1
00:00:00,000 --> 00:00:04,700
Good morning everyone and welcome to the village talk area.

2
00:00:04,700 --> 00:00:06,480
I am Tony, or ToughNuts.

3
00:00:06,480 --> 00:00:08,080
We're with the Packing Hacking Village.

4
00:00:08,080 --> 00:00:09,220
We are just across the way.

5
00:00:09,220 --> 00:00:11,280
Come, please, and visit us.

6
00:00:11,480 --> 00:00:19,880
And it is my great pleasure to introduce Will, who is going to be talking about some stuff you've found about leakage of data in social media.

7
00:00:19,880 --> 00:00:25,100
So, without much to do, if I could get everyone to please sit down.

8
00:00:25,520 --> 00:00:26,820
Hold on, D9.

9
00:00:28,340 --> 00:00:29,260
Can you?

10
00:00:31,940 --> 00:00:34,040
There we go.

11
00:00:34,040 --> 00:00:36,720
And Will, with that, thank you very much.

12
00:00:36,720 --> 00:00:37,640
Here you go.

13
00:00:39,800 --> 00:00:40,840
Hi.

14
00:00:43,580 --> 00:00:48,200
Could I get a show of hands in the audience, those of you who use Reddit?

15
00:00:49,560 --> 00:00:50,240
Awesome.

16
00:00:50,240 --> 00:00:53,900
Keep your hand raised if you subscribe to the DEF CON subreddit.

17
00:00:55,300 --> 00:00:55,960
Good.

18
00:00:55,960 --> 00:00:57,460
And one last question.

19
00:00:58,180 --> 00:01:03,540
Of those few, keep your hand raised if you post or comment to the DEF CON subreddit.

20
00:01:05,100 --> 00:01:06,860
Not too many, just one.

21
00:01:06,860 --> 00:01:08,140
All right, that's okay.

22
00:01:09,760 --> 00:01:12,180
Yeah, I was hoping for a few more, but that's okay.

23
00:01:12,180 --> 00:01:12,900
None of them are here.

24
00:01:12,900 --> 00:01:14,720
We can talk as much as we want about them.

25
00:01:16,120 --> 00:01:18,080
Yeah, so you've probably read the program.

26
00:01:18,080 --> 00:01:30,500
This talk is about me spending the last 12 months, call it what it is, stalking the DEF CON subreddit, seeking people's personal information that they might be leaking online.

27
00:01:31,460 --> 00:01:32,980
So, yeah, thanks for coming.

28
00:01:33,420 --> 00:01:35,460
Language warning, maybe.

29
00:01:36,060 --> 00:01:38,820
Information warning, definitely.

30
00:01:39,380 --> 00:01:49,560
Particularly anyone who had their hand raised in the last couple of questions, there is a small chance that you individually may be on these slides in the next 45 minutes or so.

31
00:01:50,500 --> 00:01:51,800
But don't freak out.

32
00:01:52,120 --> 00:01:54,740
I've done my due diligence.

33
00:01:55,100 --> 00:01:59,160
There's plenty of black text all over the slides, so you probably wanted to recognise yourself.

34
00:02:02,520 --> 00:02:15,000
Yeah, I just wanted to say before I started as well, this is meant to be a no prior knowledge required speech, so if you have any questions, can we just leave them till the end, and I'm happy to answer them after that.

35
00:02:15,660 --> 00:02:17,700
Let's get on with it.

36
00:02:19,240 --> 00:02:21,380
Yeah, obligatory intro slide.

37
00:02:21,380 --> 00:02:22,520
My name's Will.

38
00:02:22,620 --> 00:02:24,080
I'm from Australia.

39
00:02:24,520 --> 00:02:27,380
I work for a cyber security firm there.

40
00:02:27,680 --> 00:02:31,840
My day job is actually not related to this at all.

41
00:02:31,840 --> 00:02:35,160
I work in cyber security in the supply chain.

42
00:02:35,840 --> 00:02:42,800
So if you're looking through the brief and you're wondering what the hell has this got to do with that, you'd be 100% right.

43
00:02:42,920 --> 00:02:44,600
It has nothing to do with it.

44
00:02:46,280 --> 00:02:55,680
And also, yeah, I guess when I started as well, I had almost none of the skills required and probably still don't to do this kind of thing.

45
00:02:56,340 --> 00:02:58,700
I'm not a coder .

46
00:02:58,780 --> 00:03:00,120
Pretty shit at code.

47
00:03:00,160 --> 00:03:02,000
I'm not a social engineer.

48
00:03:02,760 --> 00:03:08,160
And you probably get that from my shitty dad jokes and my poor taste in memes.

