WEBVTT

1
00:00:00.031 --> 00:00:01.752
And it's not a memory corruption issue.

2
00:00:02.052 --> 00:00:03.934
It's not a race condition.

3
00:00:03.974 --> 00:00:06.576
It's not a business logic issue.

4
00:00:06.596 --> 00:00:07.516
I guess technically it kind of is.

5
00:00:07.556 --> 00:00:10.879
It's literally, we forgot that these two functions did a certain thing.

6
00:00:10.999 --> 00:00:14.782
And now when they do the correct thing, it means you logged in.

7
00:00:15.963 --> 00:00:16.503
That's the bug.

8
00:00:17.023 --> 00:00:18.004
Insane vulnerability.

9
00:00:18.304 --> 00:00:19.025
Insane thing.

10
00:00:23.334 --> 00:00:25.217
Good morning, hackers.

11
00:00:25.257 --> 00:00:30.685
Welcome back to the best podcast on the internet for hackers, The Lowdown, presented by Mays.

12
00:00:30.725 --> 00:00:32.368
Matt, I got a bone to pick with you, bro.

13
00:00:32.388 --> 00:00:33.649
Okay.

14
00:00:34.310 --> 00:00:35.532
You were late to the podcast.

15
00:00:36.273 --> 00:00:37.215
Did you not get my email?

16
00:00:39.799 --> 00:00:41.120
I didn't get any email.

17
00:00:41.621 --> 00:00:44.785
Is there a particular reason why you would have gotten no emails today?

18
00:00:45.145 --> 00:00:46.987
More like Microsoft 364, am I right?

19
00:00:47.007 --> 00:00:47.408
Am I right?

20
00:00:47.428 --> 00:00:48.048
Microsoft 503.

21
00:00:48.068 --> 00:00:49.290
More like Microslop.

22
00:00:49.310 --> 00:00:49.791
Oh, my God.

23
00:00:56.078 --> 00:00:56.618
Yeah, bro.

24
00:00:56.899 --> 00:01:09.047
So we're filming this episode on November 13th, which at this point, I think Microsoft Outlook has been down for over 24 hours, like an entire day of global downtime.

25
00:01:09.087 --> 00:01:13.751
Not one company, not one region, global Outlook outage.

26
00:01:14.151 --> 00:01:15.572
Now, a little strip tease there.

27
00:01:15.592 --> 00:01:19.195
If you expect us to talk about that, we're not going to because no one knows any details about it.

28
00:01:19.635 --> 00:01:21.897
But that's just kind of hilarious that it's happening right now.

29
00:01:22.960 --> 00:01:30.687
Dude, I was, again, like, I measure things in our industry by how many, like, normal people in my life start reaching out to me about stuff.

30
00:01:30.867 --> 00:01:34.930
I've gotten a bunch of text messages this morning from people like, what is going on?

31
00:01:34.970 --> 00:01:36.651
That's so funny, dude.

32
00:01:36.772 --> 00:01:41.395
From, like, random other, like, dads, like, friends of my kids, like, stuff, you know.

33
00:01:41.856 --> 00:01:43.377
I figured I'd go close to the source.

34
00:01:43.397 --> 00:01:45.258
You know, Matt's probably got to know what's going on here.

35
00:01:45.278 --> 00:01:46.319
And I was like, I actually don't.

36
00:01:46.359 --> 00:01:47.020
I'm in a bunch of, like, sports.

37
00:01:47.120 --> 00:01:51.282
speculative text messages groups with even people that like should be in the know.

38
00:01:51.302 --> 00:01:53.383
And I don't know if we're just like not hearing stuff.

39
00:01:53.923 --> 00:01:57.444
I'm guessing this doesn't smell like a cybery thing to me.

40
00:01:57.744 --> 00:01:58.064
I agree.

41
00:01:59.065 --> 00:01:59.465
Whoa.

42
00:01:59.765 --> 00:02:01.626
Who's that Pokemon?

43
00:02:01.686 --> 00:02:02.286
It's just...

44
00:02:08.049 --> 00:02:09.570
A wild John Hammond has appeared.

45
00:02:09.590 --> 00:02:10.131
What's up, John?

46
00:02:10.251 --> 00:02:10.691
How you doing, man?

47
00:02:10.731 --> 00:02:11.152
Oh, hey.

48
00:02:11.492 --> 00:02:11.852
Hey, everybody.

49
00:02:11.872 --> 00:02:12.312
How are you, brother?

50
00:02:12.332 --> 00:02:12.933
Goodness gracious.

51
00:02:12.953 --> 00:02:14.114
I'm so sorry for jumping in.

52
00:02:14.194 --> 00:02:15.014
It's all good.

53
00:02:15.415 --> 00:02:18.837
I know, I've heard you may be busy with a particular Enable thing going on.

54
00:02:18.857 --> 00:02:19.878
Is that correct, or is that something else?

55
00:02:19.898 --> 00:02:25.062
Yeah, a little bit of Enable, a little bit of Screen Connect, a little bit of, you know, just the internet doing what it does.

56
00:02:25.442 --> 00:02:30.887
Yeah, but before we get to that, we're talking about the Microsoft outage, which I'm sure you may be affected by minorly at least.

57
00:02:31.187 --> 00:02:34.791
How are you aware of what the hell is going on over there?

58
00:02:35.011 --> 00:02:37.493
It sure sounds like just a big outage.

59
00:02:37.594 --> 00:02:43.519
Yeah, I think there was a latest thing that said like, yeah, some CPU just keeps falling over or something.

60
00:02:43.779 --> 00:02:44.720
Yeah, good sign.

61
00:02:44.840 --> 00:02:45.341
Love that.

62
00:02:45.641 --> 00:02:47.242
Yeah, it smelled like DNS to me, right?

63
00:02:47.262 --> 00:02:48.223
It didn't smell like cyber.

64
00:02:48.243 --> 00:02:51.204
It smelled like something sillier, right?

65
00:02:52.025 --> 00:02:52.405
Cool, man.

66
00:02:52.425 --> 00:02:58.509
Well, thanks for joining The Lowdown, you know, the best podcast for hackers on the planet, as we like to say.

67
00:02:58.529 --> 00:03:01.731
This is quite an honor to be hanging out with you all.

68
00:03:01.751 --> 00:03:07.014
If I may say, I do have another call in like 25-ish minutes, but I would love to hang out if you're willing to have me.

69
00:03:07.134 --> 00:03:08.795
Go have the footage, edit, whatever you need.

70
00:03:09.095 --> 00:03:09.836
Yeah, we're just yapping.

71
00:03:09.856 --> 00:03:10.516
We're opening the show.

72
00:03:10.556 --> 00:03:11.217
We're yapping.

73
00:03:11.877 --> 00:03:20.144
Matt was talking about how he grades the severity of an incident by, like, how much he gets asked in real life from his other friends about, like, that are outside of the industry, what they say.

74
00:03:20.625 --> 00:03:26.509
I was going to say, I haven't been getting a lot about Microsoft, but I keep getting questions about, like, have you heard that the grid is going to go down?

75
00:03:27.050 --> 00:03:28.191
I'm like, what?

76
00:03:28.351 --> 00:03:32.954
And they're like, yeah, like, world leaders keep saying we've got to stockpile food and water.

77
00:03:33.014 --> 00:03:35.937
I'm like, I have not heard that, but I'll keep an ear out.

78
00:03:36.157 --> 00:03:37.038
Have you guys gotten any of that?

79
00:03:38.042 --> 00:03:38.282
Yes.

80
00:03:38.723 --> 00:03:38.903
Yes.

81
00:03:39.603 --> 00:03:49.152
I know of some people going like full-on survivalist mode in their spook and concern of, yeah, the world changing.

82
00:03:49.232 --> 00:03:49.712
Yeah, dude.

83
00:03:49.752 --> 00:03:50.173
It's weird.

84
00:03:50.193 --> 00:03:52.655
I mean, AI is good at security and hacking.

85
00:03:52.675 --> 00:03:57.759
And as we've kind of seen, the OT community most of the time, sorry, guys, is not.

86
00:03:57.999 --> 00:03:59.221
So we'll see how that plays out.

87
00:03:59.341 --> 00:03:59.661
Yeah.

88
00:03:59.881 --> 00:04:00.321
It's not great.

89
00:04:00.362 --> 00:04:00.462
Yeah.

90
00:04:01.042 --> 00:04:05.848
So if you guys don't know John Hammond for some reason, I don't know why you wouldn't know John Hammond.

91
00:04:05.888 --> 00:04:08.831
Between the three of us, we have millions of YouTube subscribers, by the way.

92
00:04:11.914 --> 00:04:12.495
Wait a second.

93
00:04:12.956 --> 00:04:18.542
Yeah, but John, you, I mean, Ed and I, this is our job now.

94
00:04:18.782 --> 00:04:19.383
You're looking at it.

95
00:04:19.803 --> 00:04:20.544
My job is beach.

96
00:04:20.784 --> 00:04:21.725
My job is coffee.

97
00:04:21.805 --> 00:04:23.307
My job is podcasts, right?

98
00:04:23.367 --> 00:04:25.309
Very Ken in the Barbie movie of me.

99
00:04:26.070 --> 00:04:27.652
You have like an actual job.

100
00:04:27.912 --> 00:04:35.781
And your actual job is talking a lot about like one of our main stories of today, which is this Screen Connect thing and Enable thing.

101
00:04:36.962 --> 00:04:39.845
I was reading a bit on the Huntress website about both of them.

102
00:04:40.666 --> 00:04:45.930
seems like a really bad day to be an RMM world, right?

103
00:04:46.030 --> 00:04:46.991
Or an MSP.

104
00:04:47.311 --> 00:04:52.175
Can you just give us the download of like what the heck is going on with Screen Connect and Enable?

105
00:04:53.416 --> 00:04:54.517
I can do my best.

106
00:04:54.717 --> 00:05:01.823
I feel like I personally have been tracking a lot of the Screen Connect and Enable more than I have Enable, but I can try to go to either one of them.

107
00:05:02.143 --> 00:05:07.647
But the gist is, yeah, it's been a pretty rough week for our managed service providers of the world.

108
00:05:09.568 --> 00:05:14.990
Let me do my best with Enable because I can, I don't know, maybe try to answer some of that and then we'll go through it.

109
00:05:15.070 --> 00:05:23.634
But an individual out on the internet had just dumped a whole lot of proof of concepts for a whole lot of vulnerabilities for Enable.

110
00:05:23.694 --> 00:05:28.156
And you've heard that sweet, like two shell suffix that we're just adding to everything now.

111
00:05:28.176 --> 00:05:28.756
We gotta stop.

112
00:05:28.816 --> 00:05:29.436
We gotta stop.

113
00:05:29.957 --> 00:05:30.157
Yeah.

114
00:05:31.677 --> 00:05:34.820
Uh, and they call it able to shell and or agent to shell.

115
00:05:34.840 --> 00:05:37.661
Now there's forever able to shell because he's added more.

116
00:05:38.242 --> 00:05:45.667
Um, but Enable is aware, obviously they took a heck of a lot of things offline so that that wouldn't be as susceptible as it is.

117
00:05:45.687 --> 00:05:46.688
Um, and then, uh,

118
00:05:47.688 --> 00:06:05.724
And that has been, I think, from disgruntlement of the researcher not hearing back from Enable, but I don't want to be like pointing fingers or casting shade in any of those directions because I think there is a more thorough process for an actual report to Enable, not just an email and hope someone responds in your inbox.

119
00:06:05.784 --> 00:06:06.164
I don't know.

120
00:06:06.705 --> 00:06:07.706
I wish I knew.

121
00:06:07.746 --> 00:06:07.846
But-

122
00:06:11.125 --> 00:06:13.006
What else are you all worried on the edible side?

123
00:06:13.026 --> 00:06:23.209
Well, I had to, first of all, you know, new noob style here, the I'm not going to do myself any favor in the YouTube comments that called me stupid over the last few weeks, but I had to figure out what the heck edible was, right?

124
00:06:23.309 --> 00:06:25.110
I wasn't familiar with that to begin with.

125
00:06:26.150 --> 00:06:31.112
I haven't run into that in a bit, but it seemed to me, I mean, the POC itself was kind of wild, right?

126
00:06:31.152 --> 00:06:33.593
It's like go to website calc exe pops.

127
00:06:33.813 --> 00:06:37.134
It's like, okay, that's, yeah, that's, that's about as POC as you get.

128
00:06:38.695 --> 00:06:39.495
We have another one, right?

129
00:06:39.555 --> 00:06:41.516
Fortinet did the same thing this week too.

130
00:06:42.017 --> 00:06:55.145
So like, and then I was even looking at our friends over at Grey Noise and they're starting to see a ton of activity actually scanning for this edible thing as well and active exploitation start to go down.

131
00:06:56.025 --> 00:07:07.889
Yeah, I mean, it just seems like a wild one that we all got to get our head around and like, you know, this enable or end central users just need to watch this, watch this active exploitation campaign going on.

132
00:07:08.669 --> 00:07:09.689
Forever.

133
00:07:09.889 --> 00:07:12.350
Oh, number four ever of both who shall.

134
00:07:12.410 --> 00:07:14.491
So I'm so over the to shell, man.

135
00:07:14.511 --> 00:07:17.252
It's like dirty something and something to shell.

136
00:07:17.272 --> 00:07:17.932
Like, please stop.

137
00:07:18.012 --> 00:07:20.173
Just let's get some new words out there, please.

138
00:07:20.193 --> 00:07:20.453
Yeah.

139
00:07:21.733 --> 00:07:25.896
We're seeing at least one successful intrusion according to gray noise that I'm seeing right now too.

140
00:07:25.976 --> 00:07:27.297
So yeah.

141
00:07:27.717 --> 00:07:28.798
Yeah, seems like a spicy one.

142
00:07:28.878 --> 00:07:30.659
But okay, you know more about Screen Connect though.

143
00:07:31.325 --> 00:07:40.330
Yeah, I've been chasing a little bit of the Screen Connect side of the house because that one is wild and has a lot of different kind of interesting things to chase.

144
00:07:41.591 --> 00:07:47.795
What we had shared previously over on the blog, if you are willing to take a look at the old Huntress write-up.

145
00:07:47.815 --> 00:07:48.355
Yeah, I've got it up.

146
00:07:48.375 --> 00:07:52.878
There's this thing called SC Kit, the, I'm assuming Screen Connect Kit.

147
00:07:53.378 --> 00:08:07.720
And that is a worm so that, okay, when a user or technician would connect into a screen connect, like usual tech support, the good intentions of remote monitoring to managements, it would propagate to other affected hosts.

148
00:08:08.141 --> 00:08:16.642
But the way that that one came in through the door, the patient zero of this intrusion is a little bit more spicy because it's server side.

149
00:08:16.662 --> 00:08:21.903
And there's a lot of nuance on the client side versus the server side for screen connecting because –

150
00:08:22.863 --> 00:08:34.335
This actually has some of the spooky little supply chain shenanigans to it where it's a plug-in or extension for Screen Connect server that has some weak vulnerability in itself.

151
00:08:34.415 --> 00:08:45.366
This is RMM+, that would just be an open door to anyone being able to point at the Screen Connect server that's running it and then have the capability to push downstream to clients and have that worm-a-gate.

152
00:08:45.386 --> 00:08:46.427
Yeah.

153
00:08:47.447 --> 00:08:54.954
This is off the tails of some of the previous Screen Connect worms that have been in the news and the headlines lately that we were chasing previously.

154
00:08:55.074 --> 00:08:58.317
And I think the gimmick to that was a CVE.

155
00:08:58.597 --> 00:09:03.081
So specific CVE number, CVE 2026-84869.

156
00:09:03.141 --> 00:09:05.383
That's burned into my brain for the past couple of days.

157
00:09:06.464 --> 00:09:09.246
That is on the client side.

