1
00:00:00,000 --> 00:00:13,440
We are getting into part two of the cold card and also just general comments Q&A to try to answer as many questions for people wondering and details about what's been going on.

2
00:00:13,700 --> 00:00:15,640
So let's go ahead and jump in.

3
00:00:17,360 --> 00:00:24,880
So we're continuing our Q&A from the previous episode just because there were a lot of things that I still wanted to address.

4
00:00:25,540 --> 00:00:28,240
Also, just a shout out to our sponsor, to Bitbox.

5
00:00:28,760 --> 00:00:40,640
Don't forget, you can get a discount code right down in the show notes if you're looking for a good open source hardware wallet and one to replace your gold card if you're in that situation.

6
00:00:41,360 --> 00:00:44,960
But we are going to be talking quite a bit about all of that stuff.

7
00:00:44,960 --> 00:00:50,780
So we're going to go ahead and get into a few of the questions and comments underneath all the posts.

8
00:00:51,180 --> 00:00:54,800
So KFC, KF Chan says,

9
00:00:54,800 --> 00:01:01,100
why attribute to incompetence what obviously can be attributed to malice? So we're inverting our

10
00:01:01,100 --> 00:01:06,100
Hanlon's razor there. When you have consecutive wrong decisions from otherwise competent people,

11
00:01:06,100 --> 00:01:10,320
it's a pattern that shows malice. Imagine you want to do something really bad. Wouldn't you

12
00:01:10,320 --> 00:01:18,240
set it up as if it's just incompetence? Maybe that I mean, I'm not. I've been really back and

13
00:01:18,240 --> 00:01:26,020
forth about is this is it a conspiracy was it on purpose versus was it a mistake or incompetence

14
00:01:26,020 --> 00:01:36,100
um i am usually a hanlon's razor uh person for leaning that direction and i've seen a lot of

15
00:01:36,100 --> 00:01:43,040
things that could make that make sense and i in hindsight i've looked at a lot of what cold card

16
00:01:43,040 --> 00:01:52,060
has done and what CoinKite has built. And I see more and more a lot of the convoluted systems

17
00:01:52,060 --> 00:01:58,480
that the very idea of going Bitcoin only is supposed to avoid, you know, like avoiding all

18
00:01:58,480 --> 00:02:02,900
of the crypto stuff and all these extra things and super features, you know, trying to go down

19
00:02:02,900 --> 00:02:11,560
to something that's simple and reliable and focused on one job. And then I'd start digging

20
00:02:11,560 --> 00:02:18,720
into a lot of this you know what was going on in the firmware and a lot of people and developers

21
00:02:18,720 --> 00:02:24,720
have said comments about like why would you do it this way why is this being handed around through

22
00:02:24,720 --> 00:02:30,340
you know three or four different things when you just need to call the random number generator

23
00:02:30,340 --> 00:02:39,380
the trng and if it doesn't it just doesn't work and you know the firmware the three the two secure

24
00:02:39,380 --> 00:02:46,820
elements plus the firmware side of things. You know, it feels like a whole lot of security

25
00:02:46,820 --> 00:02:53,820
theater that actually causes a bigger attack surface or a bigger failure surface with the

26
00:02:53,820 --> 00:03:00,900
intent of catching edge cases. You know, it's a complex way to solve an edge case rather than a

27
00:03:00,900 --> 00:03:08,720
simple way to solve an edge case. And the way this was designed, it seems like it's very reasonable

28
00:03:08,720 --> 00:03:14,960
to say they simply missed it because of all the stuff around it.

29
00:03:15,360 --> 00:03:16,720
Maybe that's on purpose.

30
00:03:17,080 --> 00:03:18,400
Maybe that's on purpose.

31
00:03:19,360 --> 00:03:22,860
But it's a little bit deeper, and there's some other comments,

32
00:03:22,960 --> 00:03:26,620
I think, in this list that will maybe shed some light on that.

33
00:03:28,080 --> 00:03:30,100
But I'm still back and forth.