49
00:03:08,160 --> 00:03:09,900
And I'm certainly not a data scientist.

50
00:03:09,900 --> 00:03:14,040
I have no ability to wrangle large data sets.

51
00:03:14,440 --> 00:03:21,740
So yeah, if you're in the audience and you are professional in any of those areas, I apologize in advance.

52
00:03:21,740 --> 00:03:28,480
I'll probably butcher your fields of expertise and misrepresent entirely.

53
00:03:28,480 --> 00:03:33,980
But yeah, I just, I didn't have any of that stuff and just had a crack at it anyway.

54
00:03:37,130 --> 00:03:39,010
So this is my second DEF CON.

55
00:03:39,290 --> 00:03:42,830
I came here last year as like a bucket list thing with a few friends.

56
00:03:43,110 --> 00:03:45,110
We went and visited the wall of sheep.

57
00:03:45,290 --> 00:03:49,190
If you haven't been there, I recommend doing it at some point in the conference.

58
00:03:49,190 --> 00:03:56,270
And we love the idea of like a practical way of kind of showing people what they were leaking on a network, an unsecure network.

59
00:03:56,410 --> 00:03:59,830
And yeah, we thought that was an awesome idea.

60
00:03:59,830 --> 00:04:07,470
And like most things, after a few refreshing beverages, we kind of thought to ourselves, that's a really awesome idea.

61
00:04:07,470 --> 00:04:13,870
You know, during the conference, showing people over those four days, what kind of stuff they're leaking.

62
00:04:13,870 --> 00:04:23,130
But what about all the stuff that people are leaking 365 days of the year, sharing on social media, everything about themselves.

63
00:04:23,790 --> 00:04:27,950
And yeah, from there, we were like, wouldn't that be a cool way?

64
00:04:27,950 --> 00:04:32,150
Wouldn't that be a cool presentation to sort of collate all that information?

65
00:04:32,950 --> 00:04:36,330
And yeah, present it back in a kind of like wall of sheep online edition.

66
00:04:36,330 --> 00:04:38,450
So that's what we kind of came up.

67
00:04:40,130 --> 00:04:41,650
So this is what we're going to be talking about.

68
00:04:41,650 --> 00:04:44,290
What would a wall of sheep online edition look like?

69
00:04:45,150 --> 00:04:51,110
Starting with my most overconfident plans, you know, someone with no ability to do it, deciding to go and do it.

70
00:04:52,210 --> 00:04:53,830
Yeah, what even is a leak?

71
00:04:53,830 --> 00:04:56,250
Took a little bit of time to kind of like have to work that out.

72
00:04:56,250 --> 00:05:02,610
I had no idea what I was looking for or defining what that was going to be.

73
00:05:02,910 --> 00:05:05,130
I needed to learn the recon game.

74
00:05:05,130 --> 00:05:07,730
I had no kind of clue what was going on there.

75
00:05:07,910 --> 00:05:14,210
And then sort of like the rest of this talk is really just talking about how I started hunting and gathering and finding those people.

76
00:05:14,450 --> 00:05:24,550
And then I promise, if you sit through all that, the last few slides, we'll be talking about sort of the most juicy slide packs of people that I found.

77
00:05:27,520 --> 00:05:30,080
Yeah, so I was like, how hard could it be?

78
00:05:30,080 --> 00:05:32,780
You know, everyone knows people leak stuff online.

79
00:05:32,780 --> 00:05:36,140
Everyone knows people overshare all the time.

80
00:05:36,140 --> 00:05:42,860
Social media these days has us lulled into this idea that, you know, oversharing is what you should be doing on social media.

81
00:05:43,140 --> 00:05:54,260
So I was like, I'll just find some target group, come up with some things to look for and then, you know, manually sift through and just try to find interesting folks.

82
00:05:54,540 --> 00:05:58,540
And then shortlist those people and then monitor those juicier profiles.

83
00:05:58,540 --> 00:06:00,240
And that was kind of it from the beginning.

84
00:06:00,240 --> 00:06:05,320
I was like, you know, had no real grand plans to actually do a talk like this.

85
00:06:05,320 --> 00:06:08,900
It was actually just for shits and gigs in the sidelines.

86
00:06:09,100 --> 00:06:12,120
And then just sit and wait, essentially, see how long it took.

87
00:06:13,500 --> 00:06:23,960
And then at some point, we're like, oh, yeah, wouldn't it be awesome if we actually just went after the DEF CON community instead of just any random community for a few reasons.