158
00:09:09.427 --> 00:09:17.935
That is when the technician were to connect to another screen connect computer via RMM capability, it poisons them.

159
00:09:18.876 --> 00:09:19.596
The technician.

160
00:09:19.616 --> 00:09:20.838
So yeah, yeah.

161
00:09:21.018 --> 00:09:25.582
Their computer, the connecting guest would have code execution onto the technician's computer.

162
00:09:25.982 --> 00:09:30.507
So that's how I believe a lot of screen connect kit or SC kit has kind of propagated more.

163
00:09:31.047 --> 00:09:34.652
Um, and that's what I think a lot of the community is now trying to get the nuance around.

164
00:09:35.012 --> 00:09:35.213
Okay.

165
00:09:35.233 --> 00:09:42.882
The version number for the server isn't as much as a concern, but if they're patching and updating the clients that their technicians use, that has to be at this version, 26.6.5, which is now fixed.

166
00:09:42.902 --> 00:09:43.102
But yeah.

167
00:09:46.246 --> 00:09:57.178
Now there's a new strange thing that I'm going to have a blog out for today as quick as I can for other people screaming their head off on Reddit of weird cryptocurrency connections.

168
00:09:58.419 --> 00:09:59.240
With screen, having to do with this?

169
00:09:59.260 --> 00:10:00.642
Having to do with screen connection?

170
00:10:00.662 --> 00:10:01.462
Yeah, yeah, yeah.

171
00:10:01.482 --> 00:10:06.608
Here, I can drop a Reddit link for you and maybe I can give you at least some of the inside breaking news.

172
00:10:06.948 --> 00:10:07.389
Yeah.

173
00:10:07.409 --> 00:10:08.150
It's out the door soon.

174
00:10:08.670 --> 00:10:10.291
So let me get this straight.

175
00:10:10.431 --> 00:10:16.294
So Threadactor wants to Wormify via RMM software.

176
00:10:16.574 --> 00:10:28.281
They get a technician to connect to them, which then infects the technician, which then sets off this cool graphic that you guys put of 288 hosts in like minutes.

177
00:10:29.762 --> 00:10:32.363
Because that's what that technician then had access to.

178
00:10:32.383 --> 00:10:33.544
Do I got it right?

179
00:10:33.564 --> 00:10:33.744
Right.

180
00:10:34.304 --> 00:10:34.484
Yeah.

181
00:10:34.764 --> 00:10:35.125
Wow.

182
00:10:35.205 --> 00:10:35.505
Okay.

183
00:10:35.605 --> 00:10:36.366
That's a really, really cool one.

184
00:10:36.386 --> 00:10:43.331
Just so I understand as well, this originates from a vulnerability in a plugin that the technician has on their computer.

185
00:10:43.992 --> 00:10:45.133
The server has.

186
00:10:45.173 --> 00:10:45.353
Right.

187
00:10:45.373 --> 00:10:45.773
Okay.

188
00:10:45.793 --> 00:10:45.893
Okay.

189
00:10:45.913 --> 00:10:47.375
The screen connects server has it.

190
00:10:47.395 --> 00:10:57.643
The technician using the server to reach their other hosts would infect the first round of technicians and the technicians connecting to other computers continue to infect and infect and infect and infect.

191
00:10:57.843 --> 00:11:13.164
So is the threat, do we, and maybe I'm getting to a point where you're not allowed to say, but is it like the threat actor is getting the technician to connect to the computer or is the, is the computer infected by malware initially that then is, you know, doing this?

192
00:11:13.404 --> 00:11:13.845
You know what I mean?

193
00:11:13.865 --> 00:11:13.985
Right.

194
00:11:14.557 --> 00:11:18.079
I'm the former when you get to the client side.

195
00:11:18.619 --> 00:11:31.565
Um, and I think, I hope that's what some of the new stuff that I had pointed at to you will at least have a semblance of sense to it because now I think there is a running insight of, look, could we bait people?

196
00:11:31.865 --> 00:11:37.467
Could we lure and have some social engineering to have them click on the computer that will then infect them?

197
00:11:38.128 --> 00:11:42.870
Um, so that should be a fun story to tell as quick as I can.

198
00:11:43.767 --> 00:11:44.948
Yeah, it's wild, man.

199
00:11:45.229 --> 00:11:50.433
Yeah, you're well, thank you in the middle of firefighting phone calls, hopping on the lowdown.

200
00:11:50.674 --> 00:11:54.897
I know not on the highest of priorities, but we appreciate you.

201
00:11:54.917 --> 00:11:55.578
You guys are cool.

202
00:11:55.598 --> 00:11:56.619
I want to hang out with you all.

203
00:11:56.659 --> 00:12:00.463
And forgive me if I'm like too frazzled or frantic, if I'm not explaining things all that well.

204
00:12:00.523 --> 00:12:01.103
No, you're great, man.

205
00:12:03.837 --> 00:12:04.437
Dude, so...

206
00:12:04.457 --> 00:12:12.281
It's the end of the year, and a couple months of, like, it feels like back-to-back-to-back kind of insane vulnerabilities.

207
00:12:12.301 --> 00:12:14.603
So, yeah, I think we're all just, like, mentally spent, dude.

208
00:12:14.883 --> 00:12:22.346
We were, like, put the show notes together for today, and we only had, like, two CDSS 10-crit, like, volumes in major software.

209
00:12:22.366 --> 00:12:23.307
We're like, wow, slow week.

210
00:12:23.467 --> 00:12:26.208
Like, you know, this is kind of crazy.

211
00:12:26.268 --> 00:12:27.969
We might go short on this podcast today.

212
00:12:28.350 --> 00:12:32.572
Yeah, the other thing, man, I've been trying to figure out how to make some content about this.

213
00:12:33.566 --> 00:12:38.650
and we should all talk about it, is I've been talking to people in the incident response side of things.

214
00:12:38.991 --> 00:12:41.173
I stay in touch a lot with Threat Intel groups.

215
00:12:41.233 --> 00:12:43.475
I know you guys are in some of the same groups I am.

216
00:12:45.036 --> 00:12:49.760
And then talking to some incident responders, that that's what they do for a living.

217
00:12:50.340 --> 00:12:56.466
They are so far beyond maxed in terms of what they're able to do right now.

218
00:12:56.806 --> 00:12:59.228
And they are trying so desperately to hire people

219
00:12:59.388 --> 00:13:11.922
So if you're like struggling in the job market right now and you have any sort of chops to like get into incident response, they are hurting, hurting with the amount of work that they're getting called into, right?

220
00:13:11.982 --> 00:13:16.386
Because of all this stuff, we all get to chuckle about the headlines, but it turns out that like...

221
00:13:17.327 --> 00:13:21.688
When we see active exploitation and that's it, we say, oh, yeah, this is being actively exploited.

222
00:13:21.968 --> 00:13:30.851
Turns out that means there's a lot of teams out there that are literally doing these firefighting drills, calling in the lawyers, calling in the incident responders.

223
00:13:31.091 --> 00:13:37.133
And literally law firms, too, are completely maxed on, like, ability to even take on this kind of work.

224
00:13:37.173 --> 00:13:37.893
It's kind of insane.

225
00:13:40.313 --> 00:13:41.394
Very insane.

226
00:13:41.974 --> 00:13:44.835
Tough to be a cyber guy these couple days.

227
00:13:44.875 --> 00:13:45.775
Cyber person, sorry.

228
00:13:46.242 --> 00:13:48.743
Well, one of the other, I know, I know you got to hop off soon.

229
00:13:48.763 --> 00:13:57.445
One of the other stories that we were going to talk about, John, was the SSH bug, which feels like an eternity ago, but it actually came out since the last time we recorded the podcast.

230
00:13:57.665 --> 00:14:01.146
And we can get into the actual details after you hop off.

231
00:14:01.226 --> 00:14:08.368
But one of the things I would love to ask you is like, why some of these bugs like SSH one, the PHP one, like...

232
00:14:09.648 --> 00:14:18.113
I'm really interested in your take on like why we're not seeing the like absolute internet fire that like these things could have turned into.

233
00:14:18.433 --> 00:14:18.713
Right.

234
00:14:18.813 --> 00:14:25.216
It's like very widespread, questionably, easily exploited.

235
00:14:25.597 --> 00:14:28.938
Maybe that's part of the answer is like, you know, the exploitation is kind of hard.

236
00:14:29.299 --> 00:14:33.861
But why do you think like PHP and this SSH bug aren't like absolutely wildfires right now?

237
00:14:34.466 --> 00:14:34.606
Yeah.

238
00:14:34.686 --> 00:14:49.171
I'd love to run down the road on the SSH to shell, but I might need your hand holding because I think the gimmick of that was just, uh, 255, 256 or like longer password would just let you write into root on like Alpine Linux and things.

239
00:14:49.191 --> 00:14:49.611
Is that right?

240
00:14:49.952 --> 00:14:50.172
Yeah.

241
00:14:50.772 --> 00:14:51.012
Yeah.

242
00:14:52.515 --> 00:14:54.177
So many people have been poking fun.

243
00:14:54.317 --> 00:14:55.379
Oh, Terminal Tuchel.

244
00:14:55.699 --> 00:14:56.941
My buddy made a really good one.

245
00:14:57.001 --> 00:14:57.982
Pseudo Tuchel.

246
00:14:58.202 --> 00:14:59.184
Yeah, I saw that.

247
00:14:59.284 --> 00:15:00.125
It's so good, dude.

248
00:15:00.145 --> 00:15:01.547
It's so funny.

249
00:15:01.667 --> 00:15:10.618
Anyway, to your point, I probably would have a better answer on the PHP Tuchel side of it, not being as cognizant of what's been going on running around for SSH Tuchel.

250
00:15:11.499 --> 00:15:22.730
PHP to Shell specifically, which was one of those like, oh my God, the world is falling, the internet's on fire, everyone everywhere could be affected, vulnerable, exploited, why isn't there RCE, absolute Armageddon apocalypse.

251
00:15:24.271 --> 00:15:34.821
That one was like genuinely at least very, very nuanced to the environment, to the version, to the heap grooming, memory corruption, low level, low level shit.

252
00:15:35.321 --> 00:15:38.724
that I don't know if, Ed, you might have some sweet insight on and all.

253
00:15:39.344 --> 00:15:40.185
Sorry, we have an inside joke.

254
00:15:40.205 --> 00:15:42.006
Whenever someone says low level, I point to the sign.

255
00:15:42.046 --> 00:15:43.507
It's become like a tick in my videos.

256
00:15:43.527 --> 00:15:45.529
Whenever I say the word low level, I say name of the channel, by the way.

257
00:15:45.889 --> 00:15:46.369
Yeah, that's it.

258
00:15:46.389 --> 00:15:46.690
That's it.

259
00:15:47.570 --> 00:15:47.890
Yeah, man.

260
00:15:47.910 --> 00:15:56.977
So, yeah, my understanding of that is my understanding of the reason why it wasn't a super hot, like, actual threat in the wild is for that exact reason, right?

261
00:15:56.997 --> 00:16:04.583
The amount of recon you have to do that I think technically isn't really possible to know not only Apache or NGINX,

262
00:16:05.103 --> 00:16:06.684
What is the configuration of those two things?

263
00:16:06.705 --> 00:16:08.506
Are you reverse proxying to a different server?

264
00:16:08.566 --> 00:16:09.767
Are you on the same box?

265
00:16:10.087 --> 00:16:11.609
And then what version of PHP?

266
00:16:11.989 --> 00:16:16.573
And then from there, pulling off a race condition via a session establishment.

267
00:16:16.613 --> 00:16:19.556
Like, all those things have to line up perfectly to get the shell.

268
00:16:19.656 --> 00:16:26.281
So, like, I would imagine some nation-state, some highly-resourced threat actor could probably pull it off and probably has.

269
00:16:26.622 --> 00:16:27.502
But for the...

270
00:16:28.363 --> 00:16:32.604
Mass exploitation scenario, I think just statistically, it's very difficult to get correct.

271
00:16:32.644 --> 00:16:33.384
So that's probably why.

272
00:16:33.624 --> 00:16:36.765
But I think following on to that, though, the SSH one is not that.

273
00:16:36.805 --> 00:16:38.086
It's literally a logic bug.

274
00:16:38.106 --> 00:16:39.926
So I'm shocked we haven't seen more of those as well.

275
00:16:41.346 --> 00:16:44.827
Have you all seen the Gray Noise page for PHP to Shell?

276
00:16:45.408 --> 00:16:46.208
Actually, I haven't looked at that.

277
00:16:46.228 --> 00:16:46.428
No.

278
00:16:46.808 --> 00:16:46.948
Oh.

279
00:16:47.812 --> 00:16:51.594
I think I've been focusing on the census one a bit, on the patching.

280
00:16:51.614 --> 00:16:52.294
Yeah, census one is great too.

281
00:16:53.015 --> 00:16:54.616
Big love to GreyNoise though.

282
00:16:54.676 --> 00:17:05.522
They had ones that saw confirmed exploitation, and you can see a lot of the tradecraft, and a good many of those hosts are pulling down payloads from what was an open directory.

283
00:17:05.922 --> 00:17:11.985
So we had some fun shenanigans trying to see, oh, whoa, yeah, there is an exploit here.

284
00:17:12.005 --> 00:17:16.608
There are their victim data, and we got to do some sweet reporting and try to help clean this up.

285
00:17:17.128 --> 00:17:23.900
Um, but Gray noise, I know has been tracking as well as census and others, but I'm right there with you on the SSH to shell one.

286
00:17:23.940 --> 00:17:27.045
How come that hasn't been sprayed and prayed as much?

287
00:17:27.066 --> 00:17:27.326
Yeah.

288
00:17:27.366 --> 00:17:27.847
Kind of weird.

289
00:17:27.867 --> 00:17:28.067
Right.

290
00:17:28.468 --> 00:17:29.289
Um, the,

291
00:17:30.948 --> 00:17:32.910
The other was that open directory thing.

292
00:17:32.930 --> 00:17:38.534
There was a story that didn't make our show today wound up on the cutting room floor here.

293
00:17:38.594 --> 00:17:48.561
But there was a story where a threat actor directory was found open and they found like a jailbroken deep seek.

294
00:17:49.402 --> 00:17:49.662
in there.

295
00:17:49.682 --> 00:17:50.483
Did you read about this?

296
00:17:51.404 --> 00:17:53.827
I didn't, but that makes complete sense.

297
00:17:54.167 --> 00:18:03.257
Yeah, they found a jailbroken deep seek in there and then they found evidence of this person being in an unnamed European telecom company.

298
00:18:04.566 --> 00:18:09.627
who hadn't announced that they were compromised, but this directory was like how they found it.

299
00:18:10.187 --> 00:18:18.049
And so, but then like, as the, as the research dug deeper, and maybe now like we're actually talking about it, we can, we can pull some of it up.

300
00:18:18.189 --> 00:18:27.731
But as they dug deeper, all the vulnerabilities that were used in all the hacks were like old, crusty, you know, old enough to vote CVEs.

301
00:18:28.812 --> 00:18:29.292
You know what I mean?

302
00:18:29.312 --> 00:18:31.472
Like really old, boring CVEs that were actually used.

303
00:18:31.492 --> 00:18:33.893
So like the whole jailbroken deep sea thing was like, cool,

304
00:18:34.994 --> 00:18:39.920
didn't need, you know, the obliterated zero day machine to get into this European telecom.

305
00:18:40.040 --> 00:18:40.260
Right.

306
00:18:41.522 --> 00:18:43.764
Oh, this is a Malbear labs.

307
00:18:43.844 --> 00:18:44.225
This is Anna.

308
00:18:44.325 --> 00:18:44.946
This is her work.

309
00:18:45.026 --> 00:18:45.446
Excellent.

310
00:18:45.727 --> 00:18:46.007
Yes.

311
00:18:46.327 --> 00:18:51.313
That's the one that's, that's unrelated to the other open directory from the threat actor that you were just talking about.

