ten thirty one cutting off VC funding for CoinKite?
365
00:25:36.335 --> 00:25:45.759
The the investment was years ago where a minority investor, less than 10%, the majority of the company is owned and controlled by
366
00:25:46.240 --> 00:25:52.000
MBK and Peter, and we fucked up. We just completely fucked up. We we it was misplaced trust.
367
00:25:52.534 --> 00:25:53.414
Way,
368
00:25:53.414 --> 00:25:56.134
way, way too much trust, and and and
369
00:25:57.015 --> 00:25:58.214
and and
370
00:25:58.294 --> 00:25:59.494
we got fucked.
371
00:26:00.855 --> 00:26:04.294
And I'm I'm incredibly sorry to to both
372
00:26:04.455 --> 00:26:07.414
ten thirty one investors and to any users who were affected.
373
00:26:10.350 --> 00:26:16.669
And, you know, anyone to even if you didn't lose your funds, anyone I recommended this shit to that has been scrambling for the last
374
00:26:17.070 --> 00:26:20.429
three days to fucking move your shit in a panic
375
00:26:20.905 --> 00:26:22.024
It's horrible.
376
00:26:24.505 --> 00:26:25.225
Yeah.
377
00:26:28.825 --> 00:26:31.385
I mean, beyond Block and Rob, I mean,
378
00:26:35.130 --> 00:26:38.250
just to give props to to people like Wicked, BTC,
379
00:26:38.250 --> 00:26:39.850
mister Huddl, Shinobi,
380
00:26:39.930 --> 00:26:40.890
many others
381
00:26:41.850 --> 00:26:47.770
who had seen Kevin Loak of Liana? Yeah. Yeah. Kevin There's too many to mention. Yeah.
382
00:26:50.025 --> 00:26:55.465
There's there's too many to mention, but it's it's been pretty wonderful seeing the
383
00:26:56.425 --> 00:26:59.865
community come together at the very least, like a tiny bright
384
00:27:00.025 --> 00:27:00.825
spot
385
00:27:01.225 --> 00:27:01.785
here.
386
00:27:12.920 --> 00:27:13.400
Yep.
387
00:27:14.545 --> 00:27:21.345
Let's I see Jack Mitchell. If we did a 100 plus dice rolls and entered them, is that correct? Yeah. Theoretically.
388
00:27:22.305 --> 00:27:24.865
Like, if you fucked it up, though, it's not.
389
00:27:25.185 --> 00:27:25.745
So
390
00:27:28.865 --> 00:27:30.960
be aware. You know? I
391
00:27:31.920 --> 00:27:34.080
I wouldn't even use this thing with
392
00:27:34.480 --> 00:27:35.600
sufficient dice rolls.
393
00:27:42.255 --> 00:27:49.695
Like, every like, keep getting so many questions. Like, if I did this and that, blah blah blah. It's like, if you still have your funds and they're still on a cold card,
394
00:27:49.935 --> 00:27:52.174
you should move them. But, also,
395
00:27:52.735 --> 00:27:53.055
you know,
396
00:27:53.790 --> 00:27:56.190
been giving bad advice on
397
00:27:56.430 --> 00:28:00.030
using cold card in the first place for seven years, so take it with a fucking grain of salt.
398
00:28:09.165 --> 00:28:17.885
Shout out to Dumbder on Twitter. He says, I look like I'm watching the ghost of Lindsey Graham sucking off my dad. That was an interesting one on Twitter.
399
00:28:18.690 --> 00:28:20.210
We got the best supporters.
400
00:28:20.450 --> 00:28:21.409
Yeah. It's
401
00:28:29.970 --> 00:28:32.370
yeah. I mean, we've dished out a lot of shit
402
00:28:32.825 --> 00:28:34.985
in the years over the years. I mean, the people
403
00:28:39.785 --> 00:28:43.785
you know, when Block five blew up, danced on some graves there when
404
00:28:46.300 --> 00:28:48.540
other companies have made decisions,
405
00:28:50.380 --> 00:28:54.059
made comments, and it's just I mean, again, going back to reflecting all weekend, I think.
406
00:28:54.860 --> 00:28:56.860
It's one thing I'm gonna come away with is
407
00:28:58.285 --> 00:29:00.445
it's just this shit is hard.
408
00:29:02.445 --> 00:29:04.044
No one is perfect, and
409
00:29:06.525 --> 00:29:08.125
just be mindful of that.
410
00:29:08.525 --> 00:29:09.644
Be graceful.
411
00:29:09.885 --> 00:29:10.925
I'm praying a lot too.
