WEBVTT

1
00:00:00.320 --> 00:00:03.200
<v Speaker 1>Hey, Carl here, we're on a mission to find new

2
00:00:03.240 --> 00:00:05.919
<v Speaker 1>fans of the show. I mean, where else can you

3
00:00:05.960 --> 00:00:10.240
<v Speaker 1>get real life comedy and horror in one podcast. You

4
00:00:10.279 --> 00:00:13.000
<v Speaker 1>can do your part by leaving us a review on iTunes,

5
00:00:13.039 --> 00:00:16.719
<v Speaker 1>Spotify or wherever you get Security this Week. Also, you

6
00:00:16.719 --> 00:00:19.079
<v Speaker 1>can get an ad free feed by becoming a five

7
00:00:19.120 --> 00:00:22.039
<v Speaker 1>dollars a month patron. Go to Patreon dot Security this

8
00:00:22.079 --> 00:00:25.359
<v Speaker 1>week dot com and sign up, and don't forget discord.

9
00:00:25.600 --> 00:00:28.519
<v Speaker 1>Discord dot Security this week dot com. Lots of good

10
00:00:28.559 --> 00:00:32.280
<v Speaker 1>stuff happening there. Yeah, I think that's it. So on

11
00:00:32.359 --> 00:00:35.479
<v Speaker 1>with the podcast. So I'm at the gym a couple

12
00:00:35.520 --> 00:00:39.280
<v Speaker 1>of days ago and there's this lady on the exercise

13
00:00:39.359 --> 00:00:42.439
<v Speaker 1>bike and she's wearing a helmet on the exercise bike.

14
00:00:42.600 --> 00:00:44.520
<v Speaker 1>So I put on a lifejacket and got on the

15
00:00:44.560 --> 00:00:56.439
<v Speaker 1>rowing machine next to her. Hey, it's been a couple

16
00:00:56.479 --> 00:00:58.200
<v Speaker 1>of weeks since we talked to you. Welcome back to

17
00:00:58.200 --> 00:01:01.439
<v Speaker 1>Security this week. I'm Carl Franklin's the Flat and Patrick Hines.

18
00:01:02.000 --> 00:01:03.960
<v Speaker 1>So we're going to try to catch up by doing

19
00:01:04.040 --> 00:01:07.239
<v Speaker 1>last week's stories and last last week's stories.

20
00:01:07.319 --> 00:01:08.079
<v Speaker 2>Three hour show.

21
00:01:08.959 --> 00:01:12.640
<v Speaker 1>Yeah, starting with last last week's stories, but in conclusion,

22
00:01:12.959 --> 00:01:15.200
<v Speaker 1>you might conclude at the end of these stories that

23
00:01:16.200 --> 00:01:20.159
<v Speaker 1>we are kind of becoming obsolete because they're coming so

24
00:01:20.480 --> 00:01:22.599
<v Speaker 1>fast that you won't have time to patch.

25
00:01:23.319 --> 00:01:26.239
<v Speaker 2>Right, just beyond twenty four to seven security this moment,

26
00:01:26.480 --> 00:01:29.560
<v Speaker 2>just to lead your own system security right now.

27
00:01:30.079 --> 00:01:34.840
<v Speaker 1>All right, let's start with this Indian Express story, no warning,

28
00:01:34.920 --> 00:01:39.760
<v Speaker 1>no confirmation, how an AI agent deleted a startups critical data.

29
00:01:39.920 --> 00:01:42.319
<v Speaker 2>I know, I know, it's stupidity.

30
00:01:43.920 --> 00:01:47.599
<v Speaker 1>Pad CAAC problems exists between keyboard and chair. Yeah, this

31
00:01:47.719 --> 00:01:51.920
<v Speaker 1>is total stupidity. This is basically giving agency to agents

32
00:01:52.040 --> 00:01:54.319
<v Speaker 1>and giving them access to things that they shouldn't have

33
00:01:54.400 --> 00:01:57.319
<v Speaker 1>access to, something we've been warning people against.

34
00:01:58.040 --> 00:02:03.400
<v Speaker 2>If we've been developing between us probably seventy years. Okay, yeah,

35
00:02:03.400 --> 00:02:06.439
<v Speaker 2>I think you're probably right. Yeah, And I don't edit

36
00:02:06.879 --> 00:02:08.120
<v Speaker 2>in production, do you?

37
00:02:08.800 --> 00:02:09.080
<v Speaker 1>I know?

38
00:02:09.360 --> 00:02:13.719
<v Speaker 2>Now, why would you give AI access to productions? There

39
00:02:13.759 --> 00:02:16.360
<v Speaker 2>was a story back. I think it was a couple

40
00:02:16.360 --> 00:02:18.560
<v Speaker 2>of months ago. Actually it was about a month ago

41
00:02:18.599 --> 00:02:22.159
<v Speaker 2>that Amazon was down one of their systems. It wasn't

42
00:02:22.159 --> 00:02:24.960
<v Speaker 2>a primary system, it was an important system, and it's

43
00:02:25.000 --> 00:02:27.599
<v Speaker 2>because somebody pointed an AI agent at it and had

44
00:02:27.599 --> 00:02:29.639
<v Speaker 2>it do an update, and it decided I'm going to

45
00:02:29.639 --> 00:02:32.240
<v Speaker 2>delete it. First and start fresh. It was down for

46
00:02:32.240 --> 00:02:33.159
<v Speaker 2>twelve hours.

47
00:02:33.439 --> 00:02:34.000
<v Speaker 3>That makes sense.

48
00:02:34.039 --> 00:02:34.639
<v Speaker 1>Oh my god.

49
00:02:34.759 --> 00:02:37.919
<v Speaker 2>So this is a this is a bone headed mistake.

50
00:02:38.120 --> 00:02:40.439
<v Speaker 2>That is, it could only be done by somebody who

51
00:02:40.520 --> 00:02:42.400
<v Speaker 2>really doesn't get development. I don't know.

52
00:02:42.560 --> 00:02:45.360
<v Speaker 3>Sometimes you just can't trust AI. So my wife and

53
00:02:45.360 --> 00:02:47.960
<v Speaker 3>I retimes, Yeah, I know, my wife and I, you

54
00:02:47.960 --> 00:02:48.680
<v Speaker 3>can't trust AI.

55
00:02:48.719 --> 00:02:50.759
<v Speaker 2>That's the moral of the story.

56
00:02:51.000 --> 00:02:54.599
<v Speaker 3>We're looking, we're looking to make to get our deck

57
00:02:54.800 --> 00:02:56.520
<v Speaker 3>in our deck. Our porch is falling apart. So my

58
00:02:56.520 --> 00:02:58.680
<v Speaker 3>wife took a picture of the porch and put it

59
00:02:58.759 --> 00:03:01.520
<v Speaker 3>up in AI. And there's six columns on the porch,

60
00:03:01.599 --> 00:03:04.280
<v Speaker 3>six pillars holding the every foot. She wanted to take

61
00:03:04.319 --> 00:03:06.319
<v Speaker 3>the third one out. So I put the picture up

62
00:03:06.360 --> 00:03:08.479
<v Speaker 3>there and I said, hey, can you label these from

63
00:03:08.560 --> 00:03:11.319
<v Speaker 3>left to right one through six. I'd like to take

64
00:03:11.360 --> 00:03:13.240
<v Speaker 3>the third one out, just so it knows which one

65
00:03:13.240 --> 00:03:16.599
<v Speaker 3>the third one was, right. So it labeled them one

66
00:03:16.639 --> 00:03:18.840
<v Speaker 3>through six and took the third one out. Just the

67
00:03:18.840 --> 00:03:23.400
<v Speaker 3>third one had no label. All the restroom were labeled

68
00:03:23.560 --> 00:03:25.400
<v Speaker 3>except for that third one. It took the label out

69
00:03:25.400 --> 00:03:27.360
<v Speaker 3>and I was like, I don't even know what to

70
00:03:27.360 --> 00:03:27.680
<v Speaker 3>do with you.

71
00:03:27.719 --> 00:03:28.800
<v Speaker 1>I'm just get two.

72
00:03:28.639 --> 00:03:29.280
<v Speaker 2>Ballons of milk.

73
00:03:29.319 --> 00:03:30.759
<v Speaker 1>And if they have eggs, get a dozen.

74
00:03:31.240 --> 00:03:32.520
<v Speaker 2>It did it go?

75
00:03:32.680 --> 00:03:35.759
<v Speaker 3>But it was so close My wife and I just

76
00:03:35.759 --> 00:03:40.159
<v Speaker 3>looked at it and went like what never mind. Yeah,

77
00:03:40.199 --> 00:03:41.080
<v Speaker 3>so it's all good.

78
00:03:41.639 --> 00:03:45.879
<v Speaker 1>Yeah, so right now, AI must be curated by people

79
00:03:45.960 --> 00:03:49.560
<v Speaker 1>who are smart enough to know when it's incorrect.

80
00:03:49.639 --> 00:03:53.039
<v Speaker 2>Yep. You shouldn't let AI access to things that you

81
00:03:53.039 --> 00:03:57.879
<v Speaker 2>wouldn't let a a average high school student touch. Or

82
00:03:57.919 --> 00:04:00.719
<v Speaker 2>a chime production is something touch.

83
00:04:00.960 --> 00:04:02.360
<v Speaker 1>A child with a high IQ.

84
00:04:02.759 --> 00:04:03.479
<v Speaker 3>Yeah, there you go.

85
00:04:03.719 --> 00:04:05.840
<v Speaker 2>Yeah, I work at a high school.

86
00:04:05.840 --> 00:04:07.879
<v Speaker 3>It's like that. Most most of them are pretty high IQs.

87
00:04:07.960 --> 00:04:08.199
<v Speaker 1>Yeah.

88
00:04:08.360 --> 00:04:09.719
<v Speaker 2>Even they don't edit production.

89
00:04:09.919 --> 00:04:12.599
<v Speaker 3>Yeah, they don't get access to Bank of America's production.

90
00:04:13.240 --> 00:04:18.319
<v Speaker 1>Yeah. Yeah. So this is pocket Os right is the

91
00:04:18.360 --> 00:04:22.639
<v Speaker 1>company Texas based car rental business, and the incident reportedly

92
00:04:22.639 --> 00:04:25.639
<v Speaker 1>took place over the weekend. When it's autonomous AI tool

93
00:04:25.680 --> 00:04:28.879
<v Speaker 1>erase the database and all backups in nine seconds.

94
00:04:29.000 --> 00:04:30.959
<v Speaker 2>Well, at least it gets points for efficiency.

95
00:04:31.920 --> 00:04:33.920
<v Speaker 3>I mean, here's what I will tell you though. At

96
00:04:33.959 --> 00:04:37.319
<v Speaker 3>some point, somebody's gonna have a bad quarter, somebody's gonna

97
00:04:37.360 --> 00:04:39.560
<v Speaker 3>get random word, somebody's gonna have something bad that happens,

98
00:04:39.680 --> 00:04:41.120
<v Speaker 3>and they're just gonna go, oh, you know what, AI

99
00:04:41.199 --> 00:04:43.639
<v Speaker 3>did it. We lost all our stuff because of AI.

100
00:04:43.839 --> 00:04:45.560
<v Speaker 2>Right, I think that makes them look better.

101
00:04:45.920 --> 00:04:49.439
<v Speaker 3>I don't know at this point. Maybe it does where

102
00:04:49.480 --> 00:04:52.279
<v Speaker 3>they're like, oh that darn AI as opposed to yeah,

103
00:04:52.279 --> 00:04:54.360
<v Speaker 3>we didn't put the proper security things in place to

104
00:04:54.399 --> 00:04:55.600
<v Speaker 3>not get ransomword.

105
00:04:55.439 --> 00:04:59.360
<v Speaker 1>Right, So bonus points if you can name the coding

106
00:04:59.399 --> 00:05:02.959
<v Speaker 1>agent that oh oh oh. And now Dwayne's reading the

107
00:05:03.040 --> 00:05:04.040
<v Speaker 1>article more carefully.

108
00:05:04.160 --> 00:05:06.319
<v Speaker 2>I'm I'm frantically reading through the article.

109
00:05:06.399 --> 00:05:07.959
<v Speaker 1>It was cursor, It was cursor.

110
00:05:08.199 --> 00:05:08.920
<v Speaker 2>Was it cursor?

111
00:05:09.040 --> 00:05:11.519
<v Speaker 1>It was running. It was running on Claude Opus four

112
00:05:11.560 --> 00:05:13.959
<v Speaker 1>point six, which is widely known as.

113
00:05:13.920 --> 00:05:16.120
<v Speaker 3>Oh see, that's the problem. Opus four to seven's out.

114
00:05:16.120 --> 00:05:17.199
<v Speaker 3>I mean, what are you guys doing?

115
00:05:18.639 --> 00:05:21.000
<v Speaker 1>I said last week it was the most advanced and

116
00:05:21.000 --> 00:05:23.560
<v Speaker 1>capable model for coding. Apparently, according to.

117
00:05:23.439 --> 00:05:26.720
<v Speaker 3>This article, I mean, it leads in deletions. Here's the deal, though,

118
00:05:26.839 --> 00:05:30.839
<v Speaker 3>I honestly I don't use Opus for coding. Opus as

119
00:05:30.839 --> 00:05:33.319
<v Speaker 3>great as a deep thinker and a planner. Yeah, I

120
00:05:33.399 --> 00:05:37.279
<v Speaker 3>usually use Codex yeah for coding, So sonic listen, I'd

121
00:05:37.279 --> 00:05:39.120
<v Speaker 3>say pocket os coding's in the name.

122
00:05:39.439 --> 00:05:40.839
<v Speaker 2>Yeah, you gotta Claude's on.

123
00:05:40.920 --> 00:05:41.319
<v Speaker 1>It is good.

124
00:05:41.360 --> 00:05:43.480
<v Speaker 3>You got to use the proper model for the proper.

125
00:05:43.160 --> 00:05:45.439
<v Speaker 1>Thing, Claude so on, it's really good for what I.

126
00:05:45.439 --> 00:05:46.560
<v Speaker 3>Do, Claud's on, it's awesome.

127
00:05:46.839 --> 00:05:47.040
<v Speaker 2>Yeah.

128
00:05:47.040 --> 00:05:48.600
<v Speaker 3>And even if you were to do like an opus

129
00:05:48.639 --> 00:05:50.959
<v Speaker 3>planning layer where it does some really deep dive planning

130
00:05:50.959 --> 00:05:53.439
<v Speaker 3>but doesn't write any code, but then you use a

131
00:05:53.680 --> 00:05:56.839
<v Speaker 3>sonnet to do all the code writing and then maybe

132
00:05:56.879 --> 00:06:00.000
<v Speaker 3>a codex as your tester and testing layer, that's probably

133
00:06:00.040 --> 00:06:01.759
<v Speaker 3>will be way more stable than just open.

134
00:06:01.800 --> 00:06:05.360
<v Speaker 2>You'd still never let it touch production directly.

135
00:06:05.519 --> 00:06:08.240
<v Speaker 3>Oh heck no, yeah, absolutely no, it's.