312
00:18:53.154 --> 00:18:56.256
Let me ask you all, do you all run obliterated models?

313
00:18:56.896 --> 00:18:58.697
Do you have any in your back pocket?

314
00:18:59.217 --> 00:18:59.717
I do not.

315
00:18:59.737 --> 00:19:02.819
The only model that I've run, I actually haven't even ran a model locally.

316
00:19:02.879 --> 00:19:08.181
I use like Olama hosted GLM-53, but I have not used an obliterated one, no.

317
00:19:08.762 --> 00:19:09.602
I haven't had the need to.

318
00:19:09.782 --> 00:19:10.022
Yeah.

319
00:19:10.903 --> 00:19:15.465
I have gotten my QN 3.8 obliterated up and running, and it...

320
00:19:16.685 --> 00:19:17.986
It is obliterated, you know?

321
00:19:18.006 --> 00:19:18.346
It flaps.

322
00:19:18.567 --> 00:19:19.227
It flaps.

323
00:19:19.247 --> 00:19:22.650
You ask all the questions that you shouldn't ask, and it gives you the answers.

324
00:19:22.710 --> 00:19:23.430
What's your stack?

325
00:19:23.470 --> 00:19:25.812
Are you running it locally on like a GPU farm or what?

326
00:19:25.872 --> 00:19:32.657
Yeah, I have one beefy machine that can probably just fit like one or two maybe good running threads of 27 billion parameters.

327
00:19:32.777 --> 00:19:37.301
But I do have some friends in high places who have helped give me extreme compute power.

328
00:19:37.521 --> 00:19:38.782
There you go.

329
00:19:38.802 --> 00:19:40.003
So I'm running with the robots.

330
00:19:40.043 --> 00:19:41.244
I know I've told you all before.

331
00:19:41.304 --> 00:19:42.405
I'm a cyborg at this point.

332
00:19:42.765 --> 00:19:43.366
Oh, I'm with you, dude.

333
00:19:43.546 --> 00:19:50.538
I used to be, like, very anti-AI, and I think just seeing how good it is at VR, I'm like, oh, bro, if I can get one running in my house, I'm fucking going to.

334
00:19:50.558 --> 00:19:53.002
They are too good to ignore at this point, so I'm with you for sure.

335
00:19:54.512 --> 00:19:55.153
Yeah, I'm boring.

336
00:19:55.413 --> 00:19:57.774
I'm doing mostly coding stuff, less VR these days.

337
00:19:57.795 --> 00:20:00.977
So I haven't had like the need to dabble.

338
00:20:00.997 --> 00:20:05.240
I've been trying to do some more of the local or the open weight model stuff.

339
00:20:05.400 --> 00:20:08.322
I don't have the sparks on my desk or anything like that.

340
00:20:09.343 --> 00:20:15.827
I'm going to need to bully you to see if your friends in high places are also my friends in high places.

341
00:20:15.867 --> 00:20:17.068
We do run in similar circles.

342
00:20:18.029 --> 00:20:22.893
But yeah, no, I'm like a frontier model fanboy at this point.

343
00:20:23.153 --> 00:20:23.413
So...

344
00:20:23.863 --> 00:20:31.531
I mean, they're doing a great job too, but once we start to get to the super security security questions, they... Oh, no, totally.

345
00:20:31.551 --> 00:20:38.117
I mean, we were just talking off cam a bit about a project that Lola was working on.

346
00:20:38.617 --> 00:20:41.420
It's like you say the word cyber and you're in Opus 4.8 jail, right?

347
00:20:41.440 --> 00:20:41.660
Yeah.

348
00:20:42.301 --> 00:20:44.903
Yeah, I literally, like, I was writing a tool to, like, visualize firmware.

349
00:20:44.923 --> 00:20:45.884
I'm like, hey, here's the firmware.

350
00:20:45.924 --> 00:20:48.006
Show me, like, how the different processes talk.

351
00:20:48.046 --> 00:20:48.586
Is it network?

352
00:20:48.626 --> 00:20:49.067
Is it IPC?

353
00:20:49.107 --> 00:20:49.927
I just want to know.

354
00:20:50.228 --> 00:20:51.829
Draw me a diagram for security research.

355
00:20:51.849 --> 00:20:53.150
And it's like, ah, that's cyber.

356
00:20:53.170 --> 00:20:53.630
Can't do that.

357
00:20:53.650 --> 00:20:55.332
I'm like, okay, man.

358
00:20:55.352 --> 00:20:56.012
I said firmware.

359
00:20:56.032 --> 00:20:56.293
Sorry.

360
00:20:56.353 --> 00:20:56.693
My bad.

361
00:20:58.014 --> 00:20:58.695
So, whatever.

362
00:20:58.875 --> 00:20:59.415
The F word.

363
00:20:59.776 --> 00:21:00.276
Yeah, right.

364
00:21:00.316 --> 00:21:00.756
Exactly.

365
00:21:00.796 --> 00:21:00.997
Exactly.

366
00:21:01.037 --> 00:21:02.378
How dare you say firmware?

367
00:21:02.398 --> 00:21:02.518
Yeah.

368
00:21:03.792 --> 00:21:29.564
uh yeah dude well cool i appreciate you come giving us the screen connect download we'll give you uh we'll give you some of your firefighting day back and uh why don't we have you on for like yeah why don't we have you on like with with less of a short notice and like you come on and we could talk about some some cool topic that we're all making videos about some week or something absolutely good to see you all have fun john hammond always down for a three-way see you later buddy take care bye

369
00:21:32.138 --> 00:21:33.118
Yeah, let's go.

370
00:21:33.218 --> 00:21:33.698
Okay, cool.

371
00:21:33.979 --> 00:21:37.519
Hey, hackers, before we keep going, I want to take a quick break and say a quick thank you to our sponsors.

372
00:21:37.800 --> 00:21:41.321
Today's Lowdown episode is sponsored by our friends over at Maze.

373
00:21:41.801 --> 00:21:47.102
Maze has spent the last couple of years using AI to find and remediate vulnerabilities in the cloud.

374
00:21:47.202 --> 00:21:53.944
And now they've released Maze Code, AI agents that find and deeply investigate vulnerabilities in your code.

375
00:21:54.184 --> 00:21:55.025
Now we all know the deal.

376
00:21:55.385 --> 00:22:01.630
Software composition analysis tools and SaaS toolings are all historically really bad at just making more noise.

377
00:22:01.690 --> 00:22:08.535
They find issues that may not actually be real and just create a lot of problems for your engineers to deal with that don't solve real security vulnerabilities.

378
00:22:08.635 --> 00:22:09.656
I get it, we've been there.

379
00:22:10.296 --> 00:22:15.683
Maze code is a really cool tool because Maze code actually knows how your code works.

380
00:22:16.023 --> 00:22:21.109
Maze code investigates every finding with the context of how your code is actually deployed.

381
00:22:21.489 --> 00:22:29.098
Sure, you're using a version of lib.xml2 that's vulnerable, but the actual code path to hit that vulnerability is not exposed in your Docker image.

382
00:22:29.358 --> 00:22:32.880
MazeCode can find that and won't show you that that vulnerability is a big deal.

383
00:22:33.040 --> 00:22:37.602
Every vulnerability that MazeCode reports comes with evidence that it's real and a way to fix it.

384
00:22:37.942 --> 00:22:42.304
Check out Maze at go.thelowdownpod.com slash Maze.

385
00:22:42.424 --> 00:22:46.105
That's go.thelowdownpod.com slash Maze.

386
00:22:46.285 --> 00:22:46.886
Thanks again, Maze.

387
00:22:47.026 --> 00:22:47.846
Let's keep going.

388
00:22:48.386 --> 00:22:49.547
Okay, and we're back.

389
00:22:50.287 --> 00:22:52.548
Um, yeah, so we talked a lot about that SSH bug there.

390
00:22:52.568 --> 00:22:53.889
You want to actually go into what we're talking about?

391
00:22:53.909 --> 00:22:54.749
Please, please, please.

392
00:22:54.869 --> 00:22:55.209
Okay.

393
00:22:55.429 --> 00:22:58.151
Uh, again, uh, you know, I know I'm your cohost.

394
00:22:58.191 --> 00:23:00.992
I'm also, uh, just a fan of your channel.

395
00:23:01.032 --> 00:23:02.633
You did a really good video on this.

396
00:23:02.673 --> 00:23:10.876
You did, you did like, because I read all about this bug on Twitter and I was still kind of like, what, like, what, why?

397
00:23:10.896 --> 00:23:14.238
Like the joke I made was the fact that

398
00:23:14.738 --> 00:23:24.902
Three different times this year, you could have tried between one and 257 the letter A's as like, you know, to break a secure protocol.

399
00:23:25.643 --> 00:23:27.923
And it would have worked three times in the last couple months.

400
00:23:27.943 --> 00:23:29.244
It's just insane, right?

401
00:23:29.264 --> 00:23:29.424
Yeah.

402
00:23:29.724 --> 00:23:35.627
So, you know, we talked a lot about the HSM and the key generation ceremony for the last one.

403
00:23:35.647 --> 00:23:38.168
This has nothing to do with anything like that.

404
00:23:39.388 --> 00:23:43.750
What was the root cause under this SSH bug?

405
00:23:43.990 --> 00:23:55.757
Yeah, so when I was reading about this on Twitter, you know, you see the POC, the guy demonstrating, and he puts literally 257 A's into an SSH login field for root, and it works.

406
00:23:55.977 --> 00:24:00.560
And so your initial thought there is, oh my god, buffer overflow, is it 1991 all over again?

407
00:24:01.040 --> 00:24:03.441
That is actually not what happened here.

408
00:24:03.481 --> 00:24:09.525
And so this is a really, really interesting vulnerability that is a function of what I call contract drift,

409
00:24:10.005 --> 00:24:14.467
So you have two functions that you kind of assume one does one thing, one does the other.

410
00:24:14.487 --> 00:24:22.251
And as time goes on, you kind of forget how they work, and they go farther and farther apart until you get to a point where you forget that both of them depend on each other.

411
00:24:22.752 --> 00:24:27.034
You forget exactly how, and that forgetfulness creates a vulnerability.

412
00:24:27.114 --> 00:24:36.339
Okay, so I'm literally going to read from the comic that the author of the bug, the finder of the bug, made to talk about it in simpler terms.

413
00:24:36.439 --> 00:24:38.960
They have a whole write-up, and they summarize the write-up in this comic.

414
00:24:38.980 --> 00:24:39.100
Okay.

415
00:24:39.120 --> 00:24:39.260
Okay.

416
00:24:39.280 --> 00:24:40.240
This comic was great.

417
00:24:40.381 --> 00:24:44.142
Like, honestly, like for complex topics, this was super, super helpful.

418
00:24:44.422 --> 00:24:55.828
So basically back in 1980, right, when they're making the whole like password system for Linux, Bill Joy in the Etsy password schema says, hey, we need a character or a symbol

419
00:24:56.408 --> 00:24:58.870
to represent an account that does not have a password.

420
00:24:58.890 --> 00:25:04.055
Okay, in this case, it was the daemon or the daemon account that should not be able to be logged into.

421
00:25:04.095 --> 00:25:05.196
And so he says, okay, cool.

422
00:25:05.516 --> 00:25:10.181
The star, the asterisk, right, shift eight, is going to represent does not have a password.

423
00:25:10.361 --> 00:25:11.402
Okay, not a big deal.

424
00:25:12.023 --> 00:25:14.925
Fast forward 42 years, August 2012.

425
00:25:14.985 --> 00:25:16.186
So if you're not aware,

426
00:25:16.867 --> 00:25:20.131
When you're writing in C, a lot of the code that you write is not actually code you've written.

427
00:25:20.151 --> 00:25:23.195
You use functions like printf, use functions like read or whatever.

428
00:25:23.255 --> 00:25:26.700
Those functions live in the libc standard library.

429
00:25:26.780 --> 00:25:31.406
Now, there are multiple distributions or versions of this library.

430
00:25:31.446 --> 00:25:34.210
The common one is GNU libc, glibc.

431
00:25:34.510 --> 00:25:36.511
Another one is called MUSL.

432
00:25:36.671 --> 00:25:37.431
I say Moosl.

433
00:25:37.451 --> 00:25:38.972
People have been yelling at me that it's Muscle.

434
00:25:39.192 --> 00:25:40.652
Fuck you guys, it's Moosl.

435
00:25:41.032 --> 00:25:43.854
There are different variants, and this bug is only in the Moosl variant.

436
00:25:43.874 --> 00:25:52.817
But basically, there's a function called crypt that does the password hashing and creates the password hash string that you see in your now Etsy shadow files, okay?

437
00:25:53.377 --> 00:25:54.778
For a long time, up until 2012...

438
00:25:55.958 --> 00:25:59.281
That thing, if the crypt function failed, would return a null.

439
00:25:59.301 --> 00:26:00.902
It would turn literally just a null pointer.

440
00:26:01.282 --> 00:26:03.384
The problem was programmers are bad sometimes.

441
00:26:03.424 --> 00:26:04.585
Don't check the return values.

442
00:26:04.605 --> 00:26:07.407
And because crypt could fail, it would return null.

443
00:26:07.527 --> 00:26:09.409
You would dereference null would crash your program.

444
00:26:09.429 --> 00:26:13.332
So they said, okay, we need to have a better way of returning failure.

445
00:26:13.692 --> 00:26:17.655
And so instead of returning null as failure, we're going to return something special.

446
00:26:17.776 --> 00:26:20.578
We're going to turn something real cool, real neat that no one's ever heard before.

447
00:26:20.618 --> 00:26:21.639
It's also the asterisk.

448
00:26:22.219 --> 00:26:25.281
It's also the same character that Bill Joy said is no password.

449
00:26:25.901 --> 00:26:26.682
Not a big deal.

450
00:26:26.902 --> 00:26:28.022
Surely that won't bite us in the ass.

451
00:26:28.683 --> 00:26:30.404
Fast forward four more years, 2016.

452
00:26:31.124 --> 00:26:35.267
There is a timing attack that is disclosed in SSH.

453
00:26:35.307 --> 00:26:37.328
So let's say we're on a computer, right?

454
00:26:37.628 --> 00:26:39.369
There is an account for Bob.

455
00:26:39.509 --> 00:26:41.250
There is not an account for Matt.

456
00:26:41.370 --> 00:26:51.399
OK, if I want to enumerate the accounts on the box, what I can do is log into a known account or sorry, I can log into an account that does not exist like Matt.

457
00:26:51.679 --> 00:26:54.842
And I can time that it immediately comes back because there is no password.

458
00:26:54.862 --> 00:26:55.783
There's no hashing done.

459
00:26:55.803 --> 00:26:58.225
The CPU does not run for a long time.

460
00:26:58.585 --> 00:27:02.831
I can log in to Bob's account, or at least try to, and see how long that takes.

461
00:27:03.251 --> 00:27:07.016
I can use that timing to figure out what other accounts do and don't exist.

462
00:27:07.036 --> 00:27:13.785
So to fix this, they made it so for accounts that do not exist, you use the algorithm of some other random account, right?

463
00:27:13.805 --> 00:27:14.986
This will matter here in a second.

464
00:27:16.080 --> 00:27:21.106
Um, now remember how we said before, if the crypt function fails, it returns star.

465
00:27:21.647 --> 00:27:21.947
Okay.

466
00:27:22.488 --> 00:27:31.439
Well, what happens is if you try to log into an account that has no password and the crypt algorithm fails, it will return a star.

467
00:27:31.879 --> 00:27:33.540
And that star means no password.

468
00:27:33.620 --> 00:27:37.082
So when it compares to two hashes, that passes the login algorithm.

469
00:27:37.383 --> 00:27:46.368
And it turns out for SHA-512, if you give it more than 256 characters, it will fail out to prevent a DDoS.