88
00:06:24,860 --> 00:06:41,120
So as you're probably aware, if you are all those people who are in the DEF CON subreddit, there's also a whole bunch of other social media platforms that the DEF CON attendees like to use, such as the forum, such as the Discord, and yeah, and the Reddit group.

89
00:06:41,220 --> 00:06:45,880
So all of those things kind of is really helpful because they all have great history.

90
00:06:45,880 --> 00:06:47,480
There's heaps of discussion on there.

91
00:06:47,480 --> 00:06:50,240
There's all different types of people on there, different skill sets.

92
00:06:51,280 --> 00:06:56,500
And one really awesome thing about you hackers is you like to reuse your handle everywhere.

93
00:06:56,500 --> 00:07:00,100
So that was really good to kind of like just track people across those three different areas.

94
00:07:00,640 --> 00:07:02,320
And also it's on topic.

95
00:07:03,260 --> 00:07:08,940
But yeah, for the rest of the brief, though, I actually just focused on the subreddit.

96
00:07:10,900 --> 00:07:12,620
So yeah, where the hell do I start?

97
00:07:12,620 --> 00:07:25,760
I was like, it's one thing to sort of come up with this idea that wouldn't it be funny to scrape a whole bunch of social media, but without any kind of idea of what to do or where to find the information, what am I even looking for?

98
00:07:26,020 --> 00:07:32,080
I was kind of like in a bit of a pickle in terms of actually starting.

99
00:07:35,260 --> 00:07:40,160
So yeah, we're all kind of aware of what PII is, personal identifiable information.

100
00:07:40,160 --> 00:07:49,980
I was pretty full bottle on what that kind of meant, you know, your name, date of birth, social security number you guys have, all that kind of like basic stuff.

101
00:07:49,980 --> 00:07:52,160
I was pretty sure I could find that kind of thing.

102
00:07:52,380 --> 00:07:57,060
But what about all the other stuff that people leak in terms of like their own personal information?

103
00:07:58,020 --> 00:08:09,500
It was kind of a pretty wide scope and I didn't really have much direction as to how to sort of define it or where to find it or any kind of like ability to do that at that stage.

104
00:08:10,800 --> 00:08:26,720
Fortunately, back home at my company, I had a guy who's sitting right next to me who has a PhD in cognitive behaviour and sort of is a proper scientist regarding that.

105
00:08:26,720 --> 00:08:35,360
And he was actually researching at the time the tendencies, the language and the behavioural patterns of people preceding OPSEC incidents.

106
00:08:35,540 --> 00:08:41,560
And so I was like, Riley, tell me how to find this stuff or is there any manuals?

107
00:08:41,560 --> 00:08:43,400
Tell me how to like search for it.

108
00:08:44,360 --> 00:08:59,640
And he pointed me down the track of, you guys have some amazing resources here in America, your three-letter agencies like CIA, FBI, the National Centre for Counterintelligence all have really great manuals and doctrine online.

109
00:08:59,640 --> 00:09:01,560
And I was like, that's a really great place to start.

110
00:09:03,220 --> 00:09:12,380
And not, sorry, not forgetting the social engineering community here was a really awesome treasure trove of sort of ways to find individuals and sort of target individuals.

111
00:09:12,380 --> 00:09:22,060
So I used all that, all those kind of resources to kind of guide me in how I was going to find this other kind of information beyond PII.

112
00:09:23,920 --> 00:09:39,660
And yeah, I came to the, they were all sort of pointing to the idea that there are these sort of like vulnerable, sort of personal vulnerability traits or more easily manipulated life situations or sort of like leverageable information.

113
00:09:40,660 --> 00:09:43,160
Yeah, like, you know, the Russians call it kompromat.

114
00:09:43,480 --> 00:09:45,500
And I was like, ooh, kompromat.

115
00:09:45,500 --> 00:09:53,880
I think I'll add that to my list of things I'm looking for beyond the sort of normal names and addresses and things like that.

116
00:09:54,440 --> 00:10:04,960
And my kind of simple mind could sort of track that as a kind of like easy way for an attacker or a malicious actor to search for, you know, stuff to find like a soft target.

117
00:10:04,960 --> 00:10:07,100
So I was like, that, that makes sense.

118
00:10:09,660 --> 00:10:19,460
Yeah, so I came up with this idea or this term called PVI, personal vulnerability information, you heard it here first, as opposed to traditional PII.

119
00:10:20,140 --> 00:10:36,240
Traditional PII is just those like super basic attributes that we're all kind of aware of and that organizations and governments, government policy like to, or have in their, in their policies to protect.

120
00:10:37,120 --> 00:10:40,980
And your standard security awareness programs also have this stuff in it as well.