136
00:06:08.040 --> 00:06:11.160
<v Speaker 2>The fundamental takeaway here. I agree with you, but that's

137
00:06:11.199 --> 00:06:12.079
<v Speaker 2>the fundamental takeaway.

138
00:06:12.079 --> 00:06:14.079
<v Speaker 1>All right. I love the I love the stories where

139
00:06:14.079 --> 00:06:16.560
<v Speaker 1>we get to wag our finger at people that are.

140
00:06:17.560 --> 00:06:18.959
<v Speaker 3>Again, isn't that redundant?

141
00:06:18.959 --> 00:06:21.920
<v Speaker 2>Isn't that just a little isn't that just the stories?

142
00:06:23.319 --> 00:06:24.519
<v Speaker 3>Yeah?

143
00:06:24.560 --> 00:06:27.120
<v Speaker 1>All right, Duane, why don't you take this one about Kelp,

144
00:06:28.519 --> 00:06:29.519
<v Speaker 1>the Kelp.

145
00:06:30.199 --> 00:06:32.319
<v Speaker 2>My favorite Beatles albums Kelp, and.

146
00:06:32.240 --> 00:06:36.759
<v Speaker 3>It's fantastic ride and whatever. No, this is a DeFi hack,

147
00:06:37.759 --> 00:06:40.480
<v Speaker 3>so you'll see the start with the headline for those

148
00:06:40.519 --> 00:06:42.240
<v Speaker 3>of you. All right, all right, let's start with the

149
00:06:42.240 --> 00:06:46.879
<v Speaker 3>headline Hackers target DeFi Protocol KELP DAO and a massive

150
00:06:46.920 --> 00:06:48.519
<v Speaker 3>three hundred million dollar xploit.

151
00:06:48.680 --> 00:06:50.000
<v Speaker 2>Yeah.

152
00:06:50.399 --> 00:06:53.639
<v Speaker 3>So here's typically for those of you who haven't gone

153
00:06:53.639 --> 00:06:57.800
<v Speaker 3>into blockchain contracts and how it all works. If you

154
00:06:57.959 --> 00:07:02.000
<v Speaker 3>make well, obviously, the the chain is indelible, right, it's immutable,

155
00:07:02.079 --> 00:07:04.639
<v Speaker 3>it can't be changed. So if I make a transaction,

156
00:07:04.759 --> 00:07:08.079
<v Speaker 3>it's read only right when it gets to the chain.

157
00:07:08.800 --> 00:07:11.160
<v Speaker 3>But when it gets approved, it has to be approved

158
00:07:11.279 --> 00:07:14.160
<v Speaker 3>typically by multiple notes, a majority of the notes, right,

159
00:07:14.680 --> 00:07:17.040
<v Speaker 3>and and each contract is different as to what that

160
00:07:17.079 --> 00:07:19.600
<v Speaker 3>looks like, but call it a majority, right, So one

161
00:07:19.759 --> 00:07:22.879
<v Speaker 3>percent plus one? Yeah, on this security this week, two

162
00:07:22.920 --> 00:07:23.680
<v Speaker 3>of us out of three.

163
00:07:23.759 --> 00:07:26.079
<v Speaker 1>Does that mean everybody who uses the blockchain has the

164
00:07:26.199 --> 00:07:27.800
<v Speaker 1>vote on all these transactions?

165
00:07:27.839 --> 00:07:30.439
<v Speaker 3>So? No, it, Like I said, that's where it depends.

166
00:07:30.600 --> 00:07:35.120
<v Speaker 3>Sometimes there's trusted nodes, because what you don't want is, hey,

167
00:07:35.199 --> 00:07:38.439
<v Speaker 3>let's put a million different random you know, vms up there,

168
00:07:39.160 --> 00:07:41.120
<v Speaker 3>and they get they get a vote, and now I

169
00:07:41.160 --> 00:07:42.160
<v Speaker 3>can tell them how to vote.

170
00:07:42.160 --> 00:07:44.800
<v Speaker 2>Supposedly the government tried to do that early on in blockchain.

171
00:07:45.000 --> 00:07:47.040
<v Speaker 3>Yeah, I know, well they I supposedly they tried to

172
00:07:47.040 --> 00:07:48.800
<v Speaker 3>do it with onion routing as well, And I don't

173
00:07:48.800 --> 00:07:50.120
<v Speaker 3>think any of that's actually real.

174
00:07:50.040 --> 00:07:52.160
<v Speaker 1>But only onion routing that happens in this house is

175
00:07:52.199 --> 00:07:55.600
<v Speaker 1>in my kitchen when I'm cooking, especially French onion soup.

176
00:07:56.079 --> 00:07:59.639
<v Speaker 3>Mm. So it's a majority.

177
00:07:59.759 --> 00:08:00.680
<v Speaker 1>What is onion routing?

178
00:08:00.759 --> 00:08:04.920
<v Speaker 3>Tore network so yeah, tour network so oh okay, yeah.

179
00:08:04.959 --> 00:08:09.120
<v Speaker 3>So anyways, so it's a majority, or it's trusted nodes

180
00:08:09.120 --> 00:08:11.079
<v Speaker 3>that vote, et cetera, et cetera, et cetera. And that's

181
00:08:11.079 --> 00:08:13.360
<v Speaker 3>why a lot of it is considered quite you know,

182
00:08:13.600 --> 00:08:17.319
<v Speaker 3>quite reasonable, quite trustworthy, right, where a majority of us

183
00:08:17.399 --> 00:08:19.319
<v Speaker 3>have to say, yes, this is an actual like I'm

184
00:08:19.360 --> 00:08:21.439
<v Speaker 3>Scarity this week, right, the story of us. Two of

185
00:08:21.519 --> 00:08:22.519
<v Speaker 3>us have to say, you know what, this is a

186
00:08:22.519 --> 00:08:24.720
<v Speaker 3>great story for us to put it on the show.

187
00:08:25.279 --> 00:08:27.560
<v Speaker 3>You guys can decide whether all of us are wrong

188
00:08:27.680 --> 00:08:32.240
<v Speaker 3>or not, but needless to say often, yeah, often often

189
00:08:32.440 --> 00:08:36.200
<v Speaker 3>that we're wrong, but needs say in this particular case,

190
00:08:36.320 --> 00:08:40.320
<v Speaker 3>that was all working fine. And in this particular, in

191
00:08:40.360 --> 00:08:45.720
<v Speaker 3>this particular issue, there was a restake of eth where

192
00:08:45.759 --> 00:08:48.600
<v Speaker 3>you could not only have it on this particular chain,

193
00:08:49.120 --> 00:08:51.399
<v Speaker 3>but it could go cross chain. So I could say, oh,

194
00:08:51.440 --> 00:08:53.720
<v Speaker 3>we're restaking it on chain A, but let's move it

195
00:08:53.720 --> 00:08:54.320
<v Speaker 3>over to change.

196
00:08:54.360 --> 00:08:59.240
<v Speaker 2>So those who aren't Crypto Bros. So Ether. Ethereum is

197
00:08:59.240 --> 00:09:01.840
<v Speaker 2>a very prominent coin and one of the things you

198
00:09:01.879 --> 00:09:03.919
<v Speaker 2>can do is stake it. So that means if you

199
00:09:03.960 --> 00:09:06.399
<v Speaker 2>have one hundred coins, congratulations, there's a lot of money.

200
00:09:06.960 --> 00:09:09.360
<v Speaker 2>You can take those coins and stake them and earn

201
00:09:09.440 --> 00:09:12.279
<v Speaker 2>a percentage back. And it is a way to to

202
00:09:12.679 --> 00:09:18.519
<v Speaker 2>have people put skin in the game and earn earn credits,

203
00:09:18.639 --> 00:09:22.480
<v Speaker 2>earn crypto without having to have their machine pollute and.

204
00:09:22.960 --> 00:09:24.480
<v Speaker 1>Like selling shares or something.

205
00:09:24.519 --> 00:09:29.159
<v Speaker 2>It's almost like getting interest on your investment. Yep, it's

206
00:09:29.200 --> 00:09:31.399
<v Speaker 2>more like that, I have a thousand eths, so I'll

207
00:09:31.440 --> 00:09:35.080
<v Speaker 2>get a couple of eighth a month, so interest if

208
00:09:35.080 --> 00:09:35.759
<v Speaker 2>I stake it.

209
00:09:35.799 --> 00:09:37.799
<v Speaker 1>Is there any risk in staking your.

210
00:09:37.600 --> 00:09:40.960
<v Speaker 2>Eth Yes, because when you stake it, you can't unstake it.

211
00:09:41.080 --> 00:09:43.759
<v Speaker 2>You can't sell it after unstaking it for like seven

212
00:09:43.840 --> 00:09:46.759
<v Speaker 2>days or something like that. So there's a cooling off period.

213
00:09:46.799 --> 00:09:49.159
<v Speaker 2>So when you stake it, it's not liquid. You can't

214
00:09:49.200 --> 00:09:50.480
<v Speaker 2>just unstake it and sell it.

215
00:09:50.600 --> 00:09:50.960
<v Speaker 1>I see.

216
00:09:51.080 --> 00:09:54.879
<v Speaker 3>So if Ethereum tanks in the next two days.

217
00:09:54.559 --> 00:09:55.120
<v Speaker 2>You're stuck.

218
00:09:55.480 --> 00:09:56.480
<v Speaker 1>You're stuck. Yeah.

219
00:09:56.799 --> 00:09:58.600
<v Speaker 2>I don't know what the period was, and I'm sure

220
00:09:58.639 --> 00:10:02.399
<v Speaker 2>it's changed. So this is them manipulating that staking and

221
00:10:02.440 --> 00:10:04.120
<v Speaker 2>restaking process.

222
00:10:04.480 --> 00:10:07.240
<v Speaker 3>Yeah, and in this particular case, they wanted to give

223
00:10:07.279 --> 00:10:10.120
<v Speaker 3>the users the ability to use multiple chains, right, not

224
00:10:10.240 --> 00:10:13.080
<v Speaker 3>just a single chain. So there's a message that goes

225
00:10:13.120 --> 00:10:13.639
<v Speaker 3>back and forth.

226
00:10:13.679 --> 00:10:16.840
<v Speaker 1>So wait a minute, what was they the hackers this?

227
00:10:17.080 --> 00:10:17.279
<v Speaker 2>Yeah?

228
00:10:17.360 --> 00:10:20.919
<v Speaker 3>No, no, no, no, this is your your KELP d O

229
00:10:21.240 --> 00:10:25.679
<v Speaker 3>and KELP is crypto exchange, a crypto exchange, okay, yeah,

230
00:10:25.759 --> 00:10:27.200
<v Speaker 3>call it a crypto exchange, all right.

231
00:10:27.240 --> 00:10:29.559
<v Speaker 1>So these are the guys that were doing the bad stuff.

232
00:10:29.840 --> 00:10:31.759
<v Speaker 3>No, these are the guys who are do the normal stuff.

233
00:10:31.799 --> 00:10:34.200
<v Speaker 3>This is like coinbase, this is like like whatever, Like

234
00:10:34.240 --> 00:10:37.399
<v Speaker 3>we have their system. Yeah, we have customers who log

235
00:10:37.480 --> 00:10:39.679
<v Speaker 3>in and they have ETH and they they stake it

236
00:10:39.720 --> 00:10:43.240
<v Speaker 3>and restake it and do whatever and whatever. Right, the

237
00:10:43.320 --> 00:10:47.480
<v Speaker 3>bad guys, what they did is the when you try

238
00:10:47.519 --> 00:10:49.960
<v Speaker 3>and stake that ETH and either move it around to

239
00:10:50.080 --> 00:10:54.879
<v Speaker 3>different chains. The bad guys actually targeted the messaging system

240
00:10:55.279 --> 00:10:59.159
<v Speaker 3>between chains. Remember the chain he for it to get

241
00:10:59.200 --> 00:11:01.080
<v Speaker 3>on the chain and for its to be a valid transaction,

242
00:11:01.240 --> 00:11:04.600
<v Speaker 3>a majority have to agree and be that chain is

243
00:11:04.679 --> 00:11:07.240
<v Speaker 3>read only. But when it moves from chain to chain,

244
00:11:07.720 --> 00:11:10.240
<v Speaker 3>in this particular case, there's a message that comes in

245
00:11:10.240 --> 00:11:11.720
<v Speaker 3>and says, hey, by the way, I'm going to take

246
00:11:11.759 --> 00:11:14.360
<v Speaker 3>this steak deep. Then we're going to keep it on

247
00:11:14.360 --> 00:11:15.879
<v Speaker 3>this chain over here for a little while, and then

248
00:11:15.879 --> 00:11:16.519
<v Speaker 3>we'll move it back.

249
00:11:16.720 --> 00:11:18.559
<v Speaker 2>Does that literally mean it's off the chain?

250
00:11:18.879 --> 00:11:22.039
<v Speaker 1>I remember the early days of cryptocurrency and some very

251
00:11:22.480 --> 00:11:24.919
<v Speaker 1>smart people that Patrick and I both know, and you

252
00:11:24.919 --> 00:11:28.240
<v Speaker 1>probably know who I'm talking about, said this is so great.

253
00:11:28.320 --> 00:11:32.480
<v Speaker 1>You can't hack it. It's like stone solid. And I

254
00:11:32.639 --> 00:11:36.879
<v Speaker 1>was like, oh, yeah, sure, based on something that's digital

255
00:11:37.440 --> 00:11:42.120
<v Speaker 1>that exists nowhere except in some computer somewhere. And you're

256
00:11:42.159 --> 00:11:44.720
<v Speaker 1>telling me that it's not hackable. Come on.

257
00:11:44.960 --> 00:11:49.360
<v Speaker 2>So it's like communism. It sounds great, it just doesn't

258
00:11:49.360 --> 00:11:53.759
<v Speaker 2>work in real life now. I mean, cryptos blockchain has

259
00:11:53.799 --> 00:11:56.519
<v Speaker 2>some real powerful things, but the problem is the implementation.

260
00:11:56.600 --> 00:11:59.919
<v Speaker 2>And as we'll see when Dwayne explains or we explain,

261
00:12:01.120 --> 00:12:04.240
<v Speaker 2>there was an implementation detail that was kind of a

262
00:12:04.279 --> 00:12:08.440
<v Speaker 2>convenience that was arranged that was exploited, correct Wing.

263
00:12:08.759 --> 00:12:12.279
<v Speaker 3>Yeah, absolutely, you're right, Patrick. So just so we're completely

264
00:12:12.320 --> 00:12:17.480
<v Speaker 3>clear on how this all works, the KELP DAO is

265
00:12:17.559 --> 00:12:21.879
<v Speaker 3>a restaking protocol for DI for DeFi, so that could

266
00:12:21.879 --> 00:12:29.759
<v Speaker 3>be used by any exchange. Right. The restaking ether is

267
00:12:29.960 --> 00:12:33.799
<v Speaker 3>KELP sort of what they call a liquid restaking token.

268
00:12:33.840 --> 00:12:35.600
<v Speaker 3>So it's a token that can move from chain to chain.

269
00:12:36.120 --> 00:12:41.080
<v Speaker 3>They're using a messaging protocol called Layer zero. Now Layer

270
00:12:41.240 --> 00:12:45.559
<v Speaker 3>zero is a cross chain messaging bridge protocol used by KELP.