470
00:27:46.388 --> 00:27:52.532
Because if I gave you 48 million characters to process, it would spin the CPU.

471
00:27:52.552 --> 00:27:53.573
It would occupy the CPU.

472
00:27:53.593 --> 00:27:55.534
So I said, okay, we got to have a max out.

473
00:27:55.574 --> 00:27:56.695
We can't have too many characters.

474
00:27:57.015 --> 00:27:59.917
So what you do is for a SHA-512 account,

475
00:28:01.258 --> 00:28:02.939
You give it 256 characters.

476
00:28:03.519 --> 00:28:06.640
That fails out to zero in the child function.

477
00:28:06.760 --> 00:28:12.223
The parent function returns a star, and then that star matches the star in the EdSite shoutout file.

478
00:28:13.373 --> 00:28:13.893
That's the bug.

479
00:28:14.394 --> 00:28:14.834
Insane.

480
00:28:15.455 --> 00:28:17.196
And it's not a memory corruption issue.

481
00:28:17.516 --> 00:28:19.378
It's not a race condition.

482
00:28:19.438 --> 00:28:23.201
It's not a, like, business logic issue, I guess, technically it kind of is.

483
00:28:23.221 --> 00:28:30.587
It's literally, we forgot that these two functions did a certain thing, and now when they do the correct thing, it means you logged in.

484
00:28:31.407 --> 00:28:55.288
insane vulnerability it's wild all right so ssh not that widely used right no definitely not so to be clear this is only the open ssh server distribution in muscle c now this is why alpine was kind of the talked about hotness because the alpine account is i think um no login by default no no what is it it's like

485
00:28:57.065 --> 00:29:05.149
Yep, okay, if you log in as Alpine via this way, you get in as Alpine, but Alpine is the Alpine pseudo-equivalent, so you're basically root, effectively.

486
00:29:05.930 --> 00:29:06.970
Alpine is vulnerable to this.

487
00:29:07.610 --> 00:29:09.131
OpenWrt was talked about a lot.

488
00:29:09.211 --> 00:29:13.433
OpenWrt, if you don't know, is the firmware replacement for routers, right?

489
00:29:13.454 --> 00:29:18.816
If you don't trust your ISP's router, for example, which you shouldn't, you can flash OpenWrt onto it.

490
00:29:18.836 --> 00:29:22.618
It's a kind of mnemonic for DDWRT.

491
00:29:22.638 --> 00:29:23.619
It's like the old version of it.

492
00:29:24.693 --> 00:29:29.558
OpenWord uses Moosel C, but by default doesn't use Moosel OpenSSH.

493
00:29:29.578 --> 00:29:32.642
It uses DropBear, which is not vulnerable to this.

494
00:29:33.303 --> 00:29:34.444
It's just, it's weird.

495
00:29:34.464 --> 00:29:40.851
It's another kind of, like, supply chain-y, log for J-E issue where, like, people now have all these embedded devices all over the place.

496
00:29:40.911 --> 00:29:42.493
And, like, they're using SSH, but it's, like...

497
00:29:42.953 --> 00:29:44.295
What SSH do we use?

498
00:29:44.355 --> 00:29:45.597
Do we use OpenSSH?

499
00:29:45.657 --> 00:29:47.299
Are we using Muscle OpenSSH?

500
00:29:47.339 --> 00:29:51.144
Do I now have to go and patch my router that I bought 20 years ago?

501
00:29:51.164 --> 00:30:00.056
Like, what is, obviously, you should be patching, but, like, it's a similar issue where you don't really know if you're vulnerable to this unless you have very intimate details of your devices, which a lot of people don't.

502
00:30:00.777 --> 00:30:01.017
Yeah.

503
00:30:01.357 --> 00:30:02.718
So yeah, super weird man.

504
00:30:02.778 --> 00:30:03.539
Crazy bug for sure.

505
00:30:03.599 --> 00:30:13.684
I mean, yeah, the thing that stood out to me is like what I was talking asking John about right which was like seems pretty easy to exploit seems probably pretty widespread.

506
00:30:14.104 --> 00:30:14.945
Yeah granted.

507
00:30:15.817 --> 00:30:24.266
Alpine, probably harder to hit generally on internet-facing things than other distros, right?

508
00:30:25.668 --> 00:30:30.473
You know, it's usually... Alpine's like a very popular Docker base image, right?

509
00:30:31.293 --> 00:30:33.035
It's like where that really starts to rack up.

510
00:30:33.256 --> 00:30:33.416
Yep.

511
00:30:33.956 --> 00:30:35.017
So I wonder if there's like...

512
00:30:35.618 --> 00:30:41.524
you know, just networking complexities of, like, containerized networks versus just, like, open SSH type stuff.

513
00:30:41.584 --> 00:30:41.904
Yeah.

514
00:30:42.064 --> 00:30:48.710
I'm almost curious if, like, the Alpine Docker image is not running SSH by default, if that's, like, a feature you have to enable.

515
00:30:48.991 --> 00:30:49.491
You know what I mean?

516
00:30:49.571 --> 00:30:55.036
Like, if you're running, like, a K8 cluster and, like, you use Alpine as, like, your base image, is SSH turned on by default?

517
00:30:55.116 --> 00:30:55.697
Maybe not.

518
00:30:56.117 --> 00:30:56.217
Yeah.

519
00:30:56.237 --> 00:30:56.398
Yeah.

520
00:30:56.918 --> 00:31:01.742
So the problem with, like, I think the reporting on this, too, in terms of, like, oh, we're seeing threat actors do it.

521
00:31:02.282 --> 00:31:03.763
It's a really clean bug.

522
00:31:04.084 --> 00:31:08.147
And there isn't a lot of, like, telemetry to show you that it's happening.

523
00:31:08.207 --> 00:31:10.849
It's literally a password exchange over SSH.

524
00:31:10.889 --> 00:31:11.509
It's encrypted.

525
00:31:11.969 --> 00:31:12.730
So that's the thing.

526
00:31:12.850 --> 00:31:13.611
So that's one thing.

527
00:31:14.111 --> 00:31:23.899
Another thing is, like, the logs that they said the vulnerability creates or the exploit creates is literally just accepted password for account from IP from port.

528
00:31:24.559 --> 00:31:26.521
That's not noise.

529
00:31:26.561 --> 00:31:27.963
That is noise.

530
00:31:27.983 --> 00:31:28.864
That's not a good indicator.

531
00:31:28.884 --> 00:31:31.687
That's just like what's going to happen if you do password auth.

532
00:31:33.349 --> 00:31:38.895
You would have to know that that account doesn't have a password, which I guess is an indicator.

533
00:31:38.935 --> 00:31:43.520
But you have to be very knowledgeable about how the exploit works to actually find out you're being attacked by it.

534
00:31:43.540 --> 00:31:44.521
Yeah.

535
00:31:45.082 --> 00:31:46.042
Yeah, it's weird, man.

536
00:31:46.082 --> 00:31:47.043
I don't know.

537
00:31:47.083 --> 00:31:50.545
I think what I'm saying here is that there probably is a ton of exploitation going on.

538
00:31:50.765 --> 00:31:55.047
It's just very difficult to detect from like the traditional ways that we do it right now.

539
00:31:55.067 --> 00:31:56.007
I don't like that at all.

540
00:31:56.027 --> 00:31:57.007
I don't like that at all.

541
00:31:57.268 --> 00:31:57.488
Yeah.

542
00:31:57.708 --> 00:31:57.908
All right.

543
00:31:57.928 --> 00:31:58.488
This one's weird.

544
00:31:58.508 --> 00:32:04.571
It feels like this is an old bug at this point because of the gap between our two recordings and how fast things move.

545
00:32:05.511 --> 00:32:09.053
But I almost like being able to talk about some of these things not like day of.

546
00:32:09.493 --> 00:32:15.795
Because you do get to have this conversation about exploitation and things like that and complexities of environments and whatnot.

547
00:32:15.875 --> 00:32:18.136
So super interesting one.

548
00:32:18.156 --> 00:32:20.116
All right.

549
00:32:20.337 --> 00:32:22.877
Going... Man, I can't go...

550
00:32:22.957 --> 00:32:24.158
I just got back from a conference.

551
00:32:24.178 --> 00:32:28.079
I was speaking at a conference yesterday in Texas that I drove out to.

552
00:32:28.739 --> 00:32:33.641
And even the parts of the conference that weren't AI related were AI related.

553
00:32:33.781 --> 00:32:34.721
You know what I mean?

554
00:32:34.881 --> 00:32:36.222
This is all anyone's talking about.

555
00:32:36.242 --> 00:32:37.322
It's the whole industry right now.

556
00:32:37.462 --> 00:32:38.464
Right, right, right.

557
00:32:38.564 --> 00:32:42.010
And so we're going to put a dollar in that jar right now.

558
00:32:42.190 --> 00:32:46.157
OpenAI is officially a cybersecurity vendor.

559
00:32:46.738 --> 00:32:50.124
There's no other way to look at it, based on a few announcements that they had recently.

560
00:32:51.013 --> 00:33:01.298
So what started, you know, getting onto our show notes, and I think you made a video on your channel about this this week, was that they are doing a lot of their own vulnerability research, right?

561
00:33:01.318 --> 00:33:02.819
So they're eating their own dog food.

562
00:33:02.839 --> 00:33:04.079
They have a cyber team.

563
00:33:04.540 --> 00:33:05.780
We've got friends on that team.

564
00:33:05.960 --> 00:33:07.521
Like, you know, they're doing awesome work.

565
00:33:07.541 --> 00:33:09.362
They're obviously sucking up really good talent.

566
00:33:10.403 --> 00:33:13.064
And that good talent turns out with...

567
00:33:13.845 --> 00:33:23.738
you know, unlimited tokens and probably unguard railed frontier models that none of us have access to that they don't really talk about publicly is my guess.

568
00:33:24.258 --> 00:33:24.499
Right?

569
00:33:24.879 --> 00:33:25.240
I don't know.

570
00:33:25.320 --> 00:33:27.022
It's like Matt's speculation time.

571
00:33:27.222 --> 00:33:27.603
Maybe.

572
00:33:27.923 --> 00:33:28.184
You know?

573
00:33:28.204 --> 00:33:28.364
Yeah.

574
00:33:29.304 --> 00:33:30.945
But I think it's probably a pretty good guess, right?

575
00:33:30.965 --> 00:33:43.551
That they probably have versions of their own models with less specifically cyber guardrails on them than that they're even releasing to their cyber products or anything like that, right?

576
00:33:43.571 --> 00:33:44.411
Yeah.

577
00:33:44.851 --> 00:33:49.913
But they came out with basically a thousand LPEs this week.

578
00:33:49.933 --> 00:33:52.014
Can you, like, A, you made a video about this.

579
00:33:52.034 --> 00:33:53.115
Can you, like, walk me through this?

580
00:33:53.135 --> 00:33:56.036
This is part of, like, their whole Patch the Planet initiative, right?

581
00:33:56.556 --> 00:34:01.998
that like, you know, we've been trying to talk about some of the people that run this on the podcast for a while.

582
00:34:02.018 --> 00:34:06.019
It turns out they're busy finding thousands of vulnerabilities, whatever.

583
00:34:06.319 --> 00:34:06.799
Employed.

584
00:34:07.039 --> 00:34:08.379
John Hammond showed up.

585
00:34:08.940 --> 00:34:09.080
No.

586
00:34:09.440 --> 00:34:10.460
So... Montana.

587
00:34:10.940 --> 00:34:12.000
Yeah, come on, guys.

588
00:34:13.041 --> 00:34:15.881
But what the hell is going on?

589
00:34:16.402 --> 00:34:24.984
Can you help me understand the scope of this number of LPEs in this singular initiative as well, if you have any understanding of that?

590
00:34:25.544 --> 00:34:26.744
When you say scope, define that, I guess.

591
00:34:26.764 --> 00:34:27.345
Compared to normal.

592
00:34:27.565 --> 00:34:32.066
Like, compared to, like, how many LPEs do we find in a year normally, right?

593
00:34:32.126 --> 00:34:32.826
I have an idea.

594
00:34:32.846 --> 00:34:33.926
I kind of want your vibe on it.

595
00:34:34.206 --> 00:34:34.846
Yeah, I guess.

596
00:34:34.886 --> 00:34:43.169
So to frame that question, the information that I don't have is how universal are these LPEs?

597
00:34:43.189 --> 00:34:44.429
Okay, yeah, that's a really good point.

598
00:34:44.829 --> 00:34:54.277
The main question that you're probably asking is, like, is this similar to, like, Dirty Frag, or is this similar to Fragnesia in that it is a universal LPE or Dirty Cow, right?

599
00:34:54.717 --> 00:34:55.638
That I don't know.

600
00:34:55.758 --> 00:35:07.187
But I want to highlight to the audience here, when I say 1,000 LPEs, this is not 1,000, you know, null pointer derefs in the kernel that can be taken advantage of the DELS system.

601
00:35:07.207 --> 00:35:08.909
I'm talking literally they had...

602
00:35:09.609 --> 00:35:11.089
I'm referencing a graph here from OpenAI.

603
00:35:11.109 --> 00:35:12.350
We can have our editor put it on the screen.

604
00:35:12.730 --> 00:35:18.512
But they had 2,300 candidate vulnerabilities, so suspicious blocks of code.

605
00:35:19.212 --> 00:35:30.195
Of those 2,300 candidates, 1,700 were actual bugs, like vulnerabilities in the kernel, of which 1,200 were LPE candidates.

606
00:35:30.315 --> 00:35:34.856
948 of those turned into working LPEs, meaning you are Ubuntu on a box.

607
00:35:34.976 --> 00:35:37.037
I'm saying Ubuntu, but some Linux account.

608
00:35:37.597 --> 00:35:39.678
You are Ubuntu on a box, you run a script, you are root.

609
00:35:40.579 --> 00:35:41.819
That is it.

610
00:35:41.919 --> 00:35:47.523
And then 291 of them are not LPE pocs per se, but they're leak pocs.

611
00:35:47.663 --> 00:35:55.187
So they're, you know, leak privileged memory out of non-privileged memory or leak privileged memory out of non-privileged user mode execution.

612
00:35:56.628 --> 00:35:58.769
This is truly unprecedented.

613
00:35:59.429 --> 00:36:02.011
A, in the fact that it's volume, right?

614
00:36:02.051 --> 00:36:05.833
Just of the number of LPEs that have fallen out of the sky at once,

615
00:36:06.533 --> 00:36:09.155
And two, the fact that it was automated.

616
00:36:09.175 --> 00:36:17.922
Okay, the fact that we have now, we're not running SAS tooling, we're not running fuzzers, we are using LLMs to drive the ship and find these vulnerabilities.

617
00:36:18.323 --> 00:36:21.665
And I want to ask you a question, Matt, do you think $1.2 million is a lot of money?

618
00:36:22.846 --> 00:36:24.107
For this many vulns?

619
00:36:24.187 --> 00:36:24.988
No, just in general.

620
00:36:25.008 --> 00:36:25.068
No.

621
00:36:25.923 --> 00:36:27.944
Oh, in general, I would take it.

622
00:36:28.184 --> 00:36:30.264
What I'm saying is that is a lot of money.

623
00:36:30.324 --> 00:36:31.224
That is a fuck ton of money.

624
00:36:31.244 --> 00:36:31.885
I want to highlight that.

625
00:36:31.905 --> 00:36:33.965
People are going to say, oh, it costs them 1.2 mil.

626
00:36:34.325 --> 00:36:35.025
Totally agree.

627
00:36:35.165 --> 00:36:35.786
That's a lot of money.

628
00:36:35.926 --> 00:36:38.206
I don't have that to spend on fucking LPE finding for fun.

629
00:36:38.366 --> 00:36:38.786
No, thank you.

630
00:36:39.687 --> 00:36:41.627
I have a huge question here.