34
00:03:30,320 --> 00:03:31,560
We'll talk about it a little bit more.

35
00:03:31,680 --> 00:03:33,760
Let's go ahead and get on the next one here.

36
00:03:34,980 --> 00:03:37,960
So Rarson says,

37
00:03:37,960 --> 00:03:52,180
Yeah, I hate to be all tinfoil hat, but this is so egregious and so fundamental to the basic operation of the device that it defies belief that it could have gone unnoticed for so long, especially when there are tweets from years ago about retirement scams, etc.

38
00:03:52,500 --> 00:03:57,460
It also flies directly in the face of pretty much everything MBK boldly stated about his own devices.

39
00:03:57,880 --> 00:04:04,900
Almost like having two secure elements was actually security theater designed to distract from the fact the firmware wasn't actually using the proper random generation.

40
00:04:04,900 --> 00:04:20,060
It seems like a flaw purposefully put there, once the source was changed to only viewable conveniently enough, that was designed to be a retirement scam for someone that just so happened to be discovered early by some random person playing around with AI looking for flaws in signing device firmware.

41
00:04:20,060 --> 00:04:37,960
So this is one of the things that really shines light on the possibility because there is a lot to point to with offhanded comments like retirement scams and this sort of thing.

42
00:04:37,960 --> 00:04:40,800
And he says this flies directly in the face of pretty much everything

43
00:04:40,800 --> 00:04:43,500
NVK boldly stated about his own devices.

44
00:04:44,260 --> 00:04:47,460
Yeah, but NVK's attitude around

45
00:04:48,540 --> 00:04:50,580
addressing potential problems.

46
00:04:51,440 --> 00:05:09,153
You know I took it for a long time as if it was competence annoyed with people trying to waste his time And I see it now in hindsight as defensiveness as insecurity

47
00:05:10,093 --> 00:05:12,913
hiding behind the veil of competence.

48
00:05:13,653 --> 00:05:18,353
And I can very easily see how this would go unnoticed

49
00:05:18,353 --> 00:05:21,533
if every single time somebody pointed something out,

50
00:05:21,533 --> 00:05:29,873
NVK just ran them through the works on social and said it was FUD and said, what a ridiculous thing.

51
00:05:30,013 --> 00:05:34,613
You know, shut up. You're just yours isn't even any better. You're just trying to FUD cold card.

52
00:05:35,093 --> 00:05:42,433
If that was the attitude, I mean, do I think he has that attitude, a different attitude behind the scenes like to people on his team?

53
00:05:42,653 --> 00:05:47,913
I would probably bet not. I would probably bet that was the attitude that he carried everywhere about everything.

54
00:05:47,913 --> 00:05:54,533
and that's a really easy way for something really critical to go unnoticed for years

55
00:05:54,533 --> 00:06:03,053
and so i it is egregious but it's also what could be a context or consequence

56
00:06:03,053 --> 00:06:11,453
of the insecurity of the aggressive attacking that he has done about people trying to point out flaws

57
00:06:11,453 --> 00:06:18,293
or issues that just got people to stop caring or to stop thinking of it as something where

58
00:06:18,293 --> 00:06:27,453
there was a familiar or a welcoming response or attitude that made it worth even doing

59
00:06:27,453 --> 00:06:27,833
anything.

60
00:06:28,173 --> 00:06:32,533
You know, made it worth using the device, made it worth exploring the device or working

61
00:06:32,533 --> 00:06:33,213
on the firmware.