271
00:12:45.919 --> 00:12:47.919
<v Speaker 3>So I don't want I didn't want to go into

272
00:12:47.960 --> 00:12:49.639
<v Speaker 3>too much detail, but now that we're asking a little

273
00:12:49.639 --> 00:12:52.039
<v Speaker 3>bit more questions, I wanted to make sure like everybody

274
00:12:52.120 --> 00:12:53.639
<v Speaker 3>kind of understands what's going on. So we have a

275
00:12:53.679 --> 00:12:59.120
<v Speaker 3>protocol that's handling a liquid token. That liquid token can

276
00:12:59.240 --> 00:13:01.960
<v Speaker 3>move from chain the chain based on a message that

277
00:13:02.000 --> 00:13:03.360
<v Speaker 3>comes in. Does that make sense?

278
00:13:03.399 --> 00:13:04.559
<v Speaker 1>And what makes it liquid?

279
00:13:05.279 --> 00:13:07.519
<v Speaker 3>So that what makes it liquid, my gas would be

280
00:13:07.639 --> 00:13:10.639
<v Speaker 3>is because it can actually move around and it's not indelible.

281
00:13:10.679 --> 00:13:11.919
<v Speaker 3>But I'd have to dig into that.

282
00:13:12.960 --> 00:13:13.840
<v Speaker 1>Maybe it can change.

283
00:13:13.879 --> 00:13:18.559
<v Speaker 3>So but more really on the hack. The hack is

284
00:13:18.639 --> 00:13:22.559
<v Speaker 3>at that layer zero, that cross chain messaging. Right, So

285
00:13:22.679 --> 00:13:26.120
<v Speaker 3>getting a token on the chain, getting a token approved

286
00:13:26.200 --> 00:13:30.799
<v Speaker 3>obviously majority and all that good stuff. But getting a

287
00:13:30.840 --> 00:13:34.600
<v Speaker 3>token to move from chain to chain only required one

288
00:13:34.679 --> 00:13:38.639
<v Speaker 3>verifyer and there so one person sent a message in saying, oh,

289
00:13:38.679 --> 00:13:42.600
<v Speaker 3>by the way, I have three hundred million dollars worth

290
00:13:42.679 --> 00:13:46.480
<v Speaker 3>of restake teth put it on the chain and I

291
00:13:46.600 --> 00:13:49.759
<v Speaker 3>verify it. All the chains were like, yeah, sure, that's cool, awesome,

292
00:13:49.840 --> 00:13:50.320
<v Speaker 3>let's do that.

293
00:13:50.399 --> 00:13:52.639
<v Speaker 2>Who verifies it? I do, Oh okay, thank you.

294
00:13:52.720 --> 00:13:55.320
<v Speaker 3>Yeah it's me I verified it or my buddy Patrick

295
00:13:55.440 --> 00:13:55.840
<v Speaker 3>verified it.

296
00:13:55.879 --> 00:13:57.799
<v Speaker 2>And I'm like, oh okay. That's the problem. That's the

297
00:13:57.840 --> 00:14:00.679
<v Speaker 2>part that's where they violated the i'm directive.

298
00:14:00.960 --> 00:14:02.679
<v Speaker 1>So they got there wasn't a man in the middle

299
00:14:02.720 --> 00:14:05.679
<v Speaker 1>kind of attack where they got into the exchange of messages.

300
00:14:06.039 --> 00:14:09.519
<v Speaker 3>Now you can issue these messages, like if you're participating

301
00:14:09.559 --> 00:14:11.759
<v Speaker 3>in the change, generally you have the ability to issue

302
00:14:11.759 --> 00:14:16.000
<v Speaker 3>some of these messages. There's even with ethereum and that

303
00:14:16.039 --> 00:14:18.679
<v Speaker 3>sort of stuff. If you're a node that can make

304
00:14:19.840 --> 00:14:22.879
<v Speaker 3>you know that yes, this is a valid transaction. Lots

305
00:14:22.879 --> 00:14:27.440
<v Speaker 3>of times when we were doing when you remember mining crypto,

306
00:14:27.720 --> 00:14:30.840
<v Speaker 3>sure right, there were different ways to mind crypto. One

307
00:14:30.919 --> 00:14:33.440
<v Speaker 3>was trying to solve a mathematical problem. There's another way

308
00:14:33.440 --> 00:14:36.120
<v Speaker 3>where you could have spent your CPU cycles, which was

309
00:14:36.159 --> 00:14:39.879
<v Speaker 3>I'm going to spend CPU cycles in verifying transactions, right

310
00:14:39.919 --> 00:14:42.159
<v Speaker 3>and become part of the network. So you could be

311
00:14:42.159 --> 00:14:45.200
<v Speaker 3>a node that verifies transactions. Yeah, and I listen, I've

312
00:14:45.200 --> 00:14:47.120
<v Speaker 3>done this in a while. But I know with eth,

313
00:14:47.200 --> 00:14:49.360
<v Speaker 3>for example, if you had I think it was thirty

314
00:14:49.399 --> 00:14:54.679
<v Speaker 3>thousand dollars worth eth a in a wallet, you could

315
00:14:54.679 --> 00:14:58.559
<v Speaker 3>then be a verification node. If you staked your ETH

316
00:14:58.679 --> 00:15:01.000
<v Speaker 3>in that node and said, yeah, I stake my thirty

317
00:15:01.000 --> 00:15:05.320
<v Speaker 3>grand to say, yeah, we're making these transactions appropriately, you'd

318
00:15:05.360 --> 00:15:08.159
<v Speaker 3>earn ETH on that, and you'd earn ETH for doing

319
00:15:08.159 --> 00:15:09.879
<v Speaker 3>the verifications and that sort of stuff. So it was

320
00:15:10.000 --> 00:15:12.759
<v Speaker 3>it was when mining really wasn't viable anymore, but you

321
00:15:12.759 --> 00:15:15.200
<v Speaker 3>could still verify transactions.

322
00:15:14.600 --> 00:15:16.519
<v Speaker 1>In the days of mining crypto are over.

323
00:15:16.960 --> 00:15:21.679
<v Speaker 3>Yeah, pretty much. I mean yeah for the normal household user, yes,

324
00:15:22.080 --> 00:15:25.320
<v Speaker 3>I would say that you just don't have the amount

325
00:15:25.360 --> 00:15:30.559
<v Speaker 3>of compute and electricity to really mine anything. And there

326
00:15:30.559 --> 00:15:32.559
<v Speaker 3>are still, i mean there are still some minable tokens,

327
00:15:32.840 --> 00:15:36.679
<v Speaker 3>but you're not getting much for them. Okay, So just

328
00:15:36.720 --> 00:15:39.200
<v Speaker 3>so we're all sort of clear on this process, KELP

329
00:15:39.279 --> 00:15:43.679
<v Speaker 3>dio is a DeFi restaking protocol, not an exchange the

330
00:15:43.799 --> 00:15:46.639
<v Speaker 3>stolen asset, the thing that they created out of thin air,

331
00:15:46.759 --> 00:15:48.840
<v Speaker 3>although is an all crypto kind of created out of

332
00:15:48.840 --> 00:15:53.200
<v Speaker 3>thin air. Is this restaked eth or rs eth KELP

333
00:15:53.399 --> 00:15:58.399
<v Speaker 3>liquid restaking tokens that's what that is. And the failure

334
00:15:58.440 --> 00:16:02.440
<v Speaker 3>here happened in layer zero with that cross cross bridge

335
00:16:02.639 --> 00:16:06.919
<v Speaker 3>messaging communications right where they just required one of one verifier,

336
00:16:06.919 --> 00:16:08.639
<v Speaker 3>which is a no good.

337
00:16:09.039 --> 00:16:11.720
<v Speaker 1>And so these hackers got away with or did they

338
00:16:11.720 --> 00:16:14.279
<v Speaker 1>get caught this three hundred million dollars.

339
00:16:14.279 --> 00:16:18.200
<v Speaker 3>For three hundred million dollars of eth that didn't actually exist.

340
00:16:19.200 --> 00:16:21.039
<v Speaker 3>I don't know that they actually got caught. It doesn't

341
00:16:21.039 --> 00:16:21.840
<v Speaker 3>say in the article here.

342
00:16:22.159 --> 00:16:22.360
<v Speaker 1>Hmm.

343
00:16:22.519 --> 00:16:27.480
<v Speaker 2>Okay, they basically didn't steal it from someone. They conjured it, yeah,

344
00:16:28.080 --> 00:16:30.120
<v Speaker 2>out of thin thinner air.

345
00:16:29.919 --> 00:16:31.759
<v Speaker 1>All right, But it's not like they cashed it in

346
00:16:31.799 --> 00:16:33.519
<v Speaker 1>and fled the country or it might.

347
00:16:33.399 --> 00:16:36.519
<v Speaker 3>Have Yeah, no, no, no, no, okay, they say they say

348
00:16:36.559 --> 00:16:38.559
<v Speaker 3>the attack has been attributed to North Korea as a

349
00:16:38.639 --> 00:16:43.519
<v Speaker 3>Lazarith group, but nobody's arresting them and it's not like right,

350
00:16:43.679 --> 00:16:46.559
<v Speaker 3>so all right, some of the money was actually well,

351
00:16:46.679 --> 00:16:48.159
<v Speaker 3>well here you go, some of the money was actually

352
00:16:48.240 --> 00:16:52.279
<v Speaker 3>seized and frozen about thirty thousand eighth of it roughly

353
00:16:52.480 --> 00:16:55.519
<v Speaker 3>seventy one million, so not as much as it was taking.

354
00:16:55.600 --> 00:16:58.159
<v Speaker 1>Okay, all right, this is a real quick one. I

355
00:16:58.200 --> 00:17:00.919
<v Speaker 1>brought this up on dot net rocks. It's a Windows

356
00:17:00.919 --> 00:17:05.039
<v Speaker 1>Defender Performance Tool. It's a Windows application also a dot

357
00:17:05.079 --> 00:17:10.359
<v Speaker 1>net application that monitors Windows Defender ETW events and visualizes

358
00:17:10.400 --> 00:17:14.519
<v Speaker 1>scan durations in real time using a bar chart. And

359
00:17:14.559 --> 00:17:17.599
<v Speaker 1>it can also visualize snapshots. And you know, I only

360
00:17:17.640 --> 00:17:20.599
<v Speaker 1>mention it because there's a lot of people now and

361
00:17:20.599 --> 00:17:22.480
<v Speaker 1>a lot of friends of ours that just you know,

362
00:17:22.519 --> 00:17:24.599
<v Speaker 1>they sit behind their Windows machine and they're like, gee,

363
00:17:24.640 --> 00:17:26.680
<v Speaker 1>I wonder if I'm getting hacked. You know, there's not

364
00:17:26.799 --> 00:17:30.000
<v Speaker 1>really there's Windows Defender pops up and says, you know,

365
00:17:30.000 --> 00:17:33.400
<v Speaker 1>if there's a threat or something like that. But generally

366
00:17:33.519 --> 00:17:36.920
<v Speaker 1>we don't look into the inner workings of our machines

367
00:17:37.039 --> 00:17:41.000
<v Speaker 1>and you know, what applications are running in what could

368
00:17:41.000 --> 00:17:43.119
<v Speaker 1>be a problem in all this kind of stuff. So

369
00:17:43.160 --> 00:17:45.240
<v Speaker 1>this is a good thing just to run. If you've

370
00:17:45.240 --> 00:17:47.680
<v Speaker 1>got some extra screen space, throw it.

371
00:17:47.720 --> 00:17:49.519
<v Speaker 2>Up there if you're interested in this kind of thing.

372
00:17:49.640 --> 00:17:51.440
<v Speaker 1>If you're interested in this kind of thing. So I

373
00:17:51.440 --> 00:17:53.039
<v Speaker 1>don't want to be labor it, but that's it. I

374
00:17:53.119 --> 00:17:54.240
<v Speaker 1>just I'll put a link to that.

375
00:17:54.279 --> 00:17:56.599
<v Speaker 3>Bo didn't you guys see who the two developers are.

376
00:17:56.519 --> 00:17:58.039
<v Speaker 2>On us probably interviewed them.

377
00:17:58.119 --> 00:18:02.039
<v Speaker 1>Oh, Alexander and Claude.

378
00:18:02.799 --> 00:18:04.279
<v Speaker 3>I think you're going to start to see a lot

379
00:18:04.319 --> 00:18:07.119
<v Speaker 3>more of this. I've seen a lot of githubs attributed

380
00:18:07.440 --> 00:18:10.799
<v Speaker 3>to claud Code as the only developer now right, we're

381
00:18:10.839 --> 00:18:13.400
<v Speaker 3>the only developer doing any pushing any whatever is Claude

382
00:18:13.480 --> 00:18:15.519
<v Speaker 3>to where I think we'll see a lot of these

383
00:18:15.519 --> 00:18:17.440
<v Speaker 3>little tools. I mean, shoot, a lot of us have

384
00:18:17.480 --> 00:18:19.440
<v Speaker 3>had these ideas, right right, Hey, wouldn't it be cool

385
00:18:19.440 --> 00:18:20.880
<v Speaker 3>if I had this little thing that did this thing?

386
00:18:21.279 --> 00:18:21.680
<v Speaker 2>Right now?

387
00:18:21.720 --> 00:18:22.839
<v Speaker 3>You can get cloud code to create it?

388
00:18:22.839 --> 00:18:23.559
<v Speaker 1>So cool.

389
00:18:23.599 --> 00:18:24.279
<v Speaker 3>He's a good example.

390
00:18:24.240 --> 00:18:26.279
<v Speaker 2>You know that guy who did that thing at that

391
00:18:26.319 --> 00:18:26.960
<v Speaker 2>place at the time.

392
00:18:27.039 --> 00:18:28.640
<v Speaker 1>I remember, all right, who wants to take the CEA

393
00:18:28.680 --> 00:18:32.440
<v Speaker 1>panel story? Oh, c panel start with the headline please.

394
00:18:32.240 --> 00:18:34.960
<v Speaker 3>C panel for those of you know you know what

395
00:18:35.079 --> 00:18:39.279
<v Speaker 3>sea panel is critical C panel and WHM bug exploited

396
00:18:39.319 --> 00:18:43.799
<v Speaker 3>as a zero day POC now available. POC is now available.

397
00:18:44.079 --> 00:18:46.920
<v Speaker 3>I mean I should say it in that proof of concepts.

398
00:18:46.920 --> 00:18:50.000
<v Speaker 1>In other words, there's a there's a tool that somebody

399
00:18:50.039 --> 00:18:52.799
<v Speaker 1>has written to do to do the exploit, to do

400
00:18:52.839 --> 00:18:54.200
<v Speaker 1>the deed, to do the deed.

401
00:18:54.119 --> 00:18:54.960
<v Speaker 2>To do the deed.

402
00:18:55.799 --> 00:18:58.200
<v Speaker 3>So for those of you who have hosted websites, which

403
00:18:58.640 --> 00:19:01.559
<v Speaker 3>first off, for those of you hosting your website, shame, Yeah,

404
00:19:01.839 --> 00:19:03.200
<v Speaker 3>you probably shouldn't be.