412
00:29:12.240 --> 00:29:13.760
The people lost funds for
413
00:29:16.880 --> 00:29:18.480
people are losing confidence. It's
414
00:29:20.320 --> 00:29:23.440
again, this is I can't it's it's hard. I'm
415
00:29:24.495 --> 00:29:26.174
sleepless, speechless, and
416
00:29:30.655 --> 00:29:31.055
it's
417
00:29:35.775 --> 00:29:37.615
can't think of anything worse,
418
00:29:38.815 --> 00:29:39.375
literally.
419
00:29:40.430 --> 00:29:41.389
The whole
420
00:29:41.870 --> 00:29:43.309
point of Bitcoin and
421
00:29:43.950 --> 00:29:46.509
is the ability to have confidence that you can
422
00:29:47.390 --> 00:29:48.830
secure your own funds.
423
00:29:50.910 --> 00:29:53.870
And to be clear, like, that ability has not changed.
424
00:29:56.395 --> 00:29:58.794
Securing your own funds on cold card is
425
00:29:59.595 --> 00:30:02.474
not possible after March 2021
426
00:30:03.755 --> 00:30:07.275
without sufficient dice rolls. But, again, how can you have confidence?
427
00:30:07.435 --> 00:30:08.635
You need there's that
428
00:30:09.260 --> 00:30:12.940
introduction of the doubt of your ability to do that in the past is a
429
00:30:15.019 --> 00:30:17.979
a reason to move. No. And that was there was
430
00:30:19.419 --> 00:30:28.515
a couple people I helped this weekend that had a 100 dice rolls and that and that. It's like, I I don't know if it is. Like, I can't trust that you put that in
431
00:30:29.235 --> 00:30:29.955
perfectly.
432
00:30:32.115 --> 00:30:32.755
Yeah.
433
00:30:33.395 --> 00:30:34.435
Where do we go from here?
434
00:30:45.960 --> 00:30:46.999
Forward.
435
00:30:47.880 --> 00:30:50.840
I mean, I still think Bitcoin is needed now more than ever.
436
00:30:55.695 --> 00:30:57.615
You know, this is this is
437
00:30:58.655 --> 00:30:59.775
one company
438
00:31:00.175 --> 00:31:00.895
making
439
00:31:01.935 --> 00:31:04.015
a a a a really,
440
00:31:04.175 --> 00:31:05.055
really bad mistake.
441
00:31:09.130 --> 00:31:15.049
I think people will it will reflect on Bitcoin, but it shouldn't in long term. I I I hope it won't.
442
00:31:16.970 --> 00:31:17.610
But
443
00:31:18.330 --> 00:31:19.770
fuck, man. I'm
444
00:31:20.665 --> 00:31:21.625
there
445
00:31:22.425 --> 00:31:24.424
look. There's like I said, there's a lot of
446
00:31:25.465 --> 00:31:29.784
good people working out there on both trying to diagnose more bugs and
447
00:31:30.505 --> 00:31:31.784
make self custody
448
00:31:32.025 --> 00:31:33.465
more secure and easier.
449
00:31:34.840 --> 00:31:40.359
There's clearly no shortcuts, and I will say one of my mistakes was thinking maybe
450
00:31:40.679 --> 00:31:44.919
maybe there was a shortcut where you just had this really secure hardware wallet.
451
00:31:46.855 --> 00:31:51.174
And you could take a little bit less personal responsibility, you could be a little less technically
452
00:31:54.695 --> 00:31:57.815
competent and fucked up, man. Just fucked up.
453
00:31:58.215 --> 00:31:59.894
I trust put
454
00:32:00.090 --> 00:32:06.570
I trusted. I trusted I I shouldn't I shouldn't have trusted NVK, Peter, and CoinKite. I fucked up big time.
455
00:32:07.049 --> 00:32:07.690
Yeah.
456
00:32:07.850 --> 00:32:08.489
And
457
00:32:10.169 --> 00:32:13.770
I'm seeing a a comment here. You can verify dice
458
00:32:14.955 --> 00:32:22.394
roll. I think he's trying to say dice rolls, but you can, but you you can only verify when you're doing it if you didn't do that and have confidence.
459
00:32:22.555 --> 00:32:26.715
No. You can't. His point is there's two features on cold card,
460
00:32:27.675 --> 00:32:41.440
which is part of the reason that like, it was it was good from, like, a positive from, like, a features point of view, but it created an additional foot gun situation because people confuse the two features. But there's one that was supposed to add device entropy plus
461
00:32:42.345 --> 00:32:44.585
dice rolls. So dice rolls were additive.
462
00:32:45.865 --> 00:32:48.345
And my understanding is that's not verifiable.
463
00:32:48.505 --> 00:32:52.345
But if you do dice rolls only, then that is verifiable.
464
00:32:53.065 --> 00:33:02.539
But if you did dice plus entropy, even though the entropy was fucking horrible, you're still you you're you're you're still safer than the people that didn't, but you have to move funds as soon as fucking possible.