121
00:10:40,980 --> 00:10:43,260
It's mostly, you know, don't put this stuff online.

122
00:10:43,400 --> 00:10:59,800
But if you compare that to all the other stuff that are people leaking, and, you know, isn't necessarily protected by this legislation or policy, you kind of have to ask yourself, like, which one do I, which one would I prefer to see in a leak?

123
00:10:59,840 --> 00:11:03,820
And also, which one am I actually leaking myself all the time?

124
00:11:07,240 --> 00:11:17,040
See, I was beginning to see, I was beginning to learn that PII is sort of like just scratching the surface in terms of like online targeting of people.

125
00:11:18,780 --> 00:11:28,640
Yeah, and I was really realizing that like the real bad dudes out there are actually searching for people who maybe will leak rather than people who already have leaked.

126
00:11:29,460 --> 00:11:45,160
And so if you think about all those things on that original slide, I was talking about all those like PII versus PVI, all this stuff that you might be in the blue column over here, have this like, provides an attacker with options to target you.

127
00:11:47,720 --> 00:11:52,100
So I was like, who are these bad dudes who are using this kind of stuff?

128
00:11:52,100 --> 00:11:58,000
And in all of the research that I was doing, I was realizing that, you know, it's not just the usual suspects that we're talking about here.

129
00:11:58,000 --> 00:12:00,940
I'm not just talking about APTs and hacktivists.

130
00:12:00,940 --> 00:12:24,920
We've all heard the stuff about, you know, online radicalization of young jihadis and things like that, which actually also adding, you know, stalkers, jealous ex-partners, child predators, all of these other malicious actors online who are using the exact same methodology of trying to find vulnerable targets using the language that they're presenting online.

131
00:12:28,740 --> 00:12:29,840
Yeah, so I had the what.

132
00:12:29,840 --> 00:12:30,920
I was like, okay, that's great.

133
00:12:30,920 --> 00:12:37,540
I'll be searching for actual facts about people and then kind of like personality types, language use, and those kind of things.

134
00:12:37,540 --> 00:12:45,260
Now, how the hell do I find that stuff on, you know, large, large space of information?

135
00:12:46,040 --> 00:12:50,920
So if you've been working in cyber security for a minute, you'd be aware of the intrusion kill chain.

136
00:12:51,120 --> 00:12:52,640
Just wanted to put that up there.

137
00:12:52,900 --> 00:12:59,220
We are definitely in the reconnaissance phase here, just researching, identifying, and selecting targets.

138
00:12:59,540 --> 00:13:02,220
And I was kind of trying to work out like how I was going to do this.

139
00:13:02,220 --> 00:13:12,060
So I just decided to smoosh all this kind of like, all these different organizations, methodologies into one kind of, into one, like my plan.

140
00:13:12,360 --> 00:13:23,500
So use the CIA's methods of like spot assess and develop, and also mixed in a bit of like the military concept of find, fix, finish.

141
00:13:23,500 --> 00:13:41,000
And using those things, using those resources online, like, particularly want to add the Department of Justice has awesome annals of, you know, cases here in America, like they kind of document exactly how espionage cases unfolded and how those people were approached online.

142
00:13:42,060 --> 00:13:50,640
Yeah, and I came up with the idea of like find, fix, follow, which I realize is just, it's just pretty rudimentary, but sorry, fix, find, follow.

143
00:13:50,940 --> 00:13:55,460
So I fixed, chose the haystack, chose the DEF CON subreddit.

144
00:13:56,180 --> 00:14:07,180
I was going to find those people using those terms amongst the anonymous users, anonymous users, and then assess for leakage, leakability, and tag those people for monitoring.

145
00:14:07,720 --> 00:14:15,640
And really at this stage, I was just doing this pretty rudimentary, just control F, you know, finding, finding words.

146
00:14:15,740 --> 00:14:18,040
So it's pretty basic at this stage.

147
00:14:23,630 --> 00:14:26,730
Yeah, so sort of coming to the part where I started, started doing it.

148
00:14:26,730 --> 00:14:38,090
So I had to have some assumptions to start with, was deciding that like people who reveal in DEF CON, sorry, Reddit slash DEF CON will reveal elsewhere is a pretty basic kind of assumption.

149
00:14:39,430 --> 00:14:43,690
If you're not aware, the DEF CON Reddit is pretty small.

150
00:14:43,690 --> 00:14:48,630
It's pretty small traffic, low traffic, sorry, compared to other subreddits around there.

151
00:14:48,630 --> 00:14:52,030
And it's going to be a pretty small percentage of those users posts.