405
00:19:03.240 --> 00:19:04.759
<v Speaker 1>Either that or do you want a job?

406
00:19:05.680 --> 00:19:09.799
<v Speaker 3>Yeah, exactly, it's one of the two. So usually if

407
00:19:09.799 --> 00:19:13.480
<v Speaker 3>you host your own, if you're in a hosted website situation,

408
00:19:13.519 --> 00:19:15.839
<v Speaker 3>whether it's go Daddy or whatever, lots of times you'll

409
00:19:15.839 --> 00:19:17.359
<v Speaker 3>have a way to manage your website, right, So if

410
00:19:17.359 --> 00:19:19.039
<v Speaker 3>it's word Press, you're gonna log in a word Press,

411
00:19:19.039 --> 00:19:21.799
<v Speaker 3>you're gonna do whatever. But if it's another type of site,

412
00:19:21.839 --> 00:19:27.359
<v Speaker 3>you're gonna typically have what's called Seapanel. Sepanel is probably,

413
00:19:27.359 --> 00:19:29.119
<v Speaker 3>if I had to guess one of the most, if

414
00:19:29.160 --> 00:19:34.319
<v Speaker 3>not the most prominent panel management system.

415
00:19:33.799 --> 00:19:35.839
<v Speaker 2>For your website, be surprised if it isn't.

416
00:19:36.119 --> 00:19:38.839
<v Speaker 3>Yeah, I mean I we run into Seapanel a lot

417
00:19:39.759 --> 00:19:42.200
<v Speaker 3>when taking the customer sites and that sort of stuff.

418
00:19:43.359 --> 00:19:48.119
<v Speaker 3>So yeah, not entirely sure. But in this particular case,

419
00:19:48.680 --> 00:19:52.240
<v Speaker 3>there is a critical CVE twenty twenty six forty one

420
00:19:52.519 --> 00:19:58.319
<v Speaker 3>nine four zero, which is an authentication bypass vulnerability in Seapanel.

421
00:19:58.480 --> 00:19:58.920
<v Speaker 1>Not good.

422
00:19:59.039 --> 00:20:04.480
<v Speaker 3>So if you are running Seapanel eleven eleven or eleven one, ten,

423
00:20:04.519 --> 00:20:06.960
<v Speaker 3>eleven one, eighteen one, twenty six, one thirty two, one

424
00:20:07.039 --> 00:20:10.480
<v Speaker 3>thirty four, one thirty six, one thirty six one, you

425
00:20:10.759 --> 00:20:15.839
<v Speaker 3>have to go patch immediately because there are right now

426
00:20:16.160 --> 00:20:21.559
<v Speaker 3>one point six million sea panels that come up on

427
00:20:21.680 --> 00:20:26.640
<v Speaker 3>Showdown instantly if you look for them, and a fair

428
00:20:26.759 --> 00:20:30.160
<v Speaker 3>chunk of them are not patch right now. So if

429
00:20:30.200 --> 00:20:33.119
<v Speaker 3>you have if you have a sea panel, you want

430
00:20:33.119 --> 00:20:35.119
<v Speaker 3>to go patch this. This is probably one of one.

431
00:20:35.200 --> 00:20:38.799
<v Speaker 3>I'd say the biggest thing until we found a bigger thing,

432
00:20:38.839 --> 00:20:40.759
<v Speaker 3>but this is probably the number two big thing in

433
00:20:40.799 --> 00:20:41.440
<v Speaker 3>the last two.

434
00:20:41.319 --> 00:20:44.839
<v Speaker 2>Weeks and not a good big thing, right, I.

435
00:20:44.759 --> 00:20:46.119
<v Speaker 3>Mean, it depends on who you are, and I think

436
00:20:46.160 --> 00:20:46.880
<v Speaker 3>it's a good big thing.

437
00:20:47.359 --> 00:20:49.759
<v Speaker 2>Showed on just for those who you know don't remember

438
00:20:49.920 --> 00:20:53.000
<v Speaker 2>is a tool that we can use to find is

439
00:20:53.039 --> 00:20:55.119
<v Speaker 2>a first degree black belt it is.

440
00:20:55.200 --> 00:20:55.680
<v Speaker 1>It is.

441
00:20:57.880 --> 00:21:01.720
<v Speaker 2>Also a way to find exploits that are available.

442
00:21:01.799 --> 00:21:03.440
<v Speaker 3>Yeah, I mean think of it as think of it

443
00:21:03.519 --> 00:21:08.279
<v Speaker 3>like the Google for for devices is really what it is.

444
00:21:08.720 --> 00:21:12.599
<v Speaker 3>So it's like I want to find every webcam that's

445
00:21:12.680 --> 00:21:13.680
<v Speaker 3>open on the Internet.

446
00:21:14.119 --> 00:21:15.759
<v Speaker 1>Yeah, of course you wouldn't do that, but.

447
00:21:15.799 --> 00:21:18.799
<v Speaker 3>You go to Showdown showed on would Showdown would show

448
00:21:18.839 --> 00:21:19.720
<v Speaker 3>you those before.

449
00:21:19.599 --> 00:21:23.799
<v Speaker 2>That we'd use Google dorks basically special strings yep.

450
00:21:23.920 --> 00:21:25.759
<v Speaker 3>And I think I think that's if I were I

451
00:21:25.759 --> 00:21:28.039
<v Speaker 3>wouldn't be surprised if that's not where showdown came from.

452
00:21:28.079 --> 00:21:30.160
<v Speaker 3>Where they're like, gosh, we really hate dork and this

453
00:21:30.200 --> 00:21:32.480
<v Speaker 3>stuff all the time. Let's just create a database of it.

454
00:21:33.480 --> 00:21:35.920
<v Speaker 3>But like, if you want to go take for example,

455
00:21:36.559 --> 00:21:39.000
<v Speaker 3>how many computers are open on three three eight nine on.

456
00:21:39.000 --> 00:21:40.799
<v Speaker 1>The internet, right, is that the webcam port?

457
00:21:41.079 --> 00:21:41.160
<v Speaker 2>No?

458
00:21:41.200 --> 00:21:43.119
<v Speaker 3>Three three eight nine the remote desktop port?

459
00:21:43.319 --> 00:21:43.519
<v Speaker 1>Oh?

460
00:21:43.640 --> 00:21:46.319
<v Speaker 3>Yeah, okay, So how many computers are open on three

461
00:21:46.400 --> 00:21:52.559
<v Speaker 3>three eight nine and have in the picture of the

462
00:21:52.599 --> 00:21:55.720
<v Speaker 3>remote desktop the username administrator.

463
00:21:57.720 --> 00:22:02.119
<v Speaker 2>It will can't be it can't be more than fifteen million, yeah.

464
00:22:01.640 --> 00:22:04.200
<v Speaker 3>But it will do that. It will actually go through

465
00:22:04.359 --> 00:22:10.240
<v Speaker 3>take a screenshot of the remote desktop welcome window, and

466
00:22:10.319 --> 00:22:14.519
<v Speaker 3>then it will ocr all of the users that show

467
00:22:14.599 --> 00:22:17.519
<v Speaker 3>up there, so that you can search for a username. Wow,

468
00:22:17.559 --> 00:22:21.200
<v Speaker 3>which is awesome if you're looking to target a particular

469
00:22:21.200 --> 00:22:22.559
<v Speaker 3>system or user account.

470
00:22:22.359 --> 00:22:23.160
<v Speaker 1>Or if you're a hacker.

471
00:22:23.240 --> 00:22:26.319
<v Speaker 3>Yeah, or you're looking for Windows XP right yeah, type

472
00:22:26.359 --> 00:22:29.160
<v Speaker 3>in Windows XP and remote desktop and you'll find a

473
00:22:29.160 --> 00:22:32.200
<v Speaker 3>bunch of boxes connected to the internet that are Windows XP.

474
00:22:32.839 --> 00:22:34.079
<v Speaker 2>Okay or their honeypop?

475
00:22:34.720 --> 00:22:36.799
<v Speaker 1>Are we done with this? Were moving on to canvas. Yeah,

476
00:22:36.880 --> 00:22:40.359
<v Speaker 1>we are all go patch, go patch, Yeah, Canvas breach

477
00:22:40.480 --> 00:22:45.000
<v Speaker 1>may put two hundred and seventy five million users. That's

478
00:22:45.160 --> 00:22:47.640
<v Speaker 1>nine thousand schools at risk.

479
00:22:47.839 --> 00:22:52.000
<v Speaker 2>So this is not Canva, the AI tool. This is Canvas,

480
00:22:52.079 --> 00:22:52.880
<v Speaker 2>a school tool.

481
00:22:53.279 --> 00:22:56.240
<v Speaker 3>Yes, yeah, and I've seen some people use Canva in school.

482
00:22:56.319 --> 00:22:58.640
<v Speaker 3>But yeah, you're absolutely right, and now it's.

483
00:22:58.480 --> 00:23:02.519
<v Speaker 1>A tool too cool for school, is too cool? Yeah,

484
00:23:02.640 --> 00:23:03.319
<v Speaker 1>So what happened?

485
00:23:03.440 --> 00:23:04.279
<v Speaker 2>It's what happened here?

486
00:23:05.000 --> 00:23:08.279
<v Speaker 3>Good question? Right right now, there's an active investigation going on,

487
00:23:08.319 --> 00:23:11.440
<v Speaker 3>so they're not sure what the exploit was to garner

488
00:23:11.519 --> 00:23:17.920
<v Speaker 3>the information. However, right now, the shiny hunters who hacking

489
00:23:17.920 --> 00:23:20.200
<v Speaker 3>grew up on the Internet says that they have stolen

490
00:23:20.440 --> 00:23:24.359
<v Speaker 3>information that can contain some pi I some personally identify

491
00:23:24.559 --> 00:23:30.400
<v Speaker 3>unidentifiable information about students and teachers, and that that may

492
00:23:30.440 --> 00:23:33.400
<v Speaker 3>include your name, that may include your email address, et cetera.

493
00:23:33.599 --> 00:23:37.880
<v Speaker 3>So until we find out my gas is not password

494
00:23:38.519 --> 00:23:41.279
<v Speaker 3>right to log in, I know a lot of people

495
00:23:41.319 --> 00:23:46.880
<v Speaker 3>that are using Canvas are using they're using single sign

496
00:23:46.960 --> 00:23:50.880
<v Speaker 3>on anyways, right, they're using an O off to Google

497
00:23:51.319 --> 00:23:54.519
<v Speaker 3>or Microsoft or whatever. Right, a lot of schools are

498
00:23:54.640 --> 00:23:58.799
<v Speaker 3>using either you know, Google Workspaces or they're using Microsoft Office.

499
00:23:58.920 --> 00:24:02.440
<v Speaker 3>So it's very fast the author indication layer. But in

500
00:24:02.480 --> 00:24:06.440
<v Speaker 3>this particular case, you may start receiving a lot more

501
00:24:06.759 --> 00:24:08.720
<v Speaker 3>targeted phishing email.

502
00:24:08.480 --> 00:24:09.839
<v Speaker 2>Because they have information on you.

503
00:24:09.880 --> 00:24:11.920
<v Speaker 3>Because they have the information on you, they know you're

504
00:24:12.000 --> 00:24:15.480
<v Speaker 3>using Canvas, they know they what you're using it for,

505
00:24:16.319 --> 00:24:18.400
<v Speaker 3>they know your name, they know your email address, so

506
00:24:18.440 --> 00:24:21.200
<v Speaker 3>they'll probably start sending you very targeted emails. I mean,

507
00:24:21.480 --> 00:24:24.119
<v Speaker 3>if you receive an email from Canvas, what it looks

508
00:24:24.200 --> 00:24:26.799
<v Speaker 3>like it's from Canvas saying hey, we apologize there was

509
00:24:26.799 --> 00:24:29.119
<v Speaker 3>a breach blah blah blah. Here you know, click here

510
00:24:29.200 --> 00:24:32.519
<v Speaker 3>to enter your you know, login information so that you

511
00:24:32.599 --> 00:24:34.720
<v Speaker 3>can we can make sure you're secure or whatever.

512
00:24:34.759 --> 00:24:38.279
<v Speaker 1>It's probably the hackers, not can right, Okay, Yeah, so

513
00:24:38.400 --> 00:24:42.400
<v Speaker 1>this is just a cautionary tale. It's a story. There's

514
00:24:42.400 --> 00:24:45.640
<v Speaker 1>nothing anybody who's listening can do anything about. It just

515
00:24:46.039 --> 00:24:50.640
<v Speaker 1>happened where and understand, right, mm and it was probably

516
00:24:50.799 --> 00:24:56.839
<v Speaker 1>some ill configured security obviously. All right, let's move on

517
00:24:57.039 --> 00:24:59.119
<v Speaker 1>and then we'll take a break. This is just kind

518
00:24:59.119 --> 00:25:03.880
<v Speaker 1>of a public servicennouncement. It's nothing new, but the browser

519
00:25:04.039 --> 00:25:07.839
<v Speaker 1>is breaking your DLP. How data slips past modern controls.

520
00:25:08.119 --> 00:25:10.640
<v Speaker 1>It's a story and bleeping computer, but it underscore is

521
00:25:10.680 --> 00:25:12.720
<v Speaker 1>something that we've been talking about on the show for

522
00:25:12.799 --> 00:25:16.519
<v Speaker 1>quite a while now, which is you think you're safe

523
00:25:16.559 --> 00:25:21.640
<v Speaker 1>because you have you know, various forms of security and

524
00:25:21.680 --> 00:25:24.200
<v Speaker 1>in your browser and all that stuff. But more and

525
00:25:24.279 --> 00:25:28.640
<v Speaker 1>more things are falling through the cracks and we are

526
00:25:28.640 --> 00:25:31.200
<v Speaker 1>becoming more and more vulnerable every day.

527
00:25:31.640 --> 00:25:35.599
<v Speaker 2>And this boils down to people, and we all know this.

528
00:25:35.920 --> 00:25:38.880
<v Speaker 2>People are entering things more and more into portals and

529
00:25:38.920 --> 00:25:42.559
<v Speaker 2>browsers and the DLP systems are they're not triggering. They

530
00:25:43.200 --> 00:25:46.440
<v Speaker 2>they don't see that. So if you take critical information

531
00:25:46.559 --> 00:25:49.359
<v Speaker 2>and paste it into chat, GPT through a web interface,

532
00:25:49.440 --> 00:25:50.400
<v Speaker 2>it's not going to get caught.

533
00:25:50.680 --> 00:25:52.640
<v Speaker 1>So let's just define DLP. What is that.