62
00:06:34,073 --> 00:06:40,113
And of course, this was also changed when it went from open source to only source available,

63
00:06:40,113 --> 00:06:46,173
which is another reason why you could very potentially attribute the simple defensive

64
00:06:46,173 --> 00:06:52,793
insecure attitude to this entire problem is they were trying to figure out how to change the code

65
00:06:52,793 --> 00:06:58,193
just enough to make it so it wasn't the open source code anymore and they could change the license

66
00:06:58,193 --> 00:07:03,713
which is a really stupid reason to want to change code you know it's a it means that you're

67
00:07:03,713 --> 00:07:09,753
specifically for no particular reason not because it's better not because you want want it to be

68
00:07:09,753 --> 00:07:15,333
more secure, but you're changing it purely to be able to change the license. And I can see that as

69
00:07:15,333 --> 00:07:23,793
a prime time for accidentally introducing bugs. But there's some other elements worth noting

70
00:07:23,793 --> 00:07:31,173
about the whole situation in the CTO, which I should probably do a really full on video about

71
00:07:31,173 --> 00:07:37,333
that to get into the details. But for right now, let's let's jump to the next one. Okay,

72
00:07:37,333 --> 00:07:40,973
So this one gets a little bit more into the weeds, but it's worth pointing out.

73
00:07:41,113 --> 00:07:45,973
So Sato 417, I'll add a little note on the fallback mechanism, quote unquote,

74
00:07:46,393 --> 00:07:48,493
device was not built with an intentional fallback.

75
00:07:49,473 --> 00:07:53,173
CoinKite wanted the wallet to use its own high security hardware random number generator, TRNG.

76
00:07:53,453 --> 00:07:56,753
To do this, they told the underlying software system, MicroPython,

77
00:07:56,833 --> 00:07:59,753
to turn off its built-in generator by setting a switch to zero.

78
00:08:00,593 --> 00:08:05,273
A new security library, LibNGU, checked this switch using a command,

79
00:08:05,273 --> 00:08:12,893
if in-depth that only asks does the switch exist it did not ask is it turned on because the switch

80
00:08:12,893 --> 00:08:19,253
existed even though it was set to zero or off the library assumed the hardware was ready it proceeded

81
00:08:19,253 --> 00:08:24,733
to call a standard function rng get since the main system thought the hardware was disabled because

82
00:08:24,733 --> 00:08:31,133
of the zero it automatically routed that function call to a weak software backup yasmerang instead

83
00:08:31,133 --> 00:08:32,873
of the secure hardware chip.

84
00:08:33,393 --> 00:08:39,053
Edit, if they had set the value to one instead of zero, the device would have accidentally

85
00:08:39,053 --> 00:08:45,013
used the CoinKite hardware RNG in an unintentional way, and likely everyone's funds would be

86
00:08:45,013 --> 00:08:45,493
safe.

87
00:08:45,913 --> 00:08:49,653
That's kind of funny to think if they had actually made a mistake, it would have been

88
00:08:49,653 --> 00:08:55,533
safe, but the intent, it wouldn't have been the intended way that they wanted to use it.

89
00:08:55,533 --> 00:09:01,533
And this kind of lends itself to the fact that there was just a mistake, especially when

90
00:09:01,533 --> 00:09:04,413
Yasmerang never even should have been in this.

91
00:09:04,773 --> 00:09:05,913
There were too many options.

92
00:09:06,293 --> 00:09:07,593
There was too much going on.

93
00:09:07,593 --> 00:09:10,713
It should have been one simple structured system.

94
00:09:11,233 --> 00:09:16,753
But Rarson responds, that still seems like a huge oversight as to be intentional or such

95
00:09:16,753 --> 00:09:18,033
a huge oversight as to be intentional.

96
00:09:18,393 --> 00:09:22,073
Who implements a new security library without fully understanding how it works and then

97
00:09:22,073 --> 00:09:24,373
never bothers to test it sufficiently for five years?

98
00:09:24,373 --> 00:09:29,213
someone who is up their own ass.

99
00:09:30,053 --> 00:09:34,633
Like, you know, I never put the belligerence and stubbornness of people.

100
00:09:35,753 --> 00:09:38,113
Again, this could have been intentional.

101
00:09:38,273 --> 00:09:39,693
This absolutely could have been intentional.

102
00:09:39,693 --> 00:09:41,653
And there's a lot to potentially indicate that.