631
00:36:41.667 --> 00:36:47.069
If you divide that by a thousand, a thousand dollars per LPE is not a lot of money.

632
00:36:47.189 --> 00:36:47.829
No, totally.

633
00:36:48.229 --> 00:36:49.549
I also have a major question here.

634
00:36:49.589 --> 00:36:54.251
When we're reading about Frontier Labs talking about token costs.

635
00:36:54.411 --> 00:36:55.131
Sure, great.

636
00:36:55.731 --> 00:37:02.117
Like in my mind, is that an actual incurred cost for them?

637
00:37:02.397 --> 00:37:04.259
Because they would just be paying themselves, right?

638
00:37:04.279 --> 00:37:07.622
That's like their like advertised cost for the tokens used.

639
00:37:08.162 --> 00:37:11.465
So for them, I don't really know the answer to this, right?

640
00:37:11.505 --> 00:37:15.769
Because I don't have the inside baseball, but is that like an opportunity cost?

641
00:37:15.929 --> 00:37:18.752
Like it, because they were taking up that like,

642
00:37:19.729 --> 00:37:21.770
you know, compute basically, right.

643
00:37:21.850 --> 00:37:23.371
They were taking up that token usage.

644
00:37:23.851 --> 00:37:27.293
They couldn't go be charging for that at that exact moment in time.

645
00:37:27.313 --> 00:37:30.035
It's not like there's a queue going on here, but you know what I mean?

646
00:37:30.055 --> 00:37:30.655
Like, I understand.

647
00:37:30.675 --> 00:37:35.638
Was there some sort of opportunity cost that, that they use their own thing that they couldn't be selling?

648
00:37:35.918 --> 00:37:36.139
Right.

649
00:37:36.439 --> 00:37:36.599
Then.

650
00:37:36.659 --> 00:37:36.939
Okay.

651
00:37:36.959 --> 00:37:37.699
That's a real number.

652
00:37:37.980 --> 00:37:38.200
Right.

653
00:37:38.980 --> 00:37:41.602
Otherwise, I think we're talking electricity.

654
00:37:42.842 --> 00:37:47.306
As far as I understand it, and again, this is me guessing off of their verbiage.

655
00:37:47.366 --> 00:37:49.387
I am purely tea leaf readings.

656
00:37:49.447 --> 00:37:50.088
I don't know.

657
00:37:50.128 --> 00:37:50.368
Yeah.

658
00:37:50.588 --> 00:37:55.812
But the graph that I'm citing, they specifically said total API credits.

659
00:37:55.832 --> 00:37:57.172
But that's what they charge.

660
00:37:57.233 --> 00:37:58.353
That's what I'm saying.

661
00:37:58.433 --> 00:38:07.800
I think what they're insinuating there is this is the cost of the public facing API if you were to use our credits to do the same thing tomorrow.

662
00:38:08.240 --> 00:38:09.702
So did it cost them anything?

663
00:38:10.043 --> 00:38:10.363
No.

664
00:38:10.684 --> 00:38:11.565
Or electricity.

665
00:38:12.626 --> 00:38:14.188
But this is what it would cost.

666
00:38:14.429 --> 00:38:16.111
It would have cost somebody else.

667
00:38:16.131 --> 00:38:16.471
Totally.

668
00:38:16.832 --> 00:38:23.841
But I just want that to be crystal clear for people because some people are like well, OpenAI paid $1.2 million to better the planet.

669
00:38:23.881 --> 00:38:24.122
I'm like

670
00:38:24.868 --> 00:38:25.068
Right.

671
00:38:25.128 --> 00:38:25.748
I don't think they did.

672
00:38:26.108 --> 00:38:32.650
No, you know, they had like a, you know, a rounding error of electricity costs to like run this initiative.

673
00:38:32.910 --> 00:38:36.511
But that's what it would have caused you or me to do it on the API, right?

674
00:38:36.891 --> 00:38:41.352
And this issue that I have with like, this is a problem that's always been a thing in volant research in general, right?

675
00:38:41.732 --> 00:38:45.053
When I say the word resources here in a second, I'm going to say the word resources.

676
00:38:45.353 --> 00:38:46.933
That is resources of time.

677
00:38:47.273 --> 00:38:49.314
That is resources of money, electricity.

678
00:38:49.714 --> 00:38:53.115
When you're doing volant research, the inherent problem is that you know there are bugs.

679
00:38:53.535 --> 00:38:54.697
You just don't know where they are.

680
00:38:54.937 --> 00:38:57.821
So the question is, where do you spend your resources?

681
00:38:57.901 --> 00:39:00.305
Where do you spend your time, your electricity, your tokens?

682
00:39:00.926 --> 00:39:08.396
If you work at the token factory and get the tokens for free, and I know that that's not maybe what's happening, but like free generally.

683
00:39:10.640 --> 00:39:19.682
You know, it's a lot easier to justify let's ingest the entire Linux kernel source code, which is 40 million lines of code to find 1100 LPEs, right?

684
00:39:20.002 --> 00:39:21.263
So great work.

685
00:39:21.423 --> 00:39:22.763
Like, obviously, it's good shit.

686
00:39:22.783 --> 00:39:26.924
But I agree with you that it's not like they said let's spend $1.2 million.

687
00:39:26.984 --> 00:39:30.565
They said let's burn some electricity on this and say we spent $1.2 million.

688
00:39:30.826 --> 00:39:34.887
I also wonder, this is me postulating, because if I was in their position, right?

689
00:39:34.947 --> 00:39:39.348
If I had a shovel at the token factory and I was standing in front of the vuln furnace, right?

690
00:39:39.368 --> 00:39:40.788
And I was like, let's go, right?

691
00:39:42.608 --> 00:39:43.908
So I've tried some vuln research.

692
00:39:43.928 --> 00:39:53.650
I know we were just talking about obliterated models and I haven't gone that far, but I have done some vuln research with models to be able to speak intelligently about it enough, right?

693
00:39:53.670 --> 00:39:55.930
And just see what they were capable of and whatnot.

694
00:39:56.811 --> 00:39:58.191
I was doing some stuff with browsers

695
00:40:00.330 --> 00:40:02.011
I accidentally filled my hard drive.

696
00:40:02.031 --> 00:40:09.994
Yeah, it turns out when you're fuzzing browsers, you can make a whole lot of noise, right?

697
00:40:10.014 --> 00:40:10.775
The corpus gets very big.

698
00:40:10.915 --> 00:40:12.576
Yes, the corpus gets very big, right?

699
00:40:13.156 --> 00:40:26.322
So I wonder, right, like you and me kicking off like, hey, it's time to find some vulnerabilities in the Linux kernel would take up a ton of context in tokens just firing that like project off the ground, right?

700
00:40:26.342 --> 00:40:27.142
Yeah.

701
00:40:27.302 --> 00:40:41.638
I wonder if because they've got this long-standing Pat's the Planet initiative, right, and whatever, I wonder if they have to, like, do that instantiation every time or if they have some sort of harness, for lack of a better word, right, that's like...

702
00:40:42.839 --> 00:40:44.660
No, we paid that once, right?

703
00:40:44.740 --> 00:40:46.181
Like, we have it.

704
00:40:46.221 --> 00:40:51.143
We have the Linux kernel in something that resembles context to us, right?

705
00:40:51.784 --> 00:40:52.704
Because I don't know.

706
00:40:52.764 --> 00:40:56.406
I don't know how differently they're interacting with the things at the token factory.

707
00:40:56.466 --> 00:40:56.926
You know what I mean?

708
00:40:56.966 --> 00:40:57.586
Yeah, yeah, yeah.

709
00:40:57.606 --> 00:41:04.089
So I will highlight that in the article, they do a very good job on being a little transparent on what they're using.

710
00:41:04.109 --> 00:41:07.511
So they do have a special harness that they wrote that does a pretty cool thing, right?

711
00:41:07.611 --> 00:41:08.912
So as I highlighted...

712
00:41:09.632 --> 00:41:11.073
Linux kernel is 40 million lines of code.

713
00:41:11.293 --> 00:41:12.874
Very hard to reason about with an LLM.

714
00:41:13.855 --> 00:41:20.180
They had this system set up where they use these things that they call the survey agents where, hey, you're not going to go find me vulnerabilities.

715
00:41:20.200 --> 00:41:22.101
You're not going to go into deep thought on this.

716
00:41:22.481 --> 00:41:23.722
You're going to go look at the code.

717
00:41:23.742 --> 00:41:28.146
We're going to break it up in the subsystems and give a subsystem to every agent or group of them.

718
00:41:28.486 --> 00:41:30.808
And you're just going to find me suspicious blocks.

719
00:41:31.708 --> 00:41:52.457
uh structures that are hard to get right or code patterns that are dangerous and just say hey this is a suspicious thing and that is what the candidate findings are i believe then from there you give that suspicious candidate to a research agent i think they call that an analyst agent to do the actual you know go deep on this put it in a vm write a pock crash the kernel like write an exploit type thing

720
00:41:53.065 --> 00:41:56.248
So there is some harnessing going on that I think is accessible to the average person.

721
00:41:56.268 --> 00:42:06.037
But I think to your point, is there like a middle layer, like before the inference but after the LLM harness itself that says like, hey, we have completely embedded.

722
00:42:06.077 --> 00:42:13.463
We've contextualized and compressed the whole Linux kernel, and you're going to infer over that, not the plain text.

723
00:42:13.503 --> 00:42:14.304
That I don't know.

724
00:42:14.404 --> 00:42:15.265
Yeah, I don't know either, right?

725
00:42:16.405 --> 00:42:20.206
And maybe not because they are being so transparent with all of this, right?

726
00:42:20.306 --> 00:42:28.207
But so, I mean, it sounds a whole lot to me like when Mythos came out and we were starting to read about Project Glasswing, right?

727
00:42:28.467 --> 00:42:42.530
And I know we talked a lot about that at the time, that like Mozilla and Cloudflare and some of these early Glasswing participants who were able to be more transparent than maybe some other like large corporations that don't have as friendly of a PR department.

728
00:42:43.871 --> 00:42:48.154
And this is what like Mozilla, this is exactly what Mozilla was doing, right?

729
00:42:48.234 --> 00:42:59.462
To like, you know, they were literally file by file, like giving Mythos a file saying there is blank vulnerability in this file, gaslighting it.

730
00:42:59.542 --> 00:43:00.003
Gaslighting it.

731
00:43:00.023 --> 00:43:00.543
Yeah, you have to.

732
00:43:02.144 --> 00:43:05.247
And just saying there is this vulnerability in this file.

733
00:43:06.107 --> 00:43:07.148
find it, right?

734
00:43:07.629 --> 00:43:13.796
And then next file, next file, next file, next vulnerability class, next vulnerability class, next vulnerability class.

735
00:43:13.876 --> 00:43:18.341
So like you can imagine that like workload getting quite large, right?

736
00:43:18.822 --> 00:43:20.183
If you're doing a file on file.

737
00:43:20.203 --> 00:43:21.004
So it sounds very similar.

738
00:43:21.024 --> 00:43:21.165
Yeah.

739
00:43:21.705 --> 00:43:31.508
I was writing a harness to try to reproduce that process, and I did it to try to find one of the most recent 7-zip XZ cursor out of alignment issues.

740
00:43:32.148 --> 00:43:35.849
I lit $300 in token costs on fire so fast.

741
00:43:36.229 --> 00:43:47.352
Like, one run of the tool where I took every bug class, wrote a prompt for it, and then pointed every bug class at every file in 7-zip, which is not a lot of files, by the way, was like $100 per run.

742
00:43:47.372 --> 00:43:48.532
And I was like, all right, that's enough.

743
00:43:48.652 --> 00:43:49.252
That's enough of that.

744
00:43:49.332 --> 00:43:49.993
I'm not Mozilla.

745
00:43:50.073 --> 00:43:50.773
I can't afford this.

746
00:43:51.753 --> 00:43:53.114
So yeah.

747
00:43:53.154 --> 00:43:54.735
We're running on that podcast money.

748
00:43:54.916 --> 00:43:55.256
True.

749
00:43:55.276 --> 00:43:57.858
Which by the way, next year is coming up guys.

750
00:44:00.820 --> 00:44:01.540
Like and subscribe.

751
00:44:01.620 --> 00:44:02.341
Like and subscribe.

752
00:44:03.202 --> 00:44:05.463
The other dovetail off of this, right?

753
00:44:05.503 --> 00:44:08.746
That like, okay, OpenAI obviously doing cyber stuff.

754
00:44:08.846 --> 00:44:10.667
So like the Pats of Planet initiative got it.

755
00:44:10.727 --> 00:44:16.591
They're like trying to do all this stuff in Linux and open source, you know, project, major open source projects, stuff like this.

756
00:44:16.932 --> 00:44:18.553
But they did announce this week

757
00:44:19.033 --> 00:44:30.026
that they are just straight up competing down the line with uh cyber security vendors right they they launched daybreak basically red and blue right yeah uh daybreak defense daybreak uh red team

758
00:44:31.169 --> 00:44:34.813
And you can now hit them up and be like, let's do this.

759
00:44:34.973 --> 00:44:39.998
Like, I would like to hire you to be my offensive security agent.

760
00:44:40.098 --> 00:44:40.298
Right.

761
00:44:40.619 --> 00:44:41.259
Yeah.

762
00:44:41.560 --> 00:44:44.302
I find this super interesting because I think we all saw this coming.

763
00:44:45.483 --> 00:44:54.633
Like, there was just no way that they weren't going to turn on the money stream of cyber stuff when like you go to the black hat vendor floor and it's like, you know, a gazillion dollars.

764
00:44:55.533 --> 00:45:00.877
being thrown around saying that they do this kind of stuff and they're all just front-ending the frontier model capabilities anyway.

765
00:45:01.477 --> 00:45:03.178
So it makes sense that they were going to do this.

766
00:45:04.399 --> 00:45:07.081
This is the number one question I ask.

767
00:45:07.161 --> 00:45:10.463
We've done a few interviews now with AI security vendors.

768
00:45:10.703 --> 00:45:11.884
This is my number one question now.

769
00:45:12.264 --> 00:45:19.709
It's like, how do you continue to do what you do when the frontier models are going to start doing more and more of this out of the box?

770
00:45:20.030 --> 00:45:22.591
Not even this cyber offering that I'm talking about, right?

771
00:45:22.671 --> 00:45:22.952
Like...

772
00:45:24.032 --> 00:45:28.974
you know, Opus can do some of this stuff for 20 bucks, right?

773
00:45:29.054 --> 00:45:34.476
Like that vendors were doing as part of like their enterprise deals just a few years ago, right?

774
00:45:34.496 --> 00:45:38.817
So it's like, how do you kind of stay ahead of this as a cybersecurity industry, right?

775
00:45:38.837 --> 00:45:38.997
Yeah.

776
00:45:39.217 --> 00:45:39.757
It's difficult.

777
00:45:39.957 --> 00:45:46.240
Geohot, I think, said this in an interview talking about, this is like back in like the Sonnet 3, 7 days.

778
00:45:46.680 --> 00:45:53.663
But he was like, yeah, there's a problem where you have, he calls every company a tier X as a function of like what thing they provide to the market.

779
00:45:53.684 --> 00:45:56.225
So like tier zero is like the data center, right?

780
00:45:56.245 --> 00:46:01.387
Like they are the hardware, they are fundamental, and they are very hard to replace because data centers are expensive.

781
00:46:01.407 --> 00:46:03.648
And then you have like the tier ones that are like, ugh.

782
00:46:03.868 --> 00:46:04.928
open AI or Anthropic.

783
00:46:04.948 --> 00:46:10.069
They provide the frontier models and you kind of have like the tier twos and threes that all kind of just wrap each other.

784
00:46:10.709 --> 00:46:18.251
And his argument was that like the tier ones are going to just consume all of the other companies because they own the thing.

785
00:46:18.271 --> 00:46:18.811
You know what I mean?