534
00:25:52.960 --> 00:25:55.720
<v Speaker 3>So DLP is data loss prevention. So there's a lot

535
00:25:55.720 --> 00:25:58.759
<v Speaker 3>of DLP systems out there that will watch all your

536
00:25:58.759 --> 00:26:02.000
<v Speaker 3>web traffic. Like you know, you may work for a bank, right,

537
00:26:02.039 --> 00:26:04.480
<v Speaker 3>and one of the things DLP will look for is

538
00:26:05.640 --> 00:26:07.799
<v Speaker 3>it'll use rej axes and all the other good stuff

539
00:26:07.799 --> 00:26:09.839
<v Speaker 3>to look for credit cards, look for account numbers, look

540
00:26:09.880 --> 00:26:12.519
<v Speaker 3>for customer names. Right, So if I'm sending an email

541
00:26:12.559 --> 00:26:15.680
<v Speaker 3>with a credit card number, it will flag and say nope,

542
00:26:15.720 --> 00:26:18.839
<v Speaker 3>you can't send credit cards, you know, over email. So

543
00:26:19.000 --> 00:26:22.440
<v Speaker 3>make sure that you're not putting data you shouldn't be

544
00:26:22.519 --> 00:26:24.480
<v Speaker 3>putting in places you shouldn't be putting it.

545
00:26:25.000 --> 00:26:27.279
<v Speaker 1>One of the things that bugs me about this is

546
00:26:27.319 --> 00:26:31.359
<v Speaker 1>that the clipboard turns out to be a huge vulnerability

547
00:26:31.440 --> 00:26:34.920
<v Speaker 1>and the you know, the tools that we use, the

548
00:26:34.960 --> 00:26:39.480
<v Speaker 1>browser tools for password managers, they all work with a clipboard.

549
00:26:39.680 --> 00:26:40.480
<v Speaker 1>I mean you find your P.

550
00:26:40.680 --> 00:26:44.279
<v Speaker 2>I think clipboard is something that DLP typically running on

551
00:26:44.319 --> 00:26:48.160
<v Speaker 2>the local workstation endpoint can check. It's when like you

552
00:26:48.319 --> 00:26:50.839
<v Speaker 2>enter it into a page and then it's encrypted as

553
00:26:50.880 --> 00:26:52.119
<v Speaker 2>a as a load and sent.

554
00:26:52.559 --> 00:26:57.400
<v Speaker 1>This article says most traditional DLP solutions can inspect or control.

555
00:26:57.559 --> 00:27:01.039
<v Speaker 2>Really, Oh, that's an assumption. I might be wrong about

556
00:27:01.079 --> 00:27:02.519
<v Speaker 2>that then, yeah.

557
00:27:02.319 --> 00:27:05.440
<v Speaker 1>Right, anyway, so just be careful about that. And you know,

558
00:27:05.640 --> 00:27:09.279
<v Speaker 1>we used to use, I used to use a tool

559
00:27:09.440 --> 00:27:15.240
<v Speaker 1>that would remove it from the clipboard after you've pasted it, right.

560
00:27:15.559 --> 00:27:15.920
<v Speaker 2>Yeah.

561
00:27:16.000 --> 00:27:19.480
<v Speaker 1>I don't think that bitwarten does that, but that would

562
00:27:19.480 --> 00:27:21.920
<v Speaker 1>be a really great feature to turn on. You know

563
00:27:22.400 --> 00:27:27.480
<v Speaker 1>if if at bitwarten, right, copy this to the clipboard,

564
00:27:27.880 --> 00:27:31.039
<v Speaker 1>monitor the clipboard after I pasted it, remove it from

565
00:27:31.039 --> 00:27:32.160
<v Speaker 1>the clipboard, right.

566
00:27:32.279 --> 00:27:34.759
<v Speaker 2>You'd think that they do that. You'd think if you

567
00:27:34.839 --> 00:27:37.640
<v Speaker 2>hit the pill button and then go past somewhere.

568
00:27:37.960 --> 00:27:40.839
<v Speaker 1>Yeah, but we know listeners know not to use the

569
00:27:40.839 --> 00:27:44.880
<v Speaker 1>pill button because that has been exploited, yes, big time.

570
00:27:45.240 --> 00:27:47.400
<v Speaker 2>So yeah, we don't do that with the one in Bitwarden,

571
00:27:47.440 --> 00:27:48.160
<v Speaker 2>the one in the browser.

572
00:27:48.200 --> 00:27:49.279
<v Speaker 1>You mean in the browser.

573
00:27:49.359 --> 00:27:50.839
<v Speaker 2>Yeah, we're talking about the one in Bitwarten.

574
00:27:51.079 --> 00:27:53.880
<v Speaker 1>Well, Bitwarten has a browser plug in that tempts you

575
00:27:53.920 --> 00:27:55.880
<v Speaker 1>to use the fill button, and hey, you want me

576
00:27:55.920 --> 00:27:58.160
<v Speaker 1>to fill this in? And I say no, because we

577
00:27:58.279 --> 00:27:59.599
<v Speaker 1>know that that has been exploited.

578
00:27:59.720 --> 00:28:04.200
<v Speaker 3>Now I will say what bitboardon does have. It will

579
00:28:04.279 --> 00:28:08.799
<v Speaker 3>clear the clipboard automatically for you if you want, but

580
00:28:08.880 --> 00:28:10.839
<v Speaker 3>it doesn't do it after pace. It does it at

581
00:28:10.839 --> 00:28:12.799
<v Speaker 3>a scheduled time, okay, and you have to turn it on,

582
00:28:13.000 --> 00:28:14.400
<v Speaker 3>so it's like I want to say, it's like ten

583
00:28:14.480 --> 00:28:18.240
<v Speaker 3>or thirty seconds, so won't sit there forever. But and

584
00:28:18.279 --> 00:28:20.480
<v Speaker 3>you do have it is something that's done nothing. It's

585
00:28:20.519 --> 00:28:21.279
<v Speaker 3>not by default.

586
00:28:21.319 --> 00:28:24.519
<v Speaker 1>So yeah, interesting. Yeah, I'm going to check that out.

587
00:28:24.839 --> 00:28:27.440
<v Speaker 1>I'm going to check that out. That's definitely a good feature.

588
00:28:28.400 --> 00:28:29.799
<v Speaker 1>All right, I guess this is a good time to

589
00:28:29.799 --> 00:28:32.119
<v Speaker 1>take a break. Like we said, this is an extra

590
00:28:32.200 --> 00:28:35.200
<v Speaker 1>long show and we'll be right back after these very

591
00:28:35.200 --> 00:28:36.000
<v Speaker 1>important messages.

592
00:28:36.240 --> 00:28:38.960
<v Speaker 3>Stand up, stretch your legs, walk around. Unless you're driving,

593
00:28:39.240 --> 00:28:42.960
<v Speaker 3>then don't run. Get out of your car and run

594
00:28:43.000 --> 00:28:45.559
<v Speaker 3>alongside it. Jump back in.

595
00:28:49.519 --> 00:28:53.400
<v Speaker 1>And we're back. It's security this week. I'm Carl, that's Dwayne,

596
00:28:53.799 --> 00:28:56.799
<v Speaker 1>and that's Patrick. And Patrick. By the way, I don't

597
00:28:56.799 --> 00:28:58.599
<v Speaker 1>know if we should say this, but we probably should.

598
00:28:59.079 --> 00:29:03.680
<v Speaker 1>He is the possession of several little baby bunny rabbits

599
00:29:03.799 --> 00:29:07.559
<v Speaker 1>five five bunny rabbit. Five minutes before the show started,

600
00:29:08.200 --> 00:29:10.559
<v Speaker 1>he and his wife found out in the yard, stranded,

601
00:29:11.400 --> 00:29:13.559
<v Speaker 1>and there they appear to be a couple of weeks old.

602
00:29:14.200 --> 00:29:15.240
<v Speaker 1>Here's a button.

603
00:29:15.640 --> 00:29:18.079
<v Speaker 2>And and out in the out, in the middle of out,

604
00:29:18.079 --> 00:29:19.559
<v Speaker 2>in the middle of everything. They would have been dead

605
00:29:19.640 --> 00:29:20.079
<v Speaker 2>by now.

606
00:29:20.599 --> 00:29:22.359
<v Speaker 3>Yeah, oh yeah. The hawks would have picked him up,

607
00:29:22.359 --> 00:29:22.920
<v Speaker 3>that's for sure.

608
00:29:23.039 --> 00:29:24.960
<v Speaker 2>So we put him in a box. We're gonna as

609
00:29:24.960 --> 00:29:26.599
<v Speaker 2>soon as the show's over, I'm going to take them

610
00:29:26.599 --> 00:29:28.240
<v Speaker 2>to a refuge and they're going to take good care

611
00:29:28.240 --> 00:29:28.559
<v Speaker 2>of them.

612
00:29:28.640 --> 00:29:29.680
<v Speaker 1>Are they sleeping soundly?

613
00:29:30.119 --> 00:29:32.200
<v Speaker 2>I think so. Give the box a shake.

614
00:29:32.359 --> 00:29:38.799
<v Speaker 1>Yeah, that's terrible, but it's funny that, you know, he's

615
00:29:38.920 --> 00:29:41.519
<v Speaker 1>very protective of his little bunnies. He is, he is.

616
00:29:41.559 --> 00:29:43.440
<v Speaker 3>He's only had them for a couple of minutes.

617
00:29:43.160 --> 00:29:45.920
<v Speaker 2>But for very protective of my wards in all cases.

618
00:29:46.279 --> 00:29:49.920
<v Speaker 1>Uh huh yeah, all right, so, uh, let's this is

619
00:29:49.960 --> 00:29:53.440
<v Speaker 1>a good one. Guy finds Google Chrome is quietly installing

620
00:29:53.839 --> 00:29:57.799
<v Speaker 1>a four gigabyte AI model on our devices. Yikes. I

621
00:29:57.799 --> 00:30:00.359
<v Speaker 1>guess it's not so bad, but it just sounds bad.

622
00:30:00.799 --> 00:30:02.920
<v Speaker 3>Yeah, I mean it sounds nefarious where you're like, oh

623
00:30:02.960 --> 00:30:06.680
<v Speaker 3>my god for gig and it's interesting. So it says

624
00:30:06.759 --> 00:30:12.319
<v Speaker 3>installing a four gig AI model called Gemini Nano. The

625
00:30:12.400 --> 00:30:15.559
<v Speaker 3>key files there, the large file there is called weights

626
00:30:15.599 --> 00:30:18.920
<v Speaker 3>dot bin. So for anybody who's ever done like computer

627
00:30:19.039 --> 00:30:21.680
<v Speaker 3>vision or trains your own models and that sort of stuff,

628
00:30:21.720 --> 00:30:24.240
<v Speaker 3>the weight's bin is a I mean, it's honestly, it's

629
00:30:24.279 --> 00:30:28.519
<v Speaker 3>where all the post training is. It's how it identifies things. Right,

630
00:30:29.839 --> 00:30:33.759
<v Speaker 3>So that bin file probably you know, has the weights

631
00:30:33.799 --> 00:30:37.480
<v Speaker 3>for different words or whatever it's doing searching on, right,

632
00:30:37.519 --> 00:30:39.680
<v Speaker 3>so I can identify and summarize pages.

633
00:30:39.720 --> 00:30:41.640
<v Speaker 2>The cash an ai cash if you will.

634
00:30:42.039 --> 00:30:46.279
<v Speaker 3>Yeah, yeah, it's probably a good way to look at it. Yeah, absolutely.

635
00:30:46.680 --> 00:30:51.960
<v Speaker 3>So they're generally large, right, and they're generally highly processed

636
00:30:52.039 --> 00:30:55.720
<v Speaker 3>data that's going to be used statistically in some way

637
00:30:55.759 --> 00:30:58.160
<v Speaker 3>to help with you know, analyzing the page so that

638
00:30:58.200 --> 00:31:00.240
<v Speaker 3>you can click on that little Gemini button in home

639
00:31:00.319 --> 00:31:02.839
<v Speaker 3>and go, so tell me about the story, right, or

640
00:31:02.880 --> 00:31:04.839
<v Speaker 3>tell me about this article or whatever it may be.

641
00:31:05.359 --> 00:31:07.799
<v Speaker 3>Then it goes through and does its whole statistical thing.

642
00:31:08.440 --> 00:31:11.519
<v Speaker 3>What's nice and not so nice? Right? What's nice is

643
00:31:11.559 --> 00:31:13.359
<v Speaker 3>if they're pushing down this weight to bin, that means

644
00:31:13.359 --> 00:31:16.400
<v Speaker 3>a lot of that processing is being done locally potentially,

645
00:31:17.640 --> 00:31:19.880
<v Speaker 3>so you're not sort of going all the way out

646
00:31:19.880 --> 00:31:23.200
<v Speaker 3>to the web and calling some form of you know, chat,

647
00:31:23.279 --> 00:31:26.400
<v Speaker 3>GBT API with every page that you're browsing, which you

648
00:31:26.440 --> 00:31:29.720
<v Speaker 3>probably don't want to do so from a security standpoint,

649
00:31:29.759 --> 00:31:32.000
<v Speaker 3>I almost like this where I'd like to do the

650
00:31:32.039 --> 00:31:35.720
<v Speaker 3>local processing. I'd like to have the model local and

651
00:31:35.799 --> 00:31:37.920
<v Speaker 3>have it call out to the internet as little as bustable.

652
00:31:38.000 --> 00:31:40.119
<v Speaker 3>I mean, this is coming from a guy who's you know,

653
00:31:40.160 --> 00:31:43.039
<v Speaker 3>I'm running my own AI cluster in the house because

654
00:31:43.039 --> 00:31:46.240
<v Speaker 3>I don't want it, you know, going out and going anywhere. Sure,

655
00:31:47.640 --> 00:31:50.680
<v Speaker 3>but that being said, you usually want to allow the

656
00:31:50.759 --> 00:31:54.960
<v Speaker 3>users to opt in, not just pre push it. So

657
00:31:55.079 --> 00:31:56.799
<v Speaker 3>I agree in this. You know, it's funny in this

658
00:31:56.880 --> 00:31:58.839
<v Speaker 3>article they make the case of like, oh my god,

659
00:31:58.839 --> 00:32:02.359
<v Speaker 3>can you imagine how much CO two this burns pushing

660
00:32:02.400 --> 00:32:05.359
<v Speaker 3>this down to every client. You know, I would counter

661
00:32:05.480 --> 00:32:09.680
<v Speaker 3>that with anytime, you know, call of duty comes out

662
00:32:10.440 --> 00:32:13.839
<v Speaker 3>and the two and a half petabytes are required.

663
00:32:14.000 --> 00:32:16.480
<v Speaker 2>Or I would I would redirect it that person still

664
00:32:16.519 --> 00:32:18.160
<v Speaker 2>eating me, still eating me?

665
00:32:18.359 --> 00:32:20.680
<v Speaker 3>Yeah, yeah, So I you know, I don't know that

666
00:32:20.680 --> 00:32:22.680
<v Speaker 3>I would use CO two is the argument here, But

667
00:32:23.599 --> 00:32:25.839
<v Speaker 3>I you know, I would say, yeah, I get it.

668
00:32:25.920 --> 00:32:28.440
<v Speaker 3>I mean, it's a good chunk of storage being used.

669
00:32:28.720 --> 00:32:33.720
<v Speaker 3>But the alternative is take every page eyebrows and when

670
00:32:33.720 --> 00:32:36.279
<v Speaker 3>I hit that little Gemini icon, it goes out to

671
00:32:36.319 --> 00:32:40.319
<v Speaker 3>the Internet, you know, in full fidelity and gets processed

672
00:32:40.359 --> 00:32:42.079
<v Speaker 3>out there. And I don't know that I want that

673
00:32:42.119 --> 00:32:44.599
<v Speaker 3>to happen either. So hopefully, because these weights are being

674
00:32:44.640 --> 00:32:47.359
<v Speaker 3>pushed down, a lot of that processing is being done locally.