152
00:14:52,550 --> 00:14:59,490
So I was like, okay, if I find situations in that, I'm just going to track where else they're posting and try and find more information from that point.

153
00:15:00,470 --> 00:15:03,710
Yeah, people who reveal big things probably do small ones also.

154
00:15:03,710 --> 00:15:04,850
That's a no-brainer.

155
00:15:05,010 --> 00:15:06,530
There is going to be old information.

156
00:15:06,530 --> 00:15:16,990
It's very old, you know, it's not very old, but as years go by, there's going to be stuff where people have changed, changed the information that they're presenting.

157
00:15:17,710 --> 00:15:24,810
And of course, there will be deception because, you know, I realise you guys, you hackers are slippery bastards.

158
00:15:25,250 --> 00:15:33,190
So I was really just searching for things like I work for, my email address is, my name is, and I live in.

159
00:15:33,190 --> 00:15:40,790
I realise that's pretty amateur hour, but that's just kind of like the small brain position I was in to start with.

160
00:15:43,490 --> 00:15:46,730
Yeah, scraping was pretty slow, had limited results.

161
00:15:46,730 --> 00:15:51,590
I was really only finding one, two, three things and sort of just slowly plodding away.

162
00:15:51,790 --> 00:15:53,410
And I was limited to those keywords.

163
00:15:53,410 --> 00:16:02,370
Well, I'm not limited to those keywords, but I was limited by those keywords, just searching for those few things that I could think of as what would be good strings to search for.

164
00:16:02,390 --> 00:16:05,230
I wasn't finding the kind of stuff I was after.

165
00:16:05,230 --> 00:16:09,810
And at that stage, I was like, I'm not really hacking this.

166
00:16:09,810 --> 00:16:11,750
I'm really just pushing forward.

167
00:16:12,570 --> 00:16:27,670
But then, you know, GPT entered the chat and I started to ask it to help me define those leaky or vulnerable language and really open up those keywords to much more things than I could consider as possible things I wanted to search for.

168
00:16:28,550 --> 00:16:34,630
And still some manual sifting needed then to, after I found those initial targets through posts.

169
00:16:34,730 --> 00:16:37,670
But those juicy individuals started to grow.

170
00:16:37,670 --> 00:16:45,450
At this stage, I think I had like two or three people with one or two little posts where I was like, okay, I think I'm starting to develop something here.

171
00:16:47,690 --> 00:16:57,890
And yeah, so if you've used chat GPT in the past or currently, you probably know it does require a bit of sort of convincing to get it going.

172
00:16:57,890 --> 00:17:03,810
I had to kind of convince it I was reviewing my own chat logs, make sure I wasn't revealing any personal information.

173
00:17:04,330 --> 00:17:10,290
And I was quite happy to come up with, you know, things that I should be looking for to try and find PII and PVI.

174
00:17:10,970 --> 00:17:12,090
And I was like, awesome.

175
00:17:12,870 --> 00:17:16,570
But can you give me the sort of like words that might precede a leak?

176
00:17:16,570 --> 00:17:21,610
That would be a bit more helpful than just giving me the concept of the kind of things that people are looking for.

177
00:17:23,450 --> 00:17:27,210
Yeah, and I was like, okay, can you give me a hundred different versions of that?

178
00:17:27,710 --> 00:17:33,350
Which was really helpful because then I could sort of just still doing my control F, still looking through.

179
00:17:33,550 --> 00:17:46,310
At some stage, I discovered there's like a year ago, it was still functioning, was a push shift IO, which is like a website we can do live queries of Reddit data.

180
00:17:46,570 --> 00:17:55,190
And I was just really just mandrolically going through and inserting these keywords in there and trying to search and then going back onto the subreddits trying to find the information.

181
00:17:55,690 --> 00:17:57,630
It was still very, very clunky.

182
00:17:57,630 --> 00:18:07,070
As you can imagine, some of these keywords, some of these sort of preceding strings aren't exactly the kind of language that people are using in the DEF CON subreddit.

183
00:18:08,030 --> 00:18:12,090
I received an email from, it's just not the language that's really in there.

184
00:18:12,250 --> 00:18:15,150
But I was finding some stuff here and there.

185
00:18:18,100 --> 00:18:34,000
Yeah, everything was going pretty great until, if you're a user of Reddit or if you're aware, Reddit halfway through last year or halfway through this year just decided to cut all access to third-party apps and my super powerful search tool just kind of went dark.

186
00:18:34,640 --> 00:18:46,060
And I was left sort of... motivation died, regretted not finishing earlier, not producing any slides because I had already actually sent this presentation to DEF CON.

187
00:18:46,060 --> 00:18:48,640
So I was like, had nothing.

188
00:18:49,120 --> 00:18:51,860
And I was kind of like, six weeks of just like, whatever.

189
00:18:51,860 --> 00:18:52,980
It was a smart idea.

190
00:18:52,980 --> 00:18:55,300
Well, I thought it was a smart idea to begin with.

191
00:18:55,800 --> 00:19:00,040
And I was like, it's probably not going to be possible now.

192
00:19:01,440 --> 00:19:12,180
But then, from total failure, I realised that some saint had been archiving Reddit data since the beginning of time.

193
00:19:12,180 --> 00:19:21,500
So there was literally every single post, every single comment that had ever been made in Reddit in one gigantic text file.

194
00:19:22,580 --> 00:19:23,960
And so I was like, sweet.

195
00:19:24,040 --> 00:19:26,880
50,000 comments, 6,000 posts.

196
00:19:26,960 --> 00:19:29,160
Now I wasn't just looking at the last 12 months.

197
00:19:29,160 --> 00:19:31,080
I had 18 years worth.

198
00:19:31,540 --> 00:19:36,720
The Reddit... sorry, the DEF CON Reddit doesn't go back that far.

199
00:19:36,720 --> 00:19:47,240
But now I was really just looking at a decade's worth of comments, which gave me a much more big... sorry, a much larger data set to look through.

200
00:19:50,560 --> 00:19:51,220
Yeah, sorry.

201
00:19:51,220 --> 00:19:55,860
So at that stage, I was... yeah, so I'm... sorry, it's kind of...

202
00:19:57,840 --> 00:20:01,840
one of the problems with doing one of these briefs is you've got to remember what you were doing 12 months ago.

203
00:20:03,100 --> 00:20:04,860
Now I was working with a text file.

204
00:20:04,860 --> 00:20:16,440
So now I'm just really just searching through a text file, you know, using Python to just search for strings within the text file, rather than looking on Reddit online, using the search.

205
00:20:18,880 --> 00:20:24,460
But yeah, at some point during the year, I was listening to a Darknet Diaries episode, big Darknet fan.

206
00:20:24,700 --> 00:20:28,860
And I was listening to Sam Bent, who he presented last year at DEF CON.

207
00:20:28,920 --> 00:20:37,620
He was talking about how he was using... he was trying to circumnavigate the US Postal Service to send drugs through the mail.

208
00:20:37,620 --> 00:20:39,820
And he was trying to work out how he was going to do it.

209
00:20:39,820 --> 00:20:49,040
And he was like, well, one of the, you know, one of the things, best things about the Postal Service is they have a manual online for what good looks like, what they're searching for.

210
00:20:49,040 --> 00:20:51,160
And he was like, I'll just do the opposite.

211
00:20:51,460 --> 00:20:55,220
And I was driving along, I was like, I think I could just do the opposite.

212
00:20:56,820 --> 00:21:10,280
So I went online to the Australian government website to try and find like the kind of attributes that they are looking for when they're trying to give people security clearances, or they're looking to give people security clearances.

213
00:21:10,560 --> 00:21:16,020
And I was looking for attributes for people in that sense, because I was like, well, that's good, that's safe.

214
00:21:16,020 --> 00:21:25,240
Those are the things that they think people will present as, you know, being able to keep a secret or being able to, you know, keep the nation's secrets.

215
00:21:25,880 --> 00:21:30,920
And I was like, hey, ChachiPT, give me all the synonyms for all of those things.

216
00:21:31,660 --> 00:21:34,780
And then also give me all the antonyms for all of those as well.

217
00:21:34,780 --> 00:21:41,160
So I came up with a huge list of antonyms of like the types of people that I was looking for online.

218
00:21:42,600 --> 00:21:46,920
And then I was like, find me all variations of that.

219
00:21:46,920 --> 00:21:57,740
So sort of like just punched out hundreds and hundreds of different sentences, which might suggest the type of person who presents those kind of characteristics.

220
00:21:58,980 --> 00:22:09,180
And yeah, now I really felt like I was cooking with gas, because I was not only searching for these things, I was searching for sort of permutations or variations of those, just little chunks of those.

221
00:22:09,180 --> 00:22:13,580
And I was really finding a lot of results using this method.

222
00:22:16,540 --> 00:22:18,780
Yeah, so I also wanted to find those leak facts.

223
00:22:18,780 --> 00:22:21,560
So thinking back to the beginning, I was also trying to find PII.

224
00:22:22,060 --> 00:22:36,960
So, you know, from a slow and sort of like noisy scrape, I was now using those kind of that, you know, GPT enhanced powerful semantic search to try and find words, which sort of seemed like the person was either going to leak or had leaked.