465
00:33:04.059 --> 00:33:05.740
But like I said, I I just
466
00:33:06.700 --> 00:33:11.980
if you have to ask the question, you should move funds. Just don't use this product. It's a horrible fucking product.
467
00:33:12.815 --> 00:33:16.015
It failed at the one thing it was it needed to do.
468
00:33:39.120 --> 00:33:39.440
Yeah.
469
00:33:41.014 --> 00:33:42.854
A lot of interesting discussion about
470
00:33:43.975 --> 00:33:49.894
the future of Bitcoin custody. I think multivendor multisig is a theme that many people are pointing out as
471
00:33:51.575 --> 00:33:54.695
maybe a minimum viable requirement to
472
00:33:55.174 --> 00:33:57.414
have confidence along
473
00:33:56.730 --> 00:33:59.929
long term storage. Just a lot of interesting discussion around
474
00:34:00.250 --> 00:34:03.450
BIP 39 and whether or not that is that
475
00:34:03.450 --> 00:34:05.049
is actually a standard that
476
00:34:05.610 --> 00:34:09.290
people should have that that people should be
477
00:34:10.415 --> 00:34:11.695
organizing around.
478
00:34:14.175 --> 00:34:16.575
So, yeah, I think I think there's gonna be a lot of
479
00:34:17.855 --> 00:34:18.655
introspection
480
00:34:18.655 --> 00:34:19.295
and,
481
00:34:19.935 --> 00:34:21.935
hopefully, debates and conversation
482
00:34:21.935 --> 00:34:25.215
about about what we're actually doing here, getting back to focusing on
483
00:34:25.680 --> 00:34:28.720
securing this stuff. It's goddamn shame.
484
00:34:29.359 --> 00:34:31.120
It took something like this to
485
00:34:31.599 --> 00:34:33.840
to push the conversation in this direction.
486
00:34:36.240 --> 00:34:38.400
I mean, yeah, the other like, passphrases
487
00:34:39.359 --> 00:34:40.320
use passphrases.
488
00:34:40.835 --> 00:34:41.714
Even
489
00:34:41.714 --> 00:34:46.915
if even if it's, like, theoretically a insecure passphrase, it buys you time in this type of situation
490
00:34:49.234 --> 00:34:51.234
because an attacker has to
491
00:34:51.474 --> 00:34:54.515
then manually grind out every passphrase
492
00:34:54.515 --> 00:34:54.994
combination.
493
00:34:55.910 --> 00:34:56.550
Yeah.
494
00:35:05.990 --> 00:35:07.590
We'll get through this, freaks. I'm sorry.
495
00:35:16.535 --> 00:35:19.575
I think we should wrap and publish this. There's
496
00:35:19.735 --> 00:35:21.015
you know? There's
497
00:35:21.895 --> 00:35:33.620
If you if you if if you have friends or family that are interested in Bitcoin, call them. Make sure they're not on a cold card. Try and get the word out as much as possible. There's still a bunch of
498
00:35:33.940 --> 00:35:37.860
users out there that maybe don't even know and haven't been swept yet.
499
00:35:38.904 --> 00:35:42.905
And if you can save even just one more person from losing their money,
500
00:35:43.625 --> 00:35:45.625
then it's huge. It's worth it.
501
00:35:46.585 --> 00:35:49.785
If you have email lists or anything, email people.
502
00:35:50.265 --> 00:35:51.944
Text people. Call them. Call them.
503
00:35:52.710 --> 00:35:54.310
If you're away on travel
504
00:35:55.350 --> 00:36:00.870
if you're away on travel and you don't have access, call someone who's who's, you know,
505
00:36:01.270 --> 00:36:02.710
relatively trusted
506
00:36:03.270 --> 00:36:07.750
and have them go sweep it for you. If they and and and
507
00:36:08.765 --> 00:36:11.005
you already have an attacker trying to take it.
508
00:36:11.725 --> 00:36:18.365
Like, you that you should at least and, you know, just just send someone over. Walk them through it.
509
00:36:18.925 --> 00:36:20.525
There's a level of urgency here.
510
00:36:28.319 --> 00:36:28.640
Yeah.
511
00:36:41.905 --> 00:36:44.385
Complacency kills. We were complacent.
512
00:36:44.385 --> 00:36:45.745
I'm
513
00:36:45.825 --> 00:36:46.305
terribly,
514
00:36:47.580 --> 00:36:49.180
deeply sorry for that, Chris.
515
00:37:01.434 --> 00:37:02.875
I guess we'll wrap here.
516
00:37:03.674 --> 00:37:05.675
Yeah. Would just also say, like,
517
00:37:11.275 --> 00:37:13.674
I just love you all. Hug your families.
518
00:37:16.850 --> 00:37:17.490
Yeah.
519
00:37:21.970 --> 00:37:22.530
Alright.