675
00:32:47.400 --> 00:32:49.519
<v Speaker 1>Would just be nice if you know, some of these

676
00:32:49.559 --> 00:32:52.720
<v Speaker 1>things came with technical descriptions of actually what's happening, so

677
00:32:52.839 --> 00:32:55.079
<v Speaker 1>people like us could understand it a little better.

678
00:32:55.119 --> 00:32:57.519
<v Speaker 2>We are non average, no, it's the problem.

679
00:32:57.240 --> 00:32:58.759
<v Speaker 1>But you know, we are displaners.

680
00:32:58.839 --> 00:33:00.000
<v Speaker 2>We are below average.

681
00:33:01.160 --> 00:33:03.160
<v Speaker 1>But you know, if we understand it, we can explain

682
00:33:03.200 --> 00:33:07.119
<v Speaker 1>it to people. Story we're explainers, We're splainers.

683
00:33:07.200 --> 00:33:10.119
<v Speaker 2>That's right, it's the it's the classic ask for forgiveness

684
00:33:10.200 --> 00:33:11.079
<v Speaker 2>rather than permission.

685
00:33:11.400 --> 00:33:15.119
<v Speaker 1>Mm hmm, all right, next story.

686
00:33:15.400 --> 00:33:16.799
<v Speaker 3>That's that's our hot take on that.

687
00:33:16.839 --> 00:33:20.160
<v Speaker 1>This one I thought was pretty neat. X Labs underscore

688
00:33:20.279 --> 00:33:24.759
<v Speaker 1>v one botnet exploits a dB in IoT devices for

689
00:33:24.920 --> 00:33:29.920
<v Speaker 1>d DOS attacks targeting game servers. So what's a dB.

690
00:33:30.720 --> 00:33:35.839
<v Speaker 3>That's the Android debug port. So yeah, I was gonna say, Carl,

691
00:33:36.000 --> 00:33:37.880
<v Speaker 3>you know you used to be the goss to be

692
00:33:37.960 --> 00:33:41.519
<v Speaker 3>the police five years ago on this podcast. You'd be like, WHOA,

693
00:33:41.759 --> 00:33:44.279
<v Speaker 3>WHOA t L d R. Like there's a little bit

694
00:33:44.319 --> 00:33:45.720
<v Speaker 3>of too much acronym soup there.

695
00:33:45.839 --> 00:33:47.759
<v Speaker 2>You just I don't think you see any words in

696
00:33:47.799 --> 00:33:48.720
<v Speaker 2>that title.

697
00:33:48.759 --> 00:33:49.519
<v Speaker 1>I know it's true.

698
00:33:49.599 --> 00:33:51.200
<v Speaker 3>I was listening to it.

699
00:33:52.799 --> 00:33:55.599
<v Speaker 1>Distributed denial of service.

700
00:33:55.920 --> 00:33:58.200
<v Speaker 3>Yeah, I know what AT is.

701
00:33:58.319 --> 00:33:59.680
<v Speaker 1>I just don't know a dB is.

702
00:33:59.759 --> 00:34:02.880
<v Speaker 3>So yeah, so ADB if if you're working with Android devices,

703
00:34:02.880 --> 00:34:04.359
<v Speaker 3>and this could be an Android phone, it could be

704
00:34:04.400 --> 00:34:09.119
<v Speaker 3>a tablet, but it also could be a TV. Right, Actually,

705
00:34:09.119 --> 00:34:12.639
<v Speaker 3>that's a real fun exercise. Go to your TV. Yeah,

706
00:34:12.840 --> 00:34:15.760
<v Speaker 3>go to your TV at your house, and you can

707
00:34:15.760 --> 00:34:17.400
<v Speaker 3>go to the settings on your TV and turn on

708
00:34:17.480 --> 00:34:19.559
<v Speaker 3>debugging and if you do it actually opens up a

709
00:34:19.639 --> 00:34:23.199
<v Speaker 3>port where you can connect to that port and then

710
00:34:23.320 --> 00:34:25.519
<v Speaker 3>run Linux commands on your.

711
00:34:25.280 --> 00:34:28.800
<v Speaker 2>TV, and you too can brick your TV and YouTube.

712
00:34:28.960 --> 00:34:34.760
<v Speaker 3>Yeahs, I've fricked many a TV infecting them. But yeah,

713
00:34:35.000 --> 00:34:36.840
<v Speaker 3>so you can go in and you can run commands

714
00:34:36.840 --> 00:34:38.920
<v Speaker 3>and you can do all sorts of stuff. Oddly enough,

715
00:34:38.960 --> 00:34:42.920
<v Speaker 3>there is a microphone command built into the Android system.

716
00:34:42.960 --> 00:34:46.239
<v Speaker 3>You would think, of course because of Android tablets and phones,

717
00:34:46.280 --> 00:34:48.400
<v Speaker 3>but it's also on the TV. So if your TV

718
00:34:48.480 --> 00:34:50.760
<v Speaker 3>has a micromonic and listen in. But there's all sorts

719
00:34:50.760 --> 00:34:51.239
<v Speaker 3>of new things.

720
00:34:51.280 --> 00:34:53.239
<v Speaker 1>Wait a minute, are you saying that you could debug

721
00:34:53.519 --> 00:34:55.519
<v Speaker 1>just by talking to it or no?

722
00:34:55.559 --> 00:34:57.519
<v Speaker 3>I'm just saying, like I could turn it on. I

723
00:34:57.559 --> 00:34:58.599
<v Speaker 3>could let's use the.

724
00:34:58.599 --> 00:35:00.840
<v Speaker 2>Debug to turn on the mic spy on you.

725
00:35:00.960 --> 00:35:03.599
<v Speaker 3>Yeah, let's say I get on your network and that

726
00:35:03.719 --> 00:35:06.960
<v Speaker 3>debug ports on your TV. I could turn on the microphone.

727
00:35:07.199 --> 00:35:08.559
<v Speaker 1>And what would that do? Exactly?

728
00:35:08.719 --> 00:35:12.079
<v Speaker 3>Well, let's see your uh you know how you can

729
00:35:12.159 --> 00:35:14.320
<v Speaker 3>be like, you know, hey, Google, I want to you

730
00:35:14.360 --> 00:35:15.440
<v Speaker 3>know where can I watch?

731
00:35:16.480 --> 00:35:17.039
<v Speaker 2>I don't know.

732
00:35:18.960 --> 00:35:21.960
<v Speaker 3>Right whatever? Golden Girls, Jesus, how old are you?

733
00:35:23.760 --> 00:35:25.480
<v Speaker 2>It's on at a restaurant that I sam?

734
00:35:26.320 --> 00:35:27.920
<v Speaker 1>How about watership down Patrick.

735
00:35:28.559 --> 00:35:30.440
<v Speaker 3>Yeah, Golden Girls.

736
00:35:30.719 --> 00:35:31.159
<v Speaker 2>Wow.

737
00:35:32.239 --> 00:35:39.559
<v Speaker 3>On that note, Nicked Night yeah, mash okay, matches better.

738
00:35:39.679 --> 00:35:45.079
<v Speaker 3>It's probably older, but better. Joey, do you ever watch

739
00:35:45.079 --> 00:35:47.000
<v Speaker 3>what was it? Empty Nest? It was the spinoff of

740
00:35:47.039 --> 00:35:49.760
<v Speaker 3>Golden Girls. Now, I don't think that was the dude

741
00:35:49.760 --> 00:35:52.519
<v Speaker 3>and his dog. Yeah really, yeah, he used to go.

742
00:35:52.679 --> 00:35:53.679
<v Speaker 3>I think he was the neighbor.

743
00:35:54.639 --> 00:35:57.280
<v Speaker 1>My grandmother used to watch Golden Girls. I never watched

744
00:35:57.320 --> 00:35:58.000
<v Speaker 1>Golden Girls.

745
00:35:58.400 --> 00:36:01.199
<v Speaker 3>Fun fact. You know the old grandmother what was her name?

746
00:36:01.400 --> 00:36:04.239
<v Speaker 3>The youngest she was the Yeah, she was not the

747
00:36:04.280 --> 00:36:06.480
<v Speaker 3>oldest lady. They just made her look old, the old

748
00:36:06.480 --> 00:36:09.480
<v Speaker 3>Italian lady. Yeah, yeah, yeah, you know, I did not

749
00:36:09.599 --> 00:36:12.880
<v Speaker 3>know that. But anyways, where were we girls?

750
00:36:13.079 --> 00:36:13.559
<v Speaker 2>Thank you?

751
00:36:14.920 --> 00:36:19.159
<v Speaker 3>But adb abb back to adb Yeah. So with a

752
00:36:19.199 --> 00:36:21.920
<v Speaker 3>debug port, you do you get? You get access to

753
00:36:22.280 --> 00:36:25.119
<v Speaker 3>all sorts of information on the device. You can run

754
00:36:25.119 --> 00:36:27.440
<v Speaker 3>commands on the device. You can also back up the device.

755
00:36:28.320 --> 00:36:31.199
<v Speaker 3>But more importantly, I can push APKs to the device,

756
00:36:31.559 --> 00:36:34.119
<v Speaker 3>so I can actually create my own Android.

757
00:36:33.760 --> 00:36:38.360
<v Speaker 1>And APK yeah packs and aapp application, packaged whatever.

758
00:36:38.639 --> 00:36:41.000
<v Speaker 3>And I can create an application, and I can push

759
00:36:41.039 --> 00:36:42.639
<v Speaker 3>it to the TV, and I can set it to

760
00:36:42.760 --> 00:36:46.480
<v Speaker 3>run when the TV turns on, like maybe reaching out

761
00:36:46.480 --> 00:36:48.960
<v Speaker 3>to my website and giving me your Wi Fi password

762
00:36:49.440 --> 00:36:53.440
<v Speaker 3>or whatever. Right, so we've ever run that one. Yeah.

763
00:36:53.480 --> 00:36:55.400
<v Speaker 3>ADB ports are fun.

764
00:36:55.239 --> 00:36:57.039
<v Speaker 1>And nothing is there to stop you.

765
00:36:57.079 --> 00:37:00.440
<v Speaker 2>No bugs like a double oh seven designation kill.

766
00:37:01.679 --> 00:37:02.880
<v Speaker 1>So is this what happened?

767
00:37:02.960 --> 00:37:05.679
<v Speaker 3>I mean, oh, you want the actual story?

768
00:37:05.920 --> 00:37:09.280
<v Speaker 1>Yeah, the actual story, Like how does one get infected

769
00:37:09.280 --> 00:37:09.840
<v Speaker 1>with this thing?

770
00:37:10.000 --> 00:37:10.239
<v Speaker 2>Yeah?

771
00:37:10.280 --> 00:37:10.920
<v Speaker 1>Do you have to?

772
00:37:11.159 --> 00:37:11.440
<v Speaker 2>Yeah?

773
00:37:11.480 --> 00:37:15.880
<v Speaker 3>So yeah, absolutely so. The the the ADB port debug

774
00:37:15.920 --> 00:37:18.239
<v Speaker 3>port sometimes is on, sometimes it is not. Depends on

775
00:37:18.280 --> 00:37:20.840
<v Speaker 3>the model of TV and whether it was turned on

776
00:37:20.960 --> 00:37:23.280
<v Speaker 3>or not, or the model of the device, whether it

777
00:37:23.320 --> 00:37:25.719
<v Speaker 3>was turned on or not, et cetera. Phones usually are

778
00:37:25.719 --> 00:37:28.639
<v Speaker 3>pretty protected, usually shut down unless you put it in

779
00:37:28.679 --> 00:37:32.480
<v Speaker 3>a developer mode, which takes you know, going to settings

780
00:37:32.519 --> 00:37:35.159
<v Speaker 3>and clicking on the about eighty five times and then

781
00:37:35.280 --> 00:37:37.719
<v Speaker 3>saying yes and then spinning around and throwing the phone

782
00:37:37.719 --> 00:37:40.159
<v Speaker 3>in the air and all sorts of weird stuff. But

783
00:37:40.239 --> 00:37:43.239
<v Speaker 3>other devices not so much, right, whether it's a camera,

784
00:37:43.440 --> 00:37:47.119
<v Speaker 3>whether it's a you know, IVR system, whether it's a TV.

785
00:37:47.800 --> 00:37:50.760
<v Speaker 3>Lots of times those ports are left on and they're

786
00:37:50.800 --> 00:37:54.760
<v Speaker 3>little mini computers, right, so great target for a botnet

787
00:37:55.239 --> 00:37:57.760
<v Speaker 3>where I might hit that port and then take control

788
00:37:57.800 --> 00:38:01.760
<v Speaker 3>of the device and run an application. Okay, yep, how

789
00:38:01.760 --> 00:38:05.039
<v Speaker 3>do you protect against this? If you have an Android device,

790
00:38:05.559 --> 00:38:07.599
<v Speaker 3>go to settings and make sure debugging is turned off.

791
00:38:07.719 --> 00:38:11.000
<v Speaker 1>Okay, but does it require physical access for somebody to

792
00:38:11.039 --> 00:38:11.599
<v Speaker 1>get on there?

793
00:38:11.840 --> 00:38:12.039
<v Speaker 2>Nah?

794
00:38:12.159 --> 00:38:14.000
<v Speaker 3>No, no, just needs access to the network.

795
00:38:14.039 --> 00:38:14.199
<v Speaker 1>Wow.

796
00:38:14.320 --> 00:38:16.119
<v Speaker 3>So if the device is connected to the Internet, that

797
00:38:16.199 --> 00:38:19.559
<v Speaker 3>means your network is the World Wide Web. If it's

798
00:38:20.119 --> 00:38:23.159
<v Speaker 3>connected locally, that means they need to break into you know,

799
00:38:23.199 --> 00:38:24.840
<v Speaker 3>your Wi Fi or local network or whatever.

800
00:38:24.920 --> 00:38:26.480
<v Speaker 1>But let's face it, if you're a TV, you're going

801
00:38:26.559 --> 00:38:28.719
<v Speaker 1>to be on the Internet because that's how they pull

802
00:38:28.800 --> 00:38:30.280
<v Speaker 1>down shows, right Yeah.

803
00:38:30.360 --> 00:38:32.719
<v Speaker 3>Well, and a lot of people don't even think about

804
00:38:32.920 --> 00:38:34.840
<v Speaker 3>you know, oh well I just plug this TV into

805
00:38:34.840 --> 00:38:36.840
<v Speaker 3>this you know, open wifire or whatever, right at a

806
00:38:37.159 --> 00:38:40.119
<v Speaker 3>Starbucks or whatever. Right, so then you have access to it.

807
00:38:40.199 --> 00:38:42.760
<v Speaker 1>So that means your TV or device could get owned

808
00:38:43.199 --> 00:38:47.280
<v Speaker 1>or poned and used in a d DOS attack against Yeah, you.