225
00:22:37,300 --> 00:22:40,420
The search method didn't change for using the PII.

226
00:22:42,060 --> 00:22:45,300
I just had to obviously assess the leak's impact.

227
00:22:45,300 --> 00:22:51,020
So some stuff that people are leaking is really just sort of basic stuff, like down in this green area.

228
00:22:52,440 --> 00:22:59,500
You know, I just put this together to try to like kind of visualise how I was assessing the kind of things that people are leaking.

229
00:22:59,500 --> 00:23:06,400
And it's kind of what I call this death by thousand likes, because lots of the stuff, sure, whatever, it's just online.

230
00:23:06,400 --> 00:23:10,220
But you know, when you add it all up, it really is and has an accumulative effect.

231
00:23:12,500 --> 00:23:16,300
Yeah, so I had to assess the impact, but also like the frequency.

232
00:23:16,300 --> 00:23:24,240
If you're, you know, if you have 500 comments in the sort of green, yellow area, maybe that's sort of only worth like one comment in the red area.

233
00:23:24,480 --> 00:23:27,300
There's still heaps of error, as you're probably aware.

234
00:23:27,700 --> 00:23:32,620
Chachipiti made a lot of things up a lot of the time and it was still very human-driven.

235
00:23:32,620 --> 00:23:34,940
I was still searching through and finding.

236
00:23:35,400 --> 00:23:42,520
I still had to go through and look through the results and actually kind of prove or disprove whether it was actually what I was looking for.

237
00:23:45,380 --> 00:23:48,060
Yeah, as you can probably imagine, 18 years is a long time.

238
00:23:48,200 --> 00:23:54,980
I had lots of false positives, lots of like comments that people were making and I was just like, that's nothing.

239
00:23:55,340 --> 00:23:57,040
So that was taking quite a lot of my time.

240
00:23:57,040 --> 00:24:03,840
It was taking quite a lot of energy and it was pretty boring looking through lots and like hundreds of false positives.

241
00:24:03,920 --> 00:24:05,500
There's lots of non-repeat offenders.

242
00:24:05,500 --> 00:24:14,500
So like you might make one mistake like one time 10 years ago and post something about yourself and I just sort of like wasn't interested in that.

243
00:24:15,440 --> 00:24:17,780
And there was lots of target drop-off as well.

244
00:24:17,780 --> 00:24:22,460
So like lots of the people I was looking at, lots of the results was happening a long time ago.

245
00:24:22,460 --> 00:24:25,600
Those people might have just like moved on and not even been into security anymore.

246
00:24:25,600 --> 00:24:27,340
So I kind of wasn't interested in that.

247
00:24:27,940 --> 00:24:31,440
The one thing I want to say about that is like it's only bad in this context, right?

248
00:24:31,440 --> 00:24:36,500
So I'm trying to find people, I'm trying to produce slides that are interesting for you guys to look at.

249
00:24:36,500 --> 00:24:48,980
If it's a malicious actor, if it's like those stalkers or predators or whatever or more advanced enemies where like you need to imagine like they only need one slip-up or one thing that they're looking for.

250
00:24:49,100 --> 00:24:51,040
So it's only bad in this context.

251
00:24:52,540 --> 00:24:55,080
Yeah, and today we're only going to look at high-quantity packs.

252
00:24:55,080 --> 00:24:58,660
I didn't really put anything together where someone had just made one slip.

253
00:25:00,820 --> 00:25:06,280
But I was also thinking now I'm actually, you know, now I've got big brain, big brain mode.

254
00:25:07,080 --> 00:25:14,740
And then, yeah, sort of like a couple of months ago, as you're probably aware, OpenAI gave us access to the API.

255
00:25:15,120 --> 00:25:16,380
So this was huge.

256
00:25:16,380 --> 00:25:19,660
I was like, whoa, god mode.

257
00:25:19,860 --> 00:25:31,940
I was now able to just plug the data, the massive text file of all of those comments straight into the ChatGPT prompt and just literally ask it questions about the data.

258
00:25:35,320 --> 00:25:43,380
Yeah, so as I said at the start, this is meant to be like a no prior knowledge brief and I certainly had no prior knowledge before I started it.

259
00:25:43,380 --> 00:25:47,720
But what it meant was I didn't have to produce any search scripts anymore.

260
00:25:48,220 --> 00:25:53,460
I was no longer in charge of coming up with creative versions of the kind of stuff I was looking for.

261
00:25:53,780 --> 00:25:55,540
I no longer needed to.