786
00:46:18.831 --> 00:46:26.733
Like the data centers aren't going to become frontier labs, but the frontier labs can kind of become whatever company they want because they have the models that can do pretty much anything

787
00:46:27.333 --> 00:46:37.801
So it does beg this kind of scary question about like, how far do these companies go if they can just have a model that runs a company that, you know, does it better than the average competitor?

788
00:46:37.821 --> 00:46:38.201
It's weird.

789
00:46:38.381 --> 00:46:39.222
I'm not sure where it goes.

790
00:46:39.262 --> 00:46:42.124
I mean, this is not a cyber-only conversation.

791
00:46:42.164 --> 00:46:42.744
No, not at all.

792
00:46:42.925 --> 00:46:45.647
Like, we just saw Claude Meetings come out yesterday.

793
00:46:45.667 --> 00:46:52.332
And it's like the laundry list of startups that do like AI meeting notes type things.

794
00:46:52.512 --> 00:46:54.413
And it's just like, oh, well, now it's just in Claude.

795
00:46:54.753 --> 00:46:56.635
Like, that's just there now.

796
00:46:57.275 --> 00:46:57.495
Right.

797
00:46:58.735 --> 00:46:59.376
It's going to get weird.

798
00:46:59.396 --> 00:47:01.196
It's going to get weird here for a bit.

799
00:47:01.296 --> 00:47:02.696
But I thought it was just super fascinating.

800
00:47:02.716 --> 00:47:06.437
We don't have a ton of details about like their cyber offering out of OpenAI.

801
00:47:06.537 --> 00:47:07.277
I'm interested.

802
00:47:08.298 --> 00:47:10.778
Obviously, the red team stuff is basically everything we're talking about.

803
00:47:10.898 --> 00:47:11.118
Right.

804
00:47:11.198 --> 00:47:13.539
Like that we've been talking about is like, do you want to find bugs?

805
00:47:13.619 --> 00:47:14.119
Like, cool.

806
00:47:14.139 --> 00:47:17.280
We got a bug finding super, you know, machine here.

807
00:47:17.320 --> 00:47:17.520
Right.

808
00:47:18.320 --> 00:47:19.000
The blue team stuff.

809
00:47:19.040 --> 00:47:24.281
It's not like they are going into the AI sock space or SIM space or anything like that.

810
00:47:24.801 --> 00:47:26.562
It's just fix the bugs that we just found.

811
00:47:27.382 --> 00:47:27.542
Right.

812
00:47:27.562 --> 00:47:29.503
So they're just staying close to the code right now.

813
00:47:29.703 --> 00:47:32.544
So I think that that's a bit of runway, right?

814
00:47:32.944 --> 00:47:37.366
If you're just an AI bug finder, you're probably not happy about this, right?

815
00:47:37.426 --> 00:47:38.446
Bug finder, bug fixer.

816
00:47:39.027 --> 00:47:48.850
But I think like there's a laundry list of companies that are doing more security operations, you know, either offensively or defensively, right?

817
00:47:48.890 --> 00:47:51.651
But just like more into the security team with AI.

818
00:47:51.671 --> 00:47:55.353
And I don't think that this announcement really impacts that much at all, right?

819
00:47:55.393 --> 00:47:55.593
Yeah.

820
00:47:55.993 --> 00:48:04.875
It's not like you're hiring them to, you're not going to go to API, OpenAI directly to, you know, run your SOC alerts, right?

821
00:48:05.035 --> 00:48:07.656
Off of daybreak type stuff.

822
00:48:07.696 --> 00:48:10.797
It seems like mostly, hey, they have this model.

823
00:48:11.057 --> 00:48:18.319
And like, I want to be clear until very recently, a lot of these models were not even intentionally trained with any additional cyber intelligence.

824
00:48:18.379 --> 00:48:22.380
It's just they got continually better at code and engineering.

825
00:48:23.181 --> 00:48:34.341
that, you know, so much so that like stuff like memory bugs and injection bugs and things like that that are very, very syntactical and like, you know, whatever, that that's okay, now we're just playing in their playground at this point.

826
00:48:36.192 --> 00:49:02.697
weird stuff super weird um speaking of injection bugs and uh code that just should you know someone should just go over to fortinet and be like we're gonna one shot rewrite everything like let's do it like rewrite fortinet in rust make no mistakes how do you think do you think we would be better off or or what do you i don't know bro i actually highlighted that in my um open ai thousand lpe's video like

827
00:49:04.133 --> 00:49:05.473
Does Rust even matter anymore?

828
00:49:05.493 --> 00:49:10.675
Because if the AIs are so good at coding, do the promises that Rust offers matter?

829
00:49:11.015 --> 00:49:11.635
You know what I mean?

830
00:49:11.675 --> 00:49:16.216
Versus just like the mem copy where N is a network value can't happen.

831
00:49:16.877 --> 00:49:17.537
Are you saying?

832
00:49:17.797 --> 00:49:18.457
Are you saying?

833
00:49:18.497 --> 00:49:18.877
Hold on.

834
00:49:19.097 --> 00:49:28.520
Are you saying that we invented entire programming languages that were memory safe because of skill issues with other programming languages?

835
00:49:28.600 --> 00:49:28.840
Yes.

836
00:49:29.200 --> 00:49:29.560
I am.

837
00:49:29.760 --> 00:49:30.541
I am saying that.

838
00:49:30.781 --> 00:49:33.042
And people are like, oh, it's only a skill issue.

839
00:49:33.082 --> 00:49:34.102
Like, just be good.

840
00:49:34.122 --> 00:49:35.143
I'm like, okay, cool.

841
00:49:35.223 --> 00:49:36.783
Microsoft would love your opinion on that.

842
00:49:36.803 --> 00:49:38.024
That's awesome.

843
00:49:38.084 --> 00:49:40.665
I'm sure that they would love to hear just hire better programmers.

844
00:49:41.366 --> 00:49:42.786
Good fucking idea, dude.

845
00:49:43.086 --> 00:49:46.608
Yeah, it's been 30-something years of C code.

846
00:49:46.788 --> 00:49:48.649
Yeah, surely we'll figure it out one of these decades.

847
00:49:49.609 --> 00:50:16.520
just be better at sea oh okay so we're only going to have the best that makes perfect sense this Fortinet to shell vulnerability is interesting because when I saw Fortinet I was like oh god it's another endpoint device special on the lowdown this week interestingly enough there's actually a vulnerability in FortiClient which as far as I understand it is the client that you use to connect to an Scell VPN so Fortinet

848
00:50:17.280 --> 00:50:23.061
They're a device manufacturer where you can use their endpoint to act as like a VPN concentrator, right?

849
00:50:23.081 --> 00:50:27.562
And you can VPN into like your corporate network and do HR work from your house.

850
00:50:27.582 --> 00:50:28.782
But you have to go inside the network to do it.

851
00:50:29.442 --> 00:50:42.105
The vulnerability actually allows a malicious website to pop calc or whatever code execution on your computer because of vulnerability in the FortiPAM portable authentication module inside of FortiClient.

852
00:50:42.345 --> 00:50:42.525
Right.

853
00:50:42.585 --> 00:50:55.034
So there's literally a POC, not really a POC, but like a video of the, the, the person who found this executing the vulnerability where literally they just have for the client with for the PEM on their, their device, on their little home computer.

854
00:50:55.434 --> 00:50:58.216
They go to a website, pop calc, GG.

855
00:50:59.320 --> 00:50:59.900
That's insane.

856
00:51:00.501 --> 00:51:17.130
Now, what I don't know is if this is a command injection vulnerability, where just, like, you do shell execute a calc.exe, or there's actually, like, some ROP, some, you know, stuff going on, some memory corruption that actually they have to kind of tailor this to the exact target.

857
00:51:17.190 --> 00:51:29.036
But regardless of that, it's a really kind of neat vulnerability where, you know, the era of browser go to website, get a shell on your computer is alive and well, despite memory safety in V8, you know?

858
00:51:29.856 --> 00:51:30.736
Man, pretty weird.

859
00:51:32.197 --> 00:51:35.817
I like I reiterate my my point earlier.

860
00:51:36.098 --> 00:51:39.058
Just take the two shell suffix out back.

861
00:51:39.198 --> 00:51:42.519
And I literally brought this up and I was like, guess what they called it?

862
00:51:43.279 --> 00:51:44.540
Guess what they called it, low level.

863
00:51:44.560 --> 00:51:46.140
Yeah, yeah, yeah.

864
00:51:46.360 --> 00:51:49.541
I literally as a joke, I was like, oh, is it Fortinet to shell?

865
00:51:50.081 --> 00:51:50.821
He's like, yeah.

866
00:51:51.401 --> 00:51:57.843
So the other part, the other part that I thought was kind of funny about this is, did you say that they're just getting out of the SSL VPN business?

867
00:51:58.243 --> 00:52:09.168
Yeah, so I Googled, just to get some background context on this, I Googled Fortinet SSL VPN, and they said, yeah, as of May of 2026, we are no longer going to provide SSL VPN services.

868
00:52:09.448 --> 00:52:12.050
So they just are not going to run that part of the company anymore.

869
00:52:12.250 --> 00:52:12.890
Moon's haunted.

870
00:52:13.110 --> 00:52:14.151
They were like, you know what?

871
00:52:14.171 --> 00:52:14.691
We tried.

872
00:52:14.711 --> 00:52:15.431
We tried.

873
00:52:15.471 --> 00:52:17.252
We tried to go back, baby.

874
00:52:17.312 --> 00:52:18.373
Good effort.

875
00:52:19.073 --> 00:52:23.355
Turns out really hard to continue to do this without pumping zero days out.

876
00:52:23.595 --> 00:52:24.775
You know, okay.

877
00:52:25.296 --> 00:52:28.917
I want to pause here for a second because we dunk, we dunk, we dunk, we dunk, we dunk, we dunk.

878
00:52:29.137 --> 00:52:30.618
And by the way, more dunks are coming.

879
00:52:30.818 --> 00:52:31.558
More dunks are coming.

880
00:52:31.578 --> 00:52:31.998
That's fine.

881
00:52:32.038 --> 00:52:32.758
Yeah.

882
00:52:33.439 --> 00:52:36.120
I have like a serious, I want your serious opinion about this.

883
00:52:36.280 --> 00:52:36.560
Okay.

884
00:52:37.540 --> 00:52:39.501
For these types of companies, right?

885
00:52:39.981 --> 00:52:44.223
And I know that neither of us are like qualified to actually answer the question I'm about to pose to you.

886
00:52:44.243 --> 00:52:45.403
So like, okay, whatever.

887
00:52:45.443 --> 00:52:45.603
Right.

888
00:52:46.104 --> 00:52:47.024
But like,

889
00:52:48.217 --> 00:52:49.138
What is stopping them?

890
00:52:49.418 --> 00:52:53.259
The vulnerabilities coming out of these edge device companies, right?

891
00:52:53.480 --> 00:52:57.942
Like the Cisco SQL injection one in the email gateway that we talked about a few weeks ago.

892
00:52:57.962 --> 00:53:00.022
Fortinet constantly, right?

893
00:53:00.983 --> 00:53:01.523
What's the other one?

894
00:53:01.543 --> 00:53:02.984
Avanti constantly, right?

895
00:53:03.304 --> 00:53:05.085
Like all of these like edge devices.

896
00:53:05.505 --> 00:53:08.446
What do you think is stopping them from seriously?

897
00:53:08.546 --> 00:53:16.390
Like the types of vulnerabilities that we are finding are things that literally AI can like just will not code anymore, right?

898
00:53:16.450 --> 00:53:16.950
Like you said.

899
00:53:17.630 --> 00:53:19.291
What are we waiting for?

900
00:53:19.571 --> 00:53:25.355
If you're a publicly traded company with massive deployments of edge devices, what are we waiting for?

901
00:53:26.478 --> 00:53:31.503
Before I answer that, I will say I've been saying a lot that like AI can't produce these code patterns.

902
00:53:31.863 --> 00:53:39.251
I actually pretty recently had AI code a really trivial path traversal bug like for me on an actual product that I'm making.

903
00:53:39.691 --> 00:53:40.873
So it is possible.

904
00:53:41.273 --> 00:53:43.896
I'm not actually sure how it managed to pull this off, but it did do that.

905
00:53:43.936 --> 00:53:48.641
So we'll just kind of it's less likely that an AI will do it than a human, but it is still possible.

906
00:53:49.461 --> 00:53:53.824
Now, your question is, like, why are they not just auditing these products or rewriting them?

907
00:53:54.564 --> 00:53:56.265
I mean, man, you've ran security teams.

908
00:53:56.285 --> 00:53:57.126
I've ran security teams.

909
00:53:57.146 --> 00:53:58.747
I think it's literally just a manning issue.

910
00:53:58.787 --> 00:54:03.049
Like, they don't have enough people to be able to give a shit about all their products.

911
00:54:03.209 --> 00:54:05.671
Because at the end of the day, like, you only have 10 people.

912
00:54:05.691 --> 00:54:07.692
And if you want 11 people, you've got to get more money.

913
00:54:07.712 --> 00:54:10.774
And if you need more money, you've got to convince the CFO and the CEO to give you more money.

914
00:54:11.154 --> 00:54:17.398
I just don't think these companies are managed in a way where there are enough people that can do the thing they need to do.

915
00:54:17.736 --> 00:54:18.176
You're right.

916
00:54:18.357 --> 00:54:18.737
You're right.

917
00:54:18.817 --> 00:54:19.417
And I know this.

918
00:54:19.518 --> 00:54:23.961
And that's why I said, we're not even qualified because it's not that they don't know how to.

919
00:54:24.142 --> 00:54:27.044
It's not that they don't have the tools to do it, right?

920
00:54:27.565 --> 00:54:32.229
It is financial, political, business priorities, all this kind of stuff.

921
00:54:32.589 --> 00:54:35.071
But if you're me, right?

922
00:54:35.331 --> 00:54:43.899
And I'm the head of a publicly traded company that has these edge devices that are supposed to do security for people out there.

923
00:54:44.379 --> 00:54:44.920
I like...

924
00:54:46.282 --> 00:54:49.643
I have a really hard time trying to figure out what else my priority would be.

925
00:54:49.663 --> 00:54:54.464
I could see a ship zero days in my security edge devices.

926
00:54:54.925 --> 00:55:00.406
I could see a world where you enter kind of cognitive dissonance and you're like, ah, the AI super hacker, nothing we can do.

927
00:55:01.026 --> 00:55:01.526
You know what I mean?

928
00:55:01.566 --> 00:55:05.468
Like maybe, man, maybe it's just like, I'm not saying for anybody is like this.

929
00:55:05.508 --> 00:55:13.470
I'm just saying when you have so many competing priorities and you have to make X people, happy shareholders, board, whatever, um, you know,

930
00:55:14.625 --> 00:55:18.048
you probably find ways to justify the fact that you're not doing all of your priorities.

931
00:55:18.068 --> 00:55:18.769
Yeah, it's money.

932
00:55:19.289 --> 00:55:19.590
It's money.

933
00:55:19.610 --> 00:55:20.511
I'm just being cynical.

934
00:55:20.791 --> 00:55:21.451
No, I'm with you.

935
00:55:21.471 --> 00:55:22.713
I mean, it sucks.

936
00:55:22.733 --> 00:55:23.573
We all know this, right?

937
00:55:23.794 --> 00:55:24.614
We've highlighted this a lot.

938
00:55:24.634 --> 00:55:28.898
The technical solution is not the solution because it's always more complicated than that, unfortunately.

939
00:55:28.918 --> 00:55:30.059
And not to say that, like...

940
00:55:30.800 --> 00:55:32.361
zero days are going to stop existing.

941
00:55:32.381 --> 00:55:34.803
Like, you know, not to say that we're going to get to the bottom of this.

942
00:55:35.063 --> 00:55:35.283
Right.

943
00:55:35.403 --> 00:55:38.726
But it's just like the vulns that we're talking about, man.

944
00:55:39.066 --> 00:55:39.326
Yeah.

945
00:55:39.346 --> 00:55:39.927
You know what I mean?

946
00:55:39.967 --> 00:55:44.270
It's just stuff that like pre-AI shouldn't have happened either.

947
00:55:44.650 --> 00:55:45.011
Right.

948
00:55:45.031 --> 00:55:48.093
Like, it's not like, oh, you know, this is H1.

949
00:55:48.693 --> 00:55:51.215
That's, that's just like kind of funny and crazy.

950
00:55:51.235 --> 00:55:55.879
And like, it's been, you know, it, that whole story took decades to unfold.

951
00:55:55.939 --> 00:55:57.400
So like stub our toe on that bone.

952
00:55:57.440 --> 00:55:57.600
Right.

953
00:55:57.620 --> 00:55:59.261
It's not like, oh my God, how did we do this?

954
00:55:59.321 --> 00:56:00.282
That's not, you know what I mean?

955
00:56:00.362 --> 00:56:02.003
You can see very clearly how that happened.

956
00:56:02.023 --> 00:56:03.284
Totally, totally, totally.

957
00:56:03.304 --> 00:56:05.425
Even though it's a major bug, I'm not like dunking on it.

958
00:56:05.626 --> 00:56:06.987
Like that's just crazy, right?

959
00:56:07.407 --> 00:56:17.313
But like, man, some of these, right, that like we keep talking about in these literally, you know, multi-billion dollar publicly traded cybersecurity companies.

960
00:56:17.333 --> 00:56:24.498
Yeah, but what was the one that was like an SQL injection from a field in an email in an email security gateway?

961
00:56:24.919 --> 00:56:25.059
Yeah.

962
00:56:25.079 --> 00:56:26.640
Like how does that happen?

963
00:56:26.660 --> 00:56:26.760
Yeah.

964
00:56:27.240 --> 00:56:27.420
Right.

965
00:56:27.881 --> 00:56:32.024
Or like how many of these public management interface login portals that like pre-off?

966
00:56:32.545 --> 00:56:38.030
Like, how do you not at this point, I don't care, like budget staffing, all this kind of stuff.

967
00:56:38.130 --> 00:56:44.676
How do you not have a frontier model just banging on your public facing management portal stuff?

968
00:56:45.076 --> 00:56:45.296
Right.

969
00:56:45.516 --> 00:56:45.857
Sure.

970
00:56:46.217 --> 00:56:48.259
Most of your code is beyond that.

971
00:56:49.384 --> 00:56:54.506
Like, most of the code of these products is beyond that public-facing thing, right?

972
00:56:54.607 --> 00:56:59.509
Like, you know, most of the functionality that could be reached by an attacker is beyond that auth.

973
00:56:59.769 --> 00:57:01.590
And it's like, okay, whatever, right?

974
00:57:02.030 --> 00:57:04.251
Like, we're going to find vulns in there forever.

975
00:57:04.411 --> 00:57:11.835
But, like, how are you not just, like, throwing something at this publicly-facing part of all of these products right now?

976
00:57:11.855 --> 00:57:13.656
Yeah, so some platforms are hard to fuzz, right?

977
00:57:13.676 --> 00:57:16.297
Like, embedded stuff in particular is hard to fuzz because of, like, the...

978
00:57:17.602 --> 00:57:19.643
hardware peripherals it may depend on.

979
00:57:19.683 --> 00:57:26.085
Like in the Citrix case, right, they're literally an application gateway, so their performance is super important to them.

980
00:57:26.105 --> 00:57:30.106
So the way they architect it is that, like, all the code goes for the kernel first, and, like, some of it's dispatched out.

981
00:57:30.466 --> 00:57:31.447
Maybe that's hard to fuzz.

982
00:57:31.747 --> 00:57:33.067
Maybe their tool chain is super old.

983
00:57:33.087 --> 00:57:38.169
Like, if you look at it, it was running FreeBSD, and, like, PIE wasn't enabled, and, like, ASLR wasn't on.

984
00:57:38.189 --> 00:57:40.610
So to me, that just implies very old tool chain,

985
00:57:41.190 --> 00:57:51.358
A company that sells a product is not incentivized to update their tool chain, which means that, A, the security mitigations aren't enabled, and, B, it's very hard to get things like ASAN or LibFuster to run on those tool chains.

986
00:57:51.378 --> 00:57:59.384
So it's just like traditional corporate inertia, traditional corporate bureaucracy kind of set the stage for those major issues, right?

987
00:57:59.628 --> 00:58:05.631
Yeah, I know I'm going to get dunked on for this section, but I'm just, it's hard, man, when we do this for a living, right?

988
00:58:05.671 --> 00:58:08.173
And we're just like looking at all these bugs going like, what's this one?

989
00:58:08.213 --> 00:58:08.813
This bug?

990
00:58:08.933 --> 00:58:11.335
Like, are we, you know, what's going on?

991
00:58:12.055 --> 00:58:13.936
All right, you want to dunk one more time?

992
00:58:14.416 --> 00:58:14.857
Yeah, dude.

993
00:58:14.937 --> 00:58:16.017
The number of the year?

994
00:58:16.778 --> 00:58:17.698
Four, baby.

995
00:58:17.818 --> 00:58:19.379
New meme number just dropped.

996
00:58:19.739 --> 00:58:20.520
Okay, so.

997
00:58:20.540 --> 00:58:24.062
A mud rub, right?

998
00:58:25.265 --> 00:58:28.950
This happened when we were at OneCon, Sentinel One's conference.

999
00:58:28.970 --> 00:58:29.270
Yeah.

1000
00:58:29.510 --> 00:58:40.565
Basically, it turned out that the private keys, plural, for eight intermediate certs for a certificate authority, eMudra, were single-digit integers, one of them being four.

1001
00:58:40.585 --> 00:58:40.685
Yeah.

1002
00:58:41.025 --> 00:58:42.406
Fucking hilarious.

1003
00:58:42.886 --> 00:58:44.828
The question becomes, how does this happen?

1004
00:58:45.168 --> 00:58:48.230
How is it possible that the cert number that the private key is for?

1005
00:58:48.810 --> 00:58:52.833
A little bit of detail came out, so we don't have the exact script that went down, right?

1006
00:58:53.353 --> 00:58:58.056
But basically what happened is, and the reason why I'm highlighting this one page, I'm going to have the editor put it on the screen.

1007
00:58:58.916 --> 00:59:04.100
It's like Emudra's FAQ for their HSM security.

1008
00:59:04.700 --> 00:59:15.235
How does Emudra integrate FIPS-certified HSMs to protect private keys and ensure crypto agility for post-quantum cryptography algorithm rollouts?

1009
00:59:16.295 --> 00:59:20.157
If you know anything about FIPS, which I really didn't until I read more into this.

1010
00:59:20.177 --> 00:59:28.920
So FIPS is the Federal Information Processing Standard, which basically means if a device is FIPS compliant, it means it is allowed to process government data.

1011
00:59:28.940 --> 00:59:34.102
It is also internationally recognized as like if you are FIPS compliant, you can be a CA, for example.

1012
00:59:34.142 --> 00:59:34.303
Right.

1013
00:59:34.903 --> 00:59:41.205
Being FIPS compliant means that you implement the NIST guidelines, the National Institute for Standards and something.

1014
00:59:41.806 --> 00:59:41.946
Right.

1015
00:59:42.726 --> 00:59:43.747
On cryptography, okay?

1016
00:59:44.287 --> 00:59:53.312
One of the FIPS requirements is that you're not able to export in plain text the private key for an HSM, okay?

1017
00:59:53.492 --> 00:59:57.354
And the HSM that Imudra was using had FIPS mode enabled.

1018
00:59:57.394 --> 00:59:58.054
Isn't that great, Matt?

1019
00:59:58.074 --> 00:59:58.995
Guys, look, they're FIPS compliant.

1020
00:59:59.015 --> 00:59:59.975
How could this possibly have happened?

1021
00:59:59.995 --> 01:00:02.777
Matt, they disabled FIPS mode

1022
01:00:03.537 --> 01:00:15.806
And exported the private key in plain text, which is fair because the script they were going to use, Matt, they were going to encrypt the private key for ingestion into a new hardware security module.

1023
01:00:16.727 --> 01:00:21.050
The vulnerability that happened, the issue that happened was something in their script.

1024
01:00:21.070 --> 01:00:21.670
We don't have the details.

1025
01:00:21.690 --> 01:00:22.331
We don't have the script.

1026
01:00:23.117 --> 01:00:25.599
Something in the script truncated the key.

1027
01:00:25.739 --> 01:00:29.581
So it read out the whole key, all N bytes, right?

1028
01:00:30.002 --> 01:00:33.284
And it made it so the key was only the last byte of the key.

1029
01:00:33.924 --> 01:00:36.686
So in the case we talked about, the last byte of the key was four.

1030
01:00:36.746 --> 01:00:37.767
That's the new private key.

1031
01:00:37.807 --> 01:00:40.149
And they ingested that into their new HSM.

1032
01:00:40.669 --> 01:00:42.090
And then the new HSM was fifth mode.

1033
01:00:43.522 --> 01:00:48.105
point of an HSM is that the key doesn't leave the HSM.

1034
01:00:48.265 --> 01:00:49.426
Correct.

1035
01:00:49.666 --> 01:00:51.087
That's the whole point.

1036
01:00:51.708 --> 01:00:56.351
I'm shocked the HSM doesn't have a way to export it encrypted.

1037
01:00:56.871 --> 01:00:57.772
How is that not a thing?

1038
01:00:58.773 --> 01:00:59.933
It sounds like they got around that.

1039
01:01:00.253 --> 01:01:02.654
It sounds like they disabled that, right?

1040
01:01:02.734 --> 01:01:03.574
So why would you, though?

1041
01:01:03.594 --> 01:01:07.396
Like, again, I have never, like, physically put my hands on an HSM.

1042
01:01:07.676 --> 01:01:08.476
I can't speak to this.

1043
01:01:09.116 --> 01:01:17.339
But as far as I understand it, it's meant to ingest and cryptographically secure the key material so that no one can see it.

1044
01:01:17.899 --> 01:01:21.120
And that if it ever comes out, it comes out encrypted and signed.

1045
01:01:21.780 --> 01:01:24.441
So I would imagine HSM has that functionality, right?

1046
01:01:24.461 --> 01:01:29.724
And that's why the NIST guidelines are literally, hey, you can export private keys, but they have to come out encrypted.

1047
01:01:30.364 --> 01:01:32.846
There's no way that HSMs don't do that.

1048
01:01:33.166 --> 01:01:37.708
Maybe it's because they're going between HSM vendors, which is part of the write-up by Mozilla.

1049
01:01:37.748 --> 01:01:43.691
Like, it's going from vendor A to vendor B, and, like, maybe the export file format isn't correct.

1050
01:01:43.771 --> 01:01:45.132
So, like, they have to do it themselves.

1051
01:01:45.232 --> 01:01:45.672
I don't know.

1052
01:01:45.892 --> 01:01:52.758
Yeah, and obviously they were thinking that they were going to do something with some Python script to manipulate it, and they couldn't manipulate it if it was encrypted.

1053
01:01:52.778 --> 01:01:54.120
Right, right, right.

1054
01:01:54.180 --> 01:01:56.061
But what is the thought process here?

1055
01:01:56.262 --> 01:01:59.264
I just don't even understand the thought process.

1056
01:01:59.685 --> 01:02:03.288
Why even spend all the money on vendor A or vendor B HSMs if you're going to do this?

1057
01:02:03.328 --> 01:02:05.190
This defeats the purpose of the HSMs.

1058
01:02:05.510 --> 01:02:10.076
It is funny, though, but, like, I was about to say, why didn't they look at the private key and check?

1059
01:02:10.496 --> 01:02:15.702
Well, actually, they did the right thing by not, like, printing the private key, right?

1060
01:02:15.722 --> 01:02:15.822
Yeah.

1061
01:02:15.842 --> 01:02:16.263
They, like...

1062
01:02:16.784 --> 01:02:17.965
But they just trusted it.

1063
01:02:18.366 --> 01:02:19.126
Yeah, right.

1064
01:02:19.186 --> 01:02:21.129
I mean, maybe you print land of private key.

1065
01:02:21.229 --> 01:02:21.890
I don't know, but...

1066
01:02:22.310 --> 01:02:26.333
At least look at it when you're devving that Python tool that you were doing, right?

1067
01:02:26.373 --> 01:02:30.015
And be like, all right, did I accidentally truncate this to one character?

1068
01:02:30.035 --> 01:02:30.796
That is the frustrating part.

1069
01:02:30.816 --> 01:02:34.558
It implies that there wasn't robust testing of that Python script.

1070
01:02:34.638 --> 01:02:35.059
At all.

1071
01:02:35.239 --> 01:02:35.539
Yeah.

1072
01:02:35.859 --> 01:02:36.800
Not robust.

1073
01:02:36.840 --> 01:02:41.083
They didn't see that the output of a single character is crazy, right?

1074
01:02:41.463 --> 01:02:42.444
Never mind look at it.

1075
01:02:42.464 --> 01:02:47.767
How did the unit test not say, hey, by the way, the output of this is not an effing key?

1076
01:02:47.947 --> 01:02:48.268
Yeah.

1077
01:02:48.808 --> 01:02:58.531
To our point about AI before, too, there is an argument to be made about, hey, maybe we don't write scripts that deal with CAs, with AI for privacy concerns of the key material getting sucked into the model.

1078
01:02:58.911 --> 01:03:01.491
But at the same time, an AI would have caught this.

1079
01:03:01.931 --> 01:03:03.852
Well, you can write the script.

1080
01:03:04.012 --> 01:03:10.534
It's just not then... You don't want to be using AI in line for actual key material stuff.

1081
01:03:10.574 --> 01:03:11.854
But you can use a model to write this script

1082
01:03:16.035 --> 01:03:17.376
Why does the script exist?

1083
01:03:17.397 --> 01:03:18.017
What are we doing?

1084
01:03:18.137 --> 01:03:20.839
I don't even understand what they're doing.

1085
01:03:20.899 --> 01:03:23.681
The whole point of HSM is that that's it.

1086
01:03:24.482 --> 01:03:27.384
That no one can see the key material because it's still in there.

1087
01:03:27.685 --> 01:03:31.628
To turn off FIPS mode and then do that is just insane.

1088
01:03:32.068 --> 01:03:33.369
Further dunk.

1089
01:03:33.729 --> 01:03:35.471
I'm not trying to dunk per se.

1090
01:03:36.051 --> 01:03:41.435
The whole situation is so unbelievable that everything that I'm going to say comes off as a dunk, so my bad.

1091
01:03:42.136 --> 01:03:44.658
Regardless of our dunking, Emudra is no longer in the

1092
01:03:45.258 --> 01:03:46.079
certificate authority.

1093
01:03:46.099 --> 01:03:47.300
It's a second in a row.

1094
01:03:47.380 --> 01:03:50.041
Two vendors in a row that have failed so miserably.

1095
01:03:50.081 --> 01:03:50.802
They're like, you know what?

1096
01:03:50.822 --> 01:03:51.082
We're out.

1097
01:03:51.102 --> 01:03:51.502
We don't care.

1098
01:03:51.522 --> 01:03:51.823
We're done.

1099
01:03:51.883 --> 01:03:52.563
You know what?

1100
01:03:52.803 --> 01:03:56.486
This whole TLS certificate authority thing for us, not for us.

1101
01:03:56.526 --> 01:03:56.886
We're good.

1102
01:03:56.906 --> 01:03:57.486
Check it out.

1103
01:03:57.886 --> 01:03:58.727
Moon was haunted.

1104
01:03:58.807 --> 01:03:59.728
We're coming home.

1105
01:03:59.748 --> 01:04:00.668
Moon's haunted.

1106
01:04:00.728 --> 01:04:01.749
We don't even want to go back.

1107
01:04:01.889 --> 01:04:02.710
That's not a thing.

1108
01:04:02.970 --> 01:04:03.610
Moon's haunted.

1109
01:04:04.091 --> 01:04:05.652
So interesting week for sure.

1110
01:04:05.672 --> 01:04:12.436
I think we mentioned this with John, but it's just like, it's funny how only two 10.0s came out this week, so it's kind of a slow week for us.

1111
01:04:12.456 --> 01:04:13.897
You know, like, all right.

1112
01:04:14.217 --> 01:04:15.898
The most important part of the podcast, Ed.

1113
01:04:16.459 --> 01:04:16.779
Oh, no.

1114
01:04:16.979 --> 01:04:18.541
We already did the vendor cut, or the sponsor cut.

1115
01:04:18.821 --> 01:04:19.541
Nope.

1116
01:04:19.561 --> 01:04:20.522
Most important part.

1117
01:04:20.542 --> 01:04:21.002
Sorry, yes.

1118
01:04:21.042 --> 01:04:23.064
Second most important part of the podcast.

1119
01:04:23.224 --> 01:04:23.624
Okay.

1120
01:04:24.645 --> 01:04:25.386
Alliance or Horde?

1121
01:04:26.487 --> 01:04:26.647
Oh.

1122
01:04:27.427 --> 01:04:28.708
On PvP, Horde, buddy.

1123
01:04:29.249 --> 01:04:29.529
Okay.

1124
01:04:29.729 --> 01:04:30.009
Okay.

1125
01:04:30.370 --> 01:04:35.153
I'm like spiritually Horde, but I got bullied into playing an Alliance character on forever.

1126
01:04:35.914 --> 01:04:37.335
Ironically, I'm the opposite.

1127
01:04:37.795 --> 01:04:43.079
My fondest gaming memories are like being at Sentinel Hill when I was like 14.

1128
01:04:44.024 --> 01:04:44.845
Yeah.

1129
01:04:44.885 --> 01:04:46.526
So I missed WoW in my teenage years.

1130
01:04:46.927 --> 01:04:48.808
I played a little bit during COVID.

1131
01:04:49.769 --> 01:04:51.871
Some friends were like, we should raid.

1132
01:04:52.171 --> 01:04:53.353
What was the expansion that came out?

1133
01:04:53.373 --> 01:04:54.974
Battle for Azeroth, I think it was called.

1134
01:04:55.394 --> 01:04:58.357
So I played a little Battle for Azeroth, which I know is way late.

1135
01:04:58.397 --> 01:05:05.704
So I missed all the Wrath of the Lich King stuff, Burning Crusade, all the stuff that you guys all look back on fondly, I completely missed out on.

1136
01:05:06.224 --> 01:05:06.485
But...

1137
01:05:07.045 --> 01:05:10.349
I do have enough homies playing forever that I was like, all right, I'll download forever.

1138
01:05:10.589 --> 01:05:11.910
So I've been putzing around with it.

1139
01:05:12.050 --> 01:05:20.459
But hilariously, all of my algorithms have me pinned because all the videos that I'm getting recommended are wow for busy dads.

1140
01:05:20.559 --> 01:05:24.183
And like wow for losers who can't actually play wow.

1141
01:05:24.223 --> 01:05:26.265
You know, and I was like, oh God, how did they know?

1142
01:05:26.365 --> 01:05:26.725
How did they?

1143
01:05:26.826 --> 01:05:28.087
I didn't, I didn't search that.

1144
01:05:28.347 --> 01:05:30.129
Like it is true, dude.

1145
01:05:30.169 --> 01:05:34.172
Like I thought I would be hungry for like the classic feel of like the OG game.

1146
01:05:34.693 --> 01:05:37.295
And then like, I've got, I've got kiddos and this job.

1147
01:05:37.315 --> 01:05:43.961
And I realized like, Oh my God, when I was a kid, I would just degen and play for like 10 hours a day for weeks.

1148
01:05:44.681 --> 01:05:44.901
Yeah.

1149
01:05:44.921 --> 01:05:45.702
That's all we did.

1150
01:05:45.722 --> 01:05:46.202
You can't do it.

1151
01:05:46.222 --> 01:05:46.723
You can't do it.

1152
01:05:46.743 --> 01:05:47.684
I can't do that anymore.

1153
01:05:47.784 --> 01:05:48.965
I will say, I don't know.

1154
01:05:49.545 --> 01:05:51.207
You know, I didn't really have that experience.

1155
01:05:51.587 --> 01:05:51.767
Um,

1156
01:05:52.788 --> 01:05:58.471
I did do that a bit in Final Fantasy XIV again during COVID, right?

1157
01:05:58.792 --> 01:06:02.734
I just ruined my sleep schedule and got really into that as an MMO.

1158
01:06:02.754 --> 01:06:06.196
It's the only MMO that I really got like super, super, super into.

1159
01:06:06.476 --> 01:06:11.539
But I will say, I don't know, Forever feels like there's not a lot of that pressure.

1160
01:06:11.559 --> 01:06:13.840
It feels a little cozier.

1161
01:06:13.920 --> 01:06:17.682
It feels like everyone's just trying to go around and kick around.

1162
01:06:17.762 --> 01:06:20.224
I see a lot of people like fishing, you know what I mean?

1163
01:06:20.244 --> 01:06:20.484
Like...

1164
01:06:20.724 --> 01:06:21.004
Yeah.

1165
01:06:21.284 --> 01:06:38.291
So the problem with retail and with WoW in general is that as the internet got more mature in like the early 2010s, the game got boiled down into a science where like if you were not doing the meta thing, you were not leveling the fastest, you didn't have the best gear, you were an idiot and like shouldn't play the game.

1166
01:06:38.831 --> 01:06:43.392
And I think WoW Forever is an attempt to return to the fact that it's like a game.

1167
01:06:43.532 --> 01:06:43.972
You know what I mean?

1168
01:06:43.992 --> 01:06:44.652
It's not a job.

1169
01:06:44.692 --> 01:06:45.512
It's not that serious.

1170
01:06:45.532 --> 01:06:50.594
Like, if it takes you four months to hit level 40 or whatever, like, that's okay.

1171
01:06:50.834 --> 01:06:52.674
Because the game's going to be here for a long time.

1172
01:06:52.694 --> 01:06:57.955
The idea is they're going away from vertical progression where, like, you've got to build, build, build, build, build, to, like, okay, cool.

1173
01:06:57.975 --> 01:06:59.675
Like, this expansion comes out now.

1174
01:07:00.356 --> 01:07:01.456
Maybe another one comes out later.

1175
01:07:01.496 --> 01:07:02.976
But, like, they don't depend on each other.

1176
01:07:02.996 --> 01:07:03.736
You know what I mean?

1177
01:07:04.196 --> 01:07:04.536
Yeah.

1178
01:07:04.737 --> 01:07:06.517
Which I think makes it closer.

1179
01:07:06.557 --> 01:07:06.697
Yeah.

1180
01:07:07.077 --> 01:07:09.119
Yeah, there was this pressure to get best in slot.

1181
01:07:09.520 --> 01:07:10.721
Sorry if you guys are not getting...

1182
01:07:10.841 --> 01:07:12.022
I got so many DMs about...

1183
01:07:12.102 --> 01:07:15.125
I made a slight joke about... Because I went and spoke at that conference yesterday.

1184
01:07:15.245 --> 01:07:15.466
Yeah.

1185
01:07:15.706 --> 01:07:17.588
And I said, I logged off for WoW for this.

1186
01:07:17.608 --> 01:07:18.989
I got so many DMs about it.

1187
01:07:19.049 --> 01:07:20.150
I was like, oh my God.

1188
01:07:20.391 --> 01:07:24.275
Our audience absolutely is on WoW forever for sure, right?

1189
01:07:24.295 --> 01:07:24.455
Yeah.

1190
01:07:24.995 --> 01:07:29.499
But yeah, like I'm like down for the cozy experience too, right?

1191
01:07:29.559 --> 01:07:33.982
I've got some add-ons from Curse Forge like to help guide me around and level.

1192
01:07:34.543 --> 01:07:37.445
But like I'm down to just like take little side quests too.

1193
01:07:37.465 --> 01:07:39.627
I'm not like trying to get the most efficient thing possible.

1194
01:07:40.087 --> 01:07:45.031
And then like I don't even know what best in slot stuff exists for anything right now.

1195
01:07:45.091 --> 01:07:46.072
You know, I have no idea.

1196
01:07:46.252 --> 01:07:47.433
So like and that's kind of cool.

1197
01:07:48.714 --> 01:07:53.638
I'm just trying to like this, like at this stage in life, I'm trying, I don't want to have a ton of gaming time.

1198
01:07:53.678 --> 01:07:57.861
So when I do, it's like, can I be on discord with some friends that live all over the world?

1199
01:07:58.342 --> 01:08:00.263
And like, can we play a game together for a little while?

1200
01:08:00.443 --> 01:08:01.064
Then that's cool.

1201
01:08:01.084 --> 01:08:01.804
That's valuable.

1202
01:08:02.065 --> 01:08:02.305
Right.

1203
01:08:02.665 --> 01:08:05.347
That's like actual useful time, not D gen time.

1204
01:08:05.607 --> 01:08:05.828
Right.

1205
01:08:05.868 --> 01:08:09.110
That's like, Oh, I'm connecting with my homies that live nowhere near me.

1206
01:08:09.130 --> 01:08:09.731
And this is great.

1207
01:08:09.751 --> 01:08:09.851
Yeah.

1208
01:08:10.411 --> 01:08:16.656
As a man in his 30s, the most fun that I've had so far has not been, like, getting a BIS item in retail WoW.

1209
01:08:16.737 --> 01:08:20.339
It's been, like, getting blown up with my friends in Outlaws.

1210
01:08:20.540 --> 01:08:21.000
You know what I mean?

1211
01:08:21.020 --> 01:08:22.081
Like, that's the fun shit.

1212
01:08:22.141 --> 01:08:22.501
Yeah.

1213
01:08:22.621 --> 01:08:23.442
That's a good time.

1214
01:08:23.842 --> 01:08:24.503
I like a grind.

1215
01:08:24.543 --> 01:08:27.846
I'm, like, if I like gaming, I like a grind game, right?

1216
01:08:28.066 --> 01:08:29.787
So I like that I'm working towards something.

1217
01:08:29.807 --> 01:08:33.710
You know, it's like, yeah, okay, let me, like, run this over and over again for a year.

1218
01:08:33.951 --> 01:08:35.312
And then, like, you know, whatever.

1219
01:08:35.392 --> 01:08:37.113
I'm kind of good with that.

1220
01:08:37.153 --> 01:08:37.894
So, yeah.

1221
01:08:38.214 --> 01:08:41.335
Anyway, I got, I got enough inbound, but I was like, we got to talk about, wow.

1222
01:08:41.355 --> 01:08:42.896
Cause I know you're playing a little bit too.

1223
01:08:43.036 --> 01:08:45.036
We're both, we're both filthy casuals.

1224
01:08:45.196 --> 01:08:47.997
I think you're 13 on like 17 or something.

1225
01:08:48.057 --> 01:08:48.317
Yeah.

1226
01:08:49.057 --> 01:08:56.540
So, uh, you know, I, I've been traveling a bunch, you know, I've, I've logged on maybe two or three nights with some, with some friends and ran some quests.

1227
01:08:56.640 --> 01:08:59.921
So it hasn't ruined my, I haven't let it ruin my sleep schedule yet.

1228
01:09:00.261 --> 01:09:00.441
Yeah.

1229
01:09:02.498 --> 01:09:06.481
Maybe when we start raiding, I'm going to, I'm of course going to roll something that we can play together.

1230
01:09:06.501 --> 01:09:08.442
Or you're going to roll something so we can play together too.

1231
01:09:08.462 --> 01:09:10.363
But yeah, maybe we can start raiding or something.

1232
01:09:10.383 --> 01:09:11.484
Because your ally PvP, you said?

1233
01:09:13.285 --> 01:09:14.086
PVE.

1234
01:09:14.526 --> 01:09:15.407
Oh, your poor PvP.

1235
01:09:15.427 --> 01:09:16.027
But we can play together.

1236
01:09:16.087 --> 01:09:17.148
Yeah.

1237
01:09:17.328 --> 01:09:18.289
PvE Alliance.

1238
01:09:18.329 --> 01:09:18.629
So yeah.

1239
01:09:18.649 --> 01:09:18.769
Okay.

1240
01:09:19.390 --> 01:09:19.550
Cool.

1241
01:09:20.270 --> 01:09:20.530
Sweet.

1242
01:09:20.671 --> 01:09:20.911
All right.

1243
01:09:21.171 --> 01:09:21.451
That's it.

1244
01:09:21.471 --> 01:09:22.232
Take us home, man.

1245
01:09:22.352 --> 01:09:23.733
As usual, we appreciate it.

1246
01:09:23.753 --> 01:09:26.316
Guys, we're wrapping up 2027 here in a little bit.

1247
01:09:26.356 --> 01:09:30.059
You know, it's been a good couple months with the podcast in our first year.

1248
01:09:30.079 --> 01:09:30.360
Episode 20.

1249
01:09:30.380 --> 01:09:32.882
And we really couldn't – oh, yeah.

1250
01:09:33.022 --> 01:09:33.623
This is number 20.

1251
01:09:33.703 --> 01:09:34.424
This is the 20th episode.

1252
01:09:34.684 --> 01:09:36.966
Guys, we really – a serious talk for a second.

1253
01:09:36.986 --> 01:09:42.071
We could not have done this podcast, started this whole thing out without the support from you guys, our listeners.

1254
01:09:42.091 --> 01:09:42.972
Really do appreciate it.

1255
01:09:44.273 --> 01:09:44.793
Next year –

1256
01:09:45.614 --> 01:09:46.895
It's going to be a stellar year for us.

1257
01:09:46.915 --> 01:09:52.439
We're going to have a lot of cool guests on here, doing a lot of cool events, maybe even hosting an event ourself.

1258
01:09:52.459 --> 01:09:53.600
That's kind of in the cards right now.

1259
01:09:54.141 --> 01:09:57.503
So if you want to be in the loop on all that stuff, hit that sub button if you're on YouTube.

1260
01:09:57.523 --> 01:10:01.366
And if you're on Spotify to actually get recommended us again, rate us five stars.

1261
01:10:01.426 --> 01:10:05.890
It actually helps push us out not only to you, but other listeners that are potentially interested as well.

1262
01:10:06.591 --> 01:10:08.492
And maybe you're a sponsor of the podcast.

1263
01:10:08.512 --> 01:10:12.055
We do appreciate your support on getting the team together and making this bad boy happen.

1264
01:10:12.135 --> 01:10:13.416
So appreciate y'all.

1265
01:10:13.616 --> 01:10:14.338
And Apple too.

1266
01:10:14.358 --> 01:10:18.084
You know, I think we're starting to see the snowball pick up on those audio platforms.

1267
01:10:18.105 --> 01:10:23.515
YouTube's weird and spiky and algorithmic and like the other ones, it like really takes you guys helping and supporting and whatever.

1268
01:10:23.595 --> 01:10:24.937
Cause there's no discoverability there.

1269
01:10:24.997 --> 01:10:25.939
So really appreciate it.

1270
01:10:25.979 --> 01:10:26.099
Yeah.

1271
01:10:26.119 --> 01:10:26.320
Yeah.

1272
01:10:26.660 --> 01:10:26.821
Yeah.

1273
01:10:27.282 --> 01:10:27.903
All right.

1274
01:10:27.923 --> 01:10:28.123
Cool.

1275
01:10:28.524 --> 01:10:29.025
Happy hacking.

1276
01:10:29.786 --> 01:10:30.548
Peace.