809
00:38:47.239 --> 00:38:50.320
<v Speaker 2>Wouldn't know it. It could also become a microphone that

810
00:38:50.400 --> 00:38:51.039
<v Speaker 2>spies on you.

811
00:38:51.599 --> 00:38:53.880
<v Speaker 3>Yeah, yep, all right, or they just at least know

812
00:38:53.880 --> 00:38:54.880
<v Speaker 3>what you're watching on TV.

813
00:38:55.320 --> 00:38:58.239
<v Speaker 1>So again, the fix is to go into debug mode

814
00:38:58.280 --> 00:39:01.280
<v Speaker 1>and make sure that oh, debug mode is off.

815
00:39:01.519 --> 00:39:04.679
<v Speaker 3>Yep, make sure debuggy mode is off. I mean, if

816
00:39:04.719 --> 00:39:07.719
<v Speaker 3>you're really paranoid and technical and map your devices in

817
00:39:07.760 --> 00:39:10.119
<v Speaker 3>the network and look for that port pretty quick way.

818
00:39:10.159 --> 00:39:13.199
<v Speaker 1>But this could be. This could be anything that runs Android.

819
00:39:13.360 --> 00:39:18.159
<v Speaker 3>Oh yeah, absolutely, refrigerator, refrigerator, yeah, washing, dishwasher, you name

820
00:39:18.199 --> 00:39:19.880
<v Speaker 3>it anything. Yeah, exactly.

821
00:39:20.320 --> 00:39:22.960
<v Speaker 2>You can take an IP address you need to check it.

822
00:39:23.159 --> 00:39:28.119
<v Speaker 1>I kind of got to believe that most home appliances

823
00:39:28.280 --> 00:39:31.599
<v Speaker 1>that are running Android are not running in debug mode.

824
00:39:31.679 --> 00:39:33.440
<v Speaker 2>Oh sure, you're so precious.

825
00:39:34.679 --> 00:39:38.639
<v Speaker 1>Do you really think that you really think that my

826
00:39:39.159 --> 00:39:41.840
<v Speaker 1>Samsung fridge they got it best Buy, has debug mode

827
00:39:41.960 --> 00:39:42.880
<v Speaker 1>enabled by default.

828
00:39:43.280 --> 00:39:46.199
<v Speaker 2>Maybe it gets a random coin toss it gets. I

829
00:39:46.239 --> 00:39:50.880
<v Speaker 2>think this is the new default password bubblem vendors, they don't.

830
00:39:51.239 --> 00:39:53.159
<v Speaker 2>They just want to sell the product, so until it

831
00:39:53.199 --> 00:39:58.599
<v Speaker 2>becomes a scandal, there's no motivation for them to police this. Yeah, geez, well,

832
00:39:58.599 --> 00:40:01.280
<v Speaker 2>I wouldn't be surprised if if it's eighty percent of device,

833
00:40:01.360 --> 00:40:04.119
<v Speaker 2>I don't know. That wouldn't surprise me at all. This

834
00:40:04.239 --> 00:40:06.880
<v Speaker 2>is just the next level, and there'll be another one

835
00:40:06.920 --> 00:40:09.599
<v Speaker 2>after this, right, be another level. The good news is

836
00:40:09.639 --> 00:40:11.880
<v Speaker 2>this is better than a hard coded password that gives

837
00:40:11.920 --> 00:40:15.280
<v Speaker 2>you because at least you could turn this off. Yeah,

838
00:40:15.400 --> 00:40:17.559
<v Speaker 2>there's always, as they say in Men in Black, there's

839
00:40:17.599 --> 00:40:21.280
<v Speaker 2>always a Corellian cruiser or a black flag or something like.

840
00:40:21.599 --> 00:40:23.800
<v Speaker 2>It's always the next thing. This just happens to be

841
00:40:23.840 --> 00:40:25.280
<v Speaker 2>the next thing, all right.

842
00:40:25.119 --> 00:40:27.840
<v Speaker 1>This next one is a bad one, Palo Alto Networks

843
00:40:27.880 --> 00:40:33.280
<v Speaker 1>warns State cluster behind. Yeah, well the last one is

844
00:40:33.320 --> 00:40:36.519
<v Speaker 1>the worst one. But Palo Alto Networks warren state linked

845
00:40:36.559 --> 00:40:41.480
<v Speaker 1>cluster behind a zero day exploit exploitation. A patch for

846
00:40:41.519 --> 00:40:44.559
<v Speaker 1>the flaw, which hackers began targeting in early April, won't

847
00:40:44.599 --> 00:40:47.440
<v Speaker 1>be ready for another week, and this was published May seventh.

848
00:40:48.280 --> 00:40:52.880
<v Speaker 1>So there's a zero day out there, and you you're

849
00:40:52.960 --> 00:40:54.239
<v Speaker 1>not protected. Yeah.

850
00:40:54.360 --> 00:40:56.880
<v Speaker 3>Not only is there is zero day, it's for a firewall.

851
00:40:57.000 --> 00:40:58.639
<v Speaker 3>Not only is it for a firewall, it means it's

852
00:40:58.679 --> 00:41:01.000
<v Speaker 3>connected to the Internet, and it's exploited by people who

853
00:41:01.079 --> 00:41:01.679
<v Speaker 3>know what they're doing.

854
00:41:01.760 --> 00:41:04.920
<v Speaker 2>So, yeah, you just don't use the internet this month.

855
00:41:05.280 --> 00:41:09.400
<v Speaker 1>Yeah, I've unplug it, unplug it, replace it with something else. Literally,

856
00:41:09.440 --> 00:41:10.840
<v Speaker 1>that's that's what you have.

857
00:41:10.880 --> 00:41:16.679
<v Speaker 2>Toces that horrible conundrum of you know your car is

858
00:41:16.719 --> 00:41:18.719
<v Speaker 2>a death trap and you know it's likely to kill

859
00:41:18.760 --> 00:41:20.880
<v Speaker 2>you any mile but you have to go to work

860
00:41:21.000 --> 00:41:23.800
<v Speaker 2>and you can't be fixed until next month. So you

861
00:41:23.880 --> 00:41:25.639
<v Speaker 2>take you roll the dice, you take your chances.

862
00:41:26.440 --> 00:41:27.800
<v Speaker 1>Call an uber, That's what I would.

863
00:41:27.840 --> 00:41:29.840
<v Speaker 2>Yeah, well you can't call an uber for your.

864
00:41:29.760 --> 00:41:32.719
<v Speaker 1>Firewall, no, no, no, But I'm saying there are ways

865
00:41:32.760 --> 00:41:35.920
<v Speaker 1>around it, which is replace the unit. Yeah, first of all,

866
00:41:35.960 --> 00:41:36.559
<v Speaker 1>turn it off.

867
00:41:36.960 --> 00:41:39.039
<v Speaker 3>Monitoring is another way to keep an eye on it.

868
00:41:39.079 --> 00:41:42.840
<v Speaker 1>Well, monitoring, but you could you could watch it get exploited.

869
00:41:42.840 --> 00:41:43.440
<v Speaker 1>Does that help?

870
00:41:43.599 --> 00:41:48.000
<v Speaker 3>That might be neat.

871
00:41:48.119 --> 00:41:52.280
<v Speaker 2>This is really yeah, that's a good Friday night for me.

872
00:41:52.480 --> 00:41:54.400
<v Speaker 2>I think this is a view of our future. I

873
00:41:54.440 --> 00:41:57.440
<v Speaker 2>think that we're going to get things that are discovered

874
00:41:57.559 --> 00:41:59.960
<v Speaker 2>as exploited and there's not going to be a fit

875
00:42:00.360 --> 00:42:03.719
<v Speaker 2>readily available right right, So you have to figure out like, oh,

876
00:42:03.800 --> 00:42:06.519
<v Speaker 2>you know what. I think this is where tabletop exercises

877
00:42:06.559 --> 00:42:08.199
<v Speaker 2>have to come in and people have to start thinking

878
00:42:08.199 --> 00:42:10.960
<v Speaker 2>about what's the worst they can happen. Now, military does

879
00:42:11.000 --> 00:42:13.280
<v Speaker 2>this all the time, at least when they're done, when

880
00:42:13.280 --> 00:42:16.119
<v Speaker 2>they're operating competently. The military does this all the time,

881
00:42:16.760 --> 00:42:18.440
<v Speaker 2>like what if they do this? What if they did that?

882
00:42:19.079 --> 00:42:23.920
<v Speaker 2>And you try to predict the unpredictable, most organizations aren't

883
00:42:24.400 --> 00:42:26.880
<v Speaker 2>geared for this, they're not ready for it. They can't

884
00:42:26.920 --> 00:42:27.360
<v Speaker 2>handle it.

885
00:42:28.159 --> 00:42:31.679
<v Speaker 1>Wow. So what is the device itself that people have

886
00:42:31.719 --> 00:42:35.559
<v Speaker 1>to worry about it? It's the PA series and VM

887
00:42:35.639 --> 00:42:39.440
<v Speaker 1>series firewalls. Yes, that's a lot of firewalls.

888
00:42:39.639 --> 00:42:42.840
<v Speaker 3>That's a lot. Yeah, it's a big line of firewalls.

889
00:42:42.840 --> 00:42:45.920
<v Speaker 1>Actually. Wow. Yeah, so this isn't a go patch because

890
00:42:45.920 --> 00:42:51.000
<v Speaker 1>there is no patch, replace it at least for a while. Okay.

891
00:42:52.239 --> 00:42:56.840
<v Speaker 1>The penultimate story today, which I thought was hilarious, student

892
00:42:57.039 --> 00:43:00.840
<v Speaker 1>hacked Taiwan high speed rail to trigger urgency breaks.

893
00:43:01.119 --> 00:43:04.800
<v Speaker 3>Isn't that just like a kid by law? By law,

894
00:43:04.800 --> 00:43:05.599
<v Speaker 3>I can't talk about this.

895
00:43:06.119 --> 00:43:07.920
<v Speaker 1>It's like it's like pulling them all.

896
00:43:08.039 --> 00:43:10.039
<v Speaker 2>I'm surprised you actually obaye that.

897
00:43:10.360 --> 00:43:12.039
<v Speaker 3>A year and a half ago though I told people

898
00:43:12.079 --> 00:43:14.239
<v Speaker 3>how to do this, but today I won't.

899
00:43:14.280 --> 00:43:16.400
<v Speaker 1>It's kind of like pulling the fire alarm, you know,

900
00:43:16.440 --> 00:43:19.039
<v Speaker 1>in grammar school, just walking by, pull the fire alarm,

901
00:43:19.119 --> 00:43:19.840
<v Speaker 1>see what happens.

902
00:43:19.880 --> 00:43:23.039
<v Speaker 3>Okay, so here's the deal. Now, it's a first off,

903
00:43:23.320 --> 00:43:26.119
<v Speaker 3>if you look at this article, just just zoom in

904
00:43:26.199 --> 00:43:27.119
<v Speaker 3>on that desk.

905
00:43:26.920 --> 00:43:27.800
<v Speaker 2>That kid has.

906
00:43:28.000 --> 00:43:31.199
<v Speaker 3>Yeah, I don't think I would ever want a picture

907
00:43:32.239 --> 00:43:36.320
<v Speaker 3>of my desk with I think that's hand cream over

908
00:43:36.360 --> 00:43:40.000
<v Speaker 3>there and maybe his braces like his twenty three year

909
00:43:40.039 --> 00:43:43.039
<v Speaker 3>old Yeah, exactly. But he's also got a lot of

910
00:43:43.079 --> 00:43:46.079
<v Speaker 3>software defined radios in there. I will tell you too.

911
00:43:47.000 --> 00:43:49.199
<v Speaker 3>And we've done this our team, I know, if you

912
00:43:49.239 --> 00:43:50.559
<v Speaker 3>go back, I think it's a year and a half,

913
00:43:50.559 --> 00:43:52.840
<v Speaker 3>two years ago. Well, my team actually went down to

914
00:43:52.920 --> 00:43:57.440
<v Speaker 3>DC and and did a session breaking into trains and one.

915
00:43:57.360 --> 00:44:01.039
<v Speaker 2>Of the governments helped to help the government try and abilities.

916
00:44:01.039 --> 00:44:03.400
<v Speaker 1>Patrick tried to stop him from talking about it, but

917
00:44:03.440 --> 00:44:04.199
<v Speaker 1>he did anyway.

918
00:44:04.559 --> 00:44:07.679
<v Speaker 3>Yeah, he's tried to stop me multiple times. And one

919
00:44:07.679 --> 00:44:11.320
<v Speaker 3>of the exploits that we deployed actually was the ability

920
00:44:11.360 --> 00:44:14.400
<v Speaker 3>to engage the E break on the trains. It didn't

921
00:44:14.480 --> 00:44:17.760
<v Speaker 3>it didn't matter. It could be high speed, right. And

922
00:44:17.800 --> 00:44:19.920
<v Speaker 3>I will tell you this. This has been deployed a

923
00:44:19.920 --> 00:44:25.760
<v Speaker 3>couple times in anger, you know, by nation states, either

924
00:44:25.800 --> 00:44:29.599
<v Speaker 3>pre attack or during attacks of nation states. So the

925
00:44:29.639 --> 00:44:32.760
<v Speaker 3>way this, the way this works is the E break

926
00:44:32.880 --> 00:44:37.039
<v Speaker 3>is actually a radio signal, right, So you can enable

927
00:44:37.119 --> 00:44:41.320
<v Speaker 3>it by playing the right frequency. And it's not just

928
00:44:41.400 --> 00:44:45.039
<v Speaker 3>a frequency, right, it's it's actually it's three frequencies within

929
00:44:45.039 --> 00:44:48.599
<v Speaker 3>a period of time or whatever. But but what I

930
00:44:48.639 --> 00:44:52.920
<v Speaker 3>will tell you is the spec for this is on

931
00:44:53.119 --> 00:44:56.519
<v Speaker 3>the internet. That the way that the train systems work,

932
00:44:56.639 --> 00:44:59.840
<v Speaker 3>if you like. For those of you who love reading

933
00:45:00.119 --> 00:45:05.400
<v Speaker 3>really boring documentation, if you go read the documentation on

934
00:45:05.599 --> 00:45:08.400
<v Speaker 3>how the rail systems work, you'll you'll find the frequencies.

935
00:45:08.400 --> 00:45:10.800
<v Speaker 3>You'll know exactly which ones to call it in when order.

936
00:45:11.880 --> 00:45:15.280
<v Speaker 3>You know, you'll see how they do train tracking. Heads up,

937
00:45:15.360 --> 00:45:18.519
<v Speaker 3>they track trains over two point four gig hertz, not

938
00:45:18.639 --> 00:45:22.800
<v Speaker 3>Wi Fi, but same band. And if you jam it,

939
00:45:22.960 --> 00:45:25.199
<v Speaker 3>they lose a train and they shut down the whole system.

940
00:45:25.199 --> 00:45:28.000
<v Speaker 3>And that was actually deployed I think against Poland at

941
00:45:28.000 --> 00:45:28.440
<v Speaker 3>one point.

942
00:45:28.599 --> 00:45:28.880
<v Speaker 1>Wow.