262
00:25:56,140 --> 00:25:57,440
Yeah, sorry I said that.

263
00:25:58,080 --> 00:26:02,620
I just left it all up to the large language model, which was awesome.

264
00:26:03,800 --> 00:26:05,740
It was really limited only by tokens.

265
00:26:05,740 --> 00:26:11,480
If you're not aware, when you're asking it questions, you are limited by a number of characters in and out.

266
00:26:11,480 --> 00:26:13,560
But that really only just slowed me down.

267
00:26:13,560 --> 00:26:18,200
I just had to like chunk up the data into smaller pieces and re-ask the questions over and over.

268
00:26:19,380 --> 00:26:36,460
And in the meantime, since creating this brief and OpenAI forging ahead with its speed of technology now, that characters that you get in and out has multiplied by a factor of I think like six or eight.

269
00:26:36,460 --> 00:26:41,220
It's huge number of... a huge token limit now.

270
00:26:42,860 --> 00:26:46,740
Yeah, so now that's an example up there of the kind of things I was asking.

271
00:26:46,740 --> 00:26:56,540
I was like, okay, so hey, this data set that I have is a huge list of... it's a play of people who are preparing to go to a hacking conference.

272
00:26:56,880 --> 00:27:00,540
Some people in there are really awesome, saying really positive things.

273
00:27:00,540 --> 00:27:07,800
And some people in there are kind of like exhibiting narcissistic or psychopathic kind of like traits.

274
00:27:07,920 --> 00:27:10,040
Give me a list of the top five.

275
00:27:10,160 --> 00:27:11,340
And it was just like an instant.

276
00:27:11,340 --> 00:27:12,160
There you go.

277
00:27:12,580 --> 00:27:17,580
Here's your five top targets, which was sweet.

278
00:27:19,560 --> 00:27:27,180
I was like, okay, so in the data, give me a list of users that are suggesting the propensity to commit crimes.

279
00:27:29,770 --> 00:27:31,310
Show me where they live.

280
00:27:31,810 --> 00:27:38,710
It wasn't actually showing me where they live, but it was sort of saying like show me all the users that talk about a town name, city name, or anything like that.

281
00:27:38,710 --> 00:27:48,630
So if you can imagine comparing to back when I was previously just using Python to search through, I would have had to look through in terms of like a dictionary search of every single town, every single place.

282
00:27:49,130 --> 00:27:54,490
But obviously, yeah, you know, just knows and goes, here are all the people that say where they live.

283
00:27:59,020 --> 00:28:01,080
Show me users that suggest fear.

284
00:28:04,490 --> 00:28:05,430
Yeah, this is an interesting one.

285
00:28:05,430 --> 00:28:12,810
Show me for a specific user over the last 10 years, when they're online, when they make their posts.

286
00:28:18,280 --> 00:28:26,480
And finally, I was like, show me anyone who's looking for anything, who has posted anything about working on anything special or out of the ordinary at their work.

287
00:28:31,970 --> 00:28:41,030
Yeah, so I guess the concept of the speech that I'm trying to get across is that the barriers of entry are completely gone.

288
00:28:41,450 --> 00:28:45,530
All those questions that I was asking in the beginning of how am I going to do it?

289
00:28:46,170 --> 00:28:48,250
You know, what even search terms am I looking for?

290
00:28:48,250 --> 00:28:48,950
What comments?

291
00:28:48,950 --> 00:28:50,990
What about things that don't match the terms?

292
00:28:51,210 --> 00:28:53,310
How do I do it at scale?

293
00:28:53,550 --> 00:28:55,510
Where is it centralised?

294
00:28:55,510 --> 00:28:57,210
And when to do it?

295
00:28:57,210 --> 00:29:06,730
All of that stuff is completely gone because of all of these things that make up what we would consider like the capabilities required to do that.

296
00:29:06,970 --> 00:29:17,930
They're all inherent and anyone here can literally go online and do online targeting of people with absolutely no experience whatsoever.

297
00:29:21,770 --> 00:29:25,410
Yeah, in some of my rehearsals, someone was like, oh, can't you TP4 just do this now?

298
00:29:25,410 --> 00:29:28,530
Like, you use plugins that you can just connect to the internet and you can just do it.

299
00:29:28,530 --> 00:29:33,930
But unfortunately, it's quite resistant to this kind of these kind of questions.

300
00:29:34,130 --> 00:29:37,790
No matter how hard I tried, TP4 is not happy to do it.

301
00:29:38,730 --> 00:29:48,490
So, you know, getting the data, saving the data down, getting it offline, putting it onto your computer, and then using the prompts to scan through it, much