103
00:09:42,733 --> 00:09:48,113
But the belligerence and confidence of people who have had something,

104
00:09:48,593 --> 00:09:51,753
the assumptions of people who have not had anything go wrong

105
00:09:51,753 --> 00:10:04,786
or have done really well and built something good in the past like the ability for people to be blind or deny or resist changing or acknowledging something is quite profound

106
00:10:05,246 --> 00:10:12,526
And I don't think this is out of the realm of normal, arrogant human behavior, I guess is the way to put it.

107
00:10:13,026 --> 00:10:23,026
But he says, I am a firm believer in Hanlon's razor, but this is immense levels of ineptitude, something that defies belief of a person who is capable of writing this kind of device firmware in the first place.

108
00:10:23,026 --> 00:10:25,866
It's the most basic and most important function of the device.

109
00:10:26,646 --> 00:10:29,406
I can't do anything but fully agree with that.

110
00:10:29,586 --> 00:10:36,306
Like that is, this is the most foundational part of what a hardware wallet does.

111
00:10:36,666 --> 00:10:49,046
And it's insane that it got this little attention and assessment with just this like hand wavy assumption that like, well, we figured that out.

112
00:10:49,246 --> 00:10:52,166
So let's move on and build all this crap on top of it.

113
00:10:52,166 --> 00:10:59,406
And that's. It's crazy that it got to as bad as it was.

114
00:10:59,746 --> 00:11:06,006
And in fact, I'll make a note here for those who didn't realize this or hadn't heard about this.

115
00:11:06,166 --> 00:11:11,086
So there were actually a number of different posts on Reddit over the years.

116
00:11:11,666 --> 00:11:18,626
This that showed or that suggested that someone had had their cold card white.

117
00:11:18,626 --> 00:11:26,466
and a bunch of people were suggesting that this bug had been found in the past and that,

118
00:11:26,466 --> 00:11:28,686
you know, somebody was wiping wallets.

119
00:11:28,806 --> 00:11:32,786
But that didn't really make sense to me because if somebody had found this bug, they wouldn't

120
00:11:32,786 --> 00:11:34,426
have just taken one random wallet.

121
00:11:34,906 --> 00:11:39,286
They would have gone on an attack and seen, you know, let me check against the blockchain

122
00:11:39,286 --> 00:11:40,926
and see if there's anything else out there.

123
00:11:40,986 --> 00:11:42,706
Like, why would they have picked out one random address?

124
00:11:42,826 --> 00:11:43,006
Right.

125
00:11:43,006 --> 00:11:52,646
and the entropy was so bad with this bug there's actually the very very real possibility of

126
00:11:52,646 --> 00:12:00,886
collisions where two different people created the exact same key and there are posts from years ago

127
00:12:00,886 --> 00:12:08,706
about people saying they created a key and there was money on it there was bitcoin on it so what

128
00:12:08,706 --> 00:12:15,766
I actually think was happening is the entropy was so bad in all of the cold cards because of this

129
00:12:15,766 --> 00:12:21,986
firmware bug that new customers, new people generating a cold card were opening up a wallet

130
00:12:21,986 --> 00:12:27,526
and seeing one Bitcoin inside of it. And they were like, what the hell do I do? Because these

131
00:12:27,526 --> 00:12:31,326
posts of people talking about like, this is surreal. This doesn't make any sense. How is

132
00:12:31,326 --> 00:12:36,006
this possible? And assuming that people just opened it up or that we're making up the story.

133
00:12:36,006 --> 00:12:49,626
But then when this gets to NDK or this gets talked about in cold card, they just come in and say, you're just fud, you're making shit up. This is ridiculous. If they didn't even bother checking into it, then of course this could go on for five years.

134
00:12:49,626 --> 00:12:59,506
and you are also in a situation where somebody something was clearly being exploited but was not

135
00:12:59,506 --> 00:13:04,246
actually being exploited in a known way that it could literally have been by accident and if

136
00:13:04,246 --> 00:13:08,426
somebody boots up a key and there's bitcoin on it what are you going to do you're probably going

137
00:13:08,426 --> 00:13:14,386
to move it to some other location most people are not honest enough to go do all the work and the

138
00:13:14,386 --> 00:13:19,646
pain of figuring out whose it was and figure out how to get it back to them. Probably a lot were

139
00:13:19,646 --> 00:13:28,666
just quietly, well, now I have $10,000 or whatever it was. And they just took it. And then there were

140
00:13:28,666 --> 00:13:36,566
some very unlucky and confused users who also got wiped out. And it was just because there's only

141
00:13:36,566 --> 00:13:43,466
like a hundred thousand, a million entropy options for what keys to produce. And people were just

142
00:13:43,466 --> 00:13:45,206
kind of accidentally producing the same ones.

143
00:13:46,026 --> 00:13:52,686
So I think that, if you want to say it,

144
00:13:52,706 --> 00:13:56,806
I think that could lend it again to incompetence,

145
00:13:57,426 --> 00:13:58,806
that it was just sitting there, nobody knew,

146
00:13:58,886 --> 00:14:00,106
and nobody bothered looking at it,

147
00:14:00,106 --> 00:14:05,746
and everything, every legitimate attempt at pointing out a bug

148
00:14:05,746 --> 00:14:10,506
was just dismissed because of insecurity or arrogance.

149
00:14:10,506 --> 00:14:16,786
okay there is no mobile app for sparrow if you find a sparrow mobile app it's a scan

150
00:14:16,786 --> 00:14:23,686
so i just heard about the fact that there were people who downloaded this sparrow app on mobile

151
00:14:23,686 --> 00:14:31,266
and uh man i wish i had even thought to say sparrow for desktop when i suggested people

152
00:14:31,266 --> 00:14:38,926
download sparrow because sparrow is the one that is the most reliable i think for just punching in

153
00:14:38,926 --> 00:14:43,746
a seed phrase or a key and then finding the wallets on it.

154
00:14:43,746 --> 00:14:47,646
In fact, my producer just told me this story that someone had six Bitcoin.

155
00:14:47,646 --> 00:14:52,366
They were trying to get it off their cold card and they downloaded Sparrow from the app

156
00:14:52,366 --> 00:15:04,739
store and sent their Bitcoin to it and then got rugged Jesus if nothing else there not much to say on this other than the fact there no mobile app for Sparrow

157
00:15:05,199 --> 00:15:07,879
And if you were ever thinking about this, remember that.

158
00:15:08,299 --> 00:15:11,279
Sparrow is a desktop only and the app is a scam.

159
00:15:11,999 --> 00:15:18,119
But it just lends itself again to so many,

160
00:15:18,439 --> 00:15:22,539
there's just so many attack vectors and so many things to consider.

161
00:15:22,759 --> 00:15:32,679
And this is why even though you need to be quick during situations like this, you also still need to take a breath and be careful.

162
00:15:32,799 --> 00:15:40,279
You need to think about it because I would assume there's probably a way, there's probably ratings or whatever.

163
00:15:40,379 --> 00:15:45,899
Like if you've been a little bit careful, it's probably easy to find out that that Sparrow mobile app is a scam.

164
00:15:45,899 --> 00:15:52,759
and when you are sending your bitcoin to something it's gone if it's not the right thing

165
00:15:52,759 --> 00:16:01,439
always do a double check on what what the thing is that you are using what the app is that you're

166
00:16:01,439 --> 00:16:09,239
using it with and that it comes from someplace reputable that it has a good rating and that's

167
00:16:09,239 --> 00:16:15,059
just the that is just misery though to think that you just saved yourself you got your funds out

168
00:16:15,059 --> 00:16:18,839
and then get rugged by the place you were trying to quickly send it to.

169
00:16:19,799 --> 00:16:23,559
Dear God, that just sucks.

170
00:16:25,239 --> 00:16:27,979
Again, you know, PSA, there's no mobile Sparrow app.

171
00:16:32,139 --> 00:16:37,299
Quote, unquote, anything is safer than a cold card is not entirely accurate,

172
00:16:37,659 --> 00:16:40,599
especially if you're one of defense cases.

173
00:16:40,759 --> 00:16:42,879
Malware hot wallets are having a field day.

174
00:16:42,879 --> 00:16:48,779
saw someone lose the six bitcoin that's the one on mobile wasabi wallet which of course was fake

175
00:16:48,779 --> 00:16:55,339
yeah there's not a mobile wasabi wallet either i don't believe man what a what a dumb attack vector

176
00:16:55,339 --> 00:17:03,399
but so easy to get people because people just think about it in like app store and so like

177
00:17:03,399 --> 00:17:07,339
say oh i use wasabi wallet all the time i use sparrow wallet all the time

178
00:17:07,339 --> 00:17:11,919
it's only on desktop it's only wasabi desktop it's only sparrow desktop

179
00:17:11,919 --> 00:17:21,939
uh thank you guys for psychros and derivation dingus and uh the other guy it

180
00:17:21,939 --> 00:17:30,979
knowledge knowledge hodl 775 for pointing that out to try to help spread that for anybody else

181
00:17:30,979 --> 00:17:36,199
because i didn't even hear about it until today um and it didn't even register that that was a

182
00:17:36,199 --> 00:17:43,099
thing like another thing to consider is the same thing was like of course of course they would just

183
00:17:43,099 --> 00:17:46,499
name it the same thing and of course the app store is going to take forever they'll take your

184
00:17:46,499 --> 00:17:54,019
uh the second you're trying to like sell something for 99 cents in your app for bitcoin or for

185
00:17:54,019 --> 00:17:59,759
something that doesn't get the 40 app store payment to apple they will take it down immediately

186
00:17:59,759 --> 00:18:05,339
without any hesitation and they'll probably have ai do it but i bet you this wasabi and sparrow

187
00:18:05,339 --> 00:18:09,479
wallet, even though they're a scam and stealing from people will be up there for months and

188
00:18:09,479 --> 00:18:17,019
months and months. Um, cause that's how, that's how it is a lot of times. Um, keep that in mind,

189
00:18:17,239 --> 00:18:22,719
keep that in mind. Your hot wallet is, is a vulnerability as much as anything else. Just

190
00:18:22,719 --> 00:18:27,099
be careful about what you download and what you're using. Um, make sure you have,

191
00:18:27,599 --> 00:18:32,199
you have a clear path to somebody that is visible and a reputation that is meaningful.

192
00:18:35,339 --> 00:18:38,559
What a thing to get hit with after you got yourself off a cold card.

193
00:18:39,239 --> 00:18:42,739
And I think we're going to close this one here because we're already getting kind of long.

194
00:18:43,199 --> 00:18:53,939
We'll actually do a part three and that will be more geared around the response, the red teaming, the best practices, the lessons learned and those sorts of things.

195
00:18:54,839 --> 00:18:57,199
Because there is a lot to pull from this.

196
00:18:57,499 --> 00:19:05,019
And I'll be doing another video more dedicated and a little bit deeper into the idea of whether or not this was malicious.

197
00:19:05,339 --> 00:19:08,839
maliciousness or incompetence.

198
00:19:09,399 --> 00:19:12,119
Because it's hard,

199
00:19:12,339 --> 00:19:13,979
it's really hard to say either way

200
00:19:13,979 --> 00:19:17,399
and I don't think either can be outright dismissed.

201
00:19:18,199 --> 00:19:19,599
So we'll see it.

202
00:19:20,659 --> 00:19:23,399
Details, links, our sponsor, Bitbox,

203
00:19:23,839 --> 00:19:27,339
open source hardware wallet is available

204
00:19:27,339 --> 00:19:28,219
in the show notes.

205
00:19:28,539 --> 00:19:30,499
So check it out and don't forget to subscribe

206
00:19:30,499 --> 00:19:32,879
to the Guys Take YouTube channel.

207
00:19:32,879 --> 00:19:37,259
we are still trying to bootstrap up and get attention to that one and that is where you

208
00:19:37,259 --> 00:19:43,279
will find this as well as a ton of other content stay subscribed i will catch you on the next

209
00:19:43,279 --> 00:19:46,519
episode and until then that is my two sats