943
00:45:29.199 --> 00:45:33.039
<v Speaker 3>So yeah, it's you know, you think like, oh my god,

944
00:45:33.079 --> 00:45:34.880
<v Speaker 3>this twenty one year old's got to be you know,

945
00:45:35.000 --> 00:45:38.400
<v Speaker 3>a genius. Twenty three not really or twenty three year

946
00:45:38.440 --> 00:45:40.079
<v Speaker 3>old not really.

947
00:45:40.760 --> 00:45:41.039
<v Speaker 2>Really.

948
00:45:41.079 --> 00:45:44.280
<v Speaker 3>All you need to do is read the RFCs.

949
00:45:44.599 --> 00:45:46.719
<v Speaker 1>Right, the request for comments.

950
00:45:46.480 --> 00:45:49.320
<v Speaker 3>Yeah, the the manual if you will, the design doc

951
00:45:49.800 --> 00:45:54.599
<v Speaker 3>for how these systems are put together, and then do

952
00:45:54.719 --> 00:45:57.400
<v Speaker 3>it right, grab a software defined radio and do it. So,

953
00:45:57.679 --> 00:46:00.199
<v Speaker 3>you know, I, although I think it's fun and it's

954
00:46:00.239 --> 00:46:02.559
<v Speaker 3>neat to kind of play around with these types of systems,

955
00:46:02.599 --> 00:46:05.440
<v Speaker 3>I don't applaud people who are like, you know, in

956
00:46:05.800 --> 00:46:10.079
<v Speaker 3>real live trains causing or any system. Right, I'll tell

957
00:46:10.119 --> 00:46:12.920
<v Speaker 3>you elevators have all sorts of systems she can play

958
00:46:12.920 --> 00:46:16.079
<v Speaker 3>with and that sort of stuff, but you shouldn't do

959
00:46:16.159 --> 00:46:16.760
<v Speaker 3>it now, right.

960
00:46:17.239 --> 00:46:20.519
<v Speaker 1>So yeah, so this is really the irony. The irony

961
00:46:20.559 --> 00:46:22.960
<v Speaker 1>of this story is as the kid was walking out

962
00:46:22.960 --> 00:46:26.119
<v Speaker 1>of the train in handcuffs, he looked on the wall

963
00:46:26.320 --> 00:46:29.760
<v Speaker 1>the window by the by the window and read the

964
00:46:29.840 --> 00:46:33.719
<v Speaker 1>sign with the cord please pull cord for emergency break.

965
00:46:34.840 --> 00:46:39.599
<v Speaker 2>He's like, oh, geez, which does the same thing. Yeah, exactly, Like.

966
00:46:42.000 --> 00:46:44.199
<v Speaker 3>So, I don't know, I, like I said, I think

967
00:46:44.320 --> 00:46:49.039
<v Speaker 3>I think it's interesting. I think I think it's fascinating

968
00:46:49.039 --> 00:46:50.679
<v Speaker 3>how all these things work and that sort of stuff.

969
00:46:50.679 --> 00:46:53.639
<v Speaker 3>But you know, to do it on a real life train, like, yeah,

970
00:46:53.719 --> 00:46:54.800
<v Speaker 3>you deserve what you get.

971
00:46:55.119 --> 00:46:59.440
<v Speaker 1>All right, We're finally to our top story and this

972
00:46:59.480 --> 00:47:04.719
<v Speaker 1>one is our top story just because it's so bad, big,

973
00:47:05.199 --> 00:47:10.079
<v Speaker 1>big and bad. Otherwise, in Dwayne speak, awesome copy fail

974
00:47:10.760 --> 00:47:13.960
<v Speaker 1>seven hundred and thirty two bytes to root on every

975
00:47:14.079 --> 00:47:17.679
<v Speaker 1>major Linux distribution. So that sounds like kind of an

976
00:47:17.719 --> 00:47:20.480
<v Speaker 1>innocuous headline. But Twain Munch, you'd tell us what this

977
00:47:20.519 --> 00:47:21.519
<v Speaker 1>sort really all about.

978
00:47:22.920 --> 00:47:29.599
<v Speaker 3>Yeah, and actually what's crazy is right now, if you're

979
00:47:29.639 --> 00:47:34.840
<v Speaker 3>on X or you're on whatever, there's tons of there's

980
00:47:34.920 --> 00:47:38.440
<v Speaker 3>a place hosting this exploit where you can run one

981
00:47:38.519 --> 00:47:41.599
<v Speaker 3>command and it pulls the exploit down, runs it in memory,

982
00:47:41.639 --> 00:47:44.599
<v Speaker 3>and gives you root access to almost any Linux box.

983
00:47:45.000 --> 00:47:48.599
<v Speaker 3>And crazy enough, this isn't just root level access. So

984
00:47:48.599 --> 00:47:50.840
<v Speaker 3>we talk about privask and how you know, a lot

985
00:47:50.840 --> 00:47:54.039
<v Speaker 3>of us listen, we don't pooh poop privesk, but we

986
00:47:54.079 --> 00:47:57.719
<v Speaker 3>say remote code execution is more dangerous, right, We've said

987
00:47:57.719 --> 00:47:59.719
<v Speaker 3>that all the time, Like, if you can remotely from

988
00:48:00.079 --> 00:48:02.519
<v Speaker 3>you know, North Korea, exploit a system here in the

989
00:48:02.599 --> 00:48:06.000
<v Speaker 3>United States, that's more dangerous than if they're on the

990
00:48:06.039 --> 00:48:08.920
<v Speaker 3>box they can become an admin. The reason we put

991
00:48:08.920 --> 00:48:12.000
<v Speaker 3>this is kind of the top story is first off,

992
00:48:12.320 --> 00:48:17.559
<v Speaker 3>because of its breadth. It's every major Linux system since

993
00:48:17.599 --> 00:48:18.400
<v Speaker 3>twenty seventeen.

994
00:48:18.519 --> 00:48:22.280
<v Speaker 2>It's not just desktops and servers. It's also like containers

995
00:48:22.440 --> 00:48:24.239
<v Speaker 2>and things like that that we don't think about.

996
00:48:24.440 --> 00:48:27.719
<v Speaker 3>Yeah, and k eights and you name it, right, So

997
00:48:27.800 --> 00:48:32.599
<v Speaker 3>this is way far reaching. It does give you the

998
00:48:32.599 --> 00:48:34.400
<v Speaker 3>ability to even jump outside of containers.

999
00:48:34.440 --> 00:48:34.679
<v Speaker 1>Wow.

1000
00:48:35.360 --> 00:48:38.039
<v Speaker 3>So a lot of people use containers as a security layer.

1001
00:48:38.119 --> 00:48:40.000
<v Speaker 3>This allows you to jump out and do all sorts

1002
00:48:40.000 --> 00:48:44.199
<v Speaker 3>of nasty things. And the really neat thing about this,

1003
00:48:44.519 --> 00:48:46.480
<v Speaker 3>For those of you who've been listening to this for

1004
00:48:46.519 --> 00:48:48.239
<v Speaker 3>a long this show for a long time, I want

1005
00:48:48.239 --> 00:48:51.599
<v Speaker 3>to say about maybe five years ago we talked about

1006
00:48:51.639 --> 00:48:54.239
<v Speaker 3>Dirty Cow, and then there was a whole bunch of

1007
00:48:54.280 --> 00:48:59.000
<v Speaker 3>Cow series that came out. But Dirty Cow was sort

1008
00:48:59.039 --> 00:49:04.400
<v Speaker 3>of role did as an amazing ability to become root

1009
00:49:04.519 --> 00:49:08.480
<v Speaker 3>level access on almost any Linux workstation, any Linux station.

1010
00:49:08.639 --> 00:49:11.719
<v Speaker 3>And what you were doing is you were exploiting a

1011
00:49:11.840 --> 00:49:17.119
<v Speaker 3>race case condition. And with Dirty Cow, when you exploited

1012
00:49:17.159 --> 00:49:22.159
<v Speaker 3>this race case condition, sometimes it worked, call it seventy

1013
00:49:22.199 --> 00:49:25.400
<v Speaker 3>percent of the time and you became root. Sometimes it

1014
00:49:25.480 --> 00:49:29.360
<v Speaker 3>did nothing, call that twenty percent of the time. And

1015
00:49:29.360 --> 00:49:32.199
<v Speaker 3>then sometimes it crashed the system. Call that ten percent

1016
00:49:32.199 --> 00:49:34.880
<v Speaker 3>of the time. So although it was it was cool

1017
00:49:35.119 --> 00:49:38.719
<v Speaker 3>and it was powerful, it wasn't as powerful as this.

1018
00:49:38.719 --> 00:49:39.320
<v Speaker 1>This is bad.

1019
00:49:40.480 --> 00:49:41.760
<v Speaker 2>Yeah. Yeah.

1020
00:49:42.360 --> 00:49:46.400
<v Speaker 3>This is a predetermined set of bytes that you can

1021
00:49:47.400 --> 00:49:52.079
<v Speaker 3>issue to a privileged area that will one hundred percent

1022
00:49:52.079 --> 00:49:54.599
<v Speaker 3>of the time give you rout and race case conditions

1023
00:49:54.639 --> 00:49:57.360
<v Speaker 3>and it is not a potential of crashing. It is

1024
00:49:57.480 --> 00:49:59.679
<v Speaker 3>super reliable, and that's one of the reasons that we

1025
00:50:00.119 --> 00:50:02.760
<v Speaker 3>of upgraded this as well, is because this is a

1026
00:50:02.960 --> 00:50:08.079
<v Speaker 3>highly reliable way of becoming root on almost any LINUXEDS

1027
00:50:08.559 --> 00:50:09.440
<v Speaker 3>planet at this point.

1028
00:50:09.519 --> 00:50:09.840
<v Speaker 2>Yeah.

1029
00:50:09.880 --> 00:50:14.559
<v Speaker 3>Absolutely, Oh, speaking of quiet, the way it yeah, I

1030
00:50:14.599 --> 00:50:15.159
<v Speaker 3>know this is.

1031
00:50:15.199 --> 00:50:16.000
<v Speaker 2>This thing's awesome.

1032
00:50:16.360 --> 00:50:19.400
<v Speaker 3>The way it injects the bites and the way you

1033
00:50:19.480 --> 00:50:24.639
<v Speaker 3>garner the token afterwards will not be picked up by

1034
00:50:24.840 --> 00:50:29.960
<v Speaker 3>ed rs, by by your anti virus, your endpoint detections.

1035
00:50:30.000 --> 00:50:33.039
<v Speaker 3>It's not even picked up by any of your logging systems.

1036
00:50:34.000 --> 00:50:38.039
<v Speaker 3>So this is not only like when you say quiet, Patrick,

1037
00:50:38.239 --> 00:50:42.320
<v Speaker 3>it is quiet, right, You're not going to see this happen.

1038
00:50:43.719 --> 00:50:45.280
<v Speaker 3>So it's it's super dangerous.

1039
00:50:45.679 --> 00:50:46.079
<v Speaker 1>You're dead.

1040
00:50:46.119 --> 00:50:46.880
<v Speaker 2>You just don't know it.

1041
00:50:47.000 --> 00:50:50.679
<v Speaker 3>Yeah, it's super dangerous, and it's it's impressive. Honestly, are

1042
00:50:50.679 --> 00:50:52.599
<v Speaker 3>there patches out there at the kernel level yet?

1043
00:50:52.599 --> 00:50:52.840
<v Speaker 2>There are?

1044
00:50:53.199 --> 00:50:54.840
<v Speaker 3>Yes, you do need to make sure you go out

1045
00:50:54.840 --> 00:50:56.119
<v Speaker 3>and patch immediately.

1046
00:50:56.440 --> 00:50:59.239
<v Speaker 1>At the end of the article, it gives you the

1047
00:50:59.280 --> 00:51:00.480
<v Speaker 1>sects and a link to the patch.

1048
00:51:00.599 --> 00:51:00.840
<v Speaker 2>Yeah.

1049
00:51:01.199 --> 00:51:04.159
<v Speaker 1>But we're which basically moves things around, doesn't it.

1050
00:51:04.119 --> 00:51:07.719
<v Speaker 3>Right, Yeah, exactly. Actually, so if you dig back why

1051
00:51:07.800 --> 00:51:12.679
<v Speaker 3>twenty seventeen. In twenty seventeen, there was an optimization added

1052
00:51:13.039 --> 00:51:15.800
<v Speaker 3>to this particular part of memory read and write and

1053
00:51:15.800 --> 00:51:19.039
<v Speaker 3>that sort of stuff that's being exploited. Now what the

1054
00:51:19.079 --> 00:51:22.480
<v Speaker 3>patch does is rolls it back, removes it. Yeah, takes

1055
00:51:22.480 --> 00:51:28.719
<v Speaker 3>that optimization out, because that optimization is what actually exposed this.

1056
00:51:28.920 --> 00:51:33.639
<v Speaker 1>And then apparently the kernel is being patch too, so

1057
00:51:33.719 --> 00:51:36.119
<v Speaker 1>there will be a version of your Linux kernel that

1058
00:51:36.199 --> 00:51:38.239
<v Speaker 1>will be safe yep, eventually.

1059
00:51:38.320 --> 00:51:40.239
<v Speaker 3>But this is if you want to read about this,

1060
00:51:40.280 --> 00:51:43.079
<v Speaker 3>there's a great article we have the link to to

1061
00:51:43.239 --> 00:51:45.920
<v Speaker 3>really dig into kind of the bits invites and how

1062
00:51:45.960 --> 00:51:51.880
<v Speaker 3>this works. I I'm wondering, I didn't see it in

1063
00:51:51.920 --> 00:51:55.480
<v Speaker 3>this article. They actually do have some code in here, but.

1064
00:51:55.880 --> 00:51:58.000
<v Speaker 1>They do they say how it works.

1065
00:51:58.280 --> 00:52:01.639
<v Speaker 3>Yeah, but there there's a super simple one liner. If

1066
00:52:01.679 --> 00:52:04.920
<v Speaker 3>you do a search for copy fail one liner, you'll

1067
00:52:04.920 --> 00:52:06.960
<v Speaker 3>find it and you can run it on a Linux

1068
00:52:07.000 --> 00:52:09.079
<v Speaker 3>box and it'll it'll give you a root.

1069
00:52:09.199 --> 00:52:09.920
<v Speaker 2>So that's sad.

1070
00:52:09.960 --> 00:52:12.519
<v Speaker 1>Note pned, owned and scrowned.

1071
00:52:14.639 --> 00:52:15.320
<v Speaker 3>This one's awesome.

1072
00:52:15.320 --> 00:52:16.280
<v Speaker 2>This is a long episode.

1073
00:52:16.440 --> 00:52:19.159
<v Speaker 1>This one's awesome, but it was good to catch up.

1074
00:52:19.320 --> 00:52:22.599
<v Speaker 1>And yeah, we're glad you stuck with us. Everybody, we'll

1075
00:52:22.639 --> 00:52:24.840
<v Speaker 1>see on discord. And that's all that happened in the

1076
00:52:24.920 --> 00:52:26.440
<v Speaker 1>last two weeks and we'll see you next week on

1077
00:52:26.519 --> 00:52:28.320
<v Speaker 1>Security this week, Bye bye guys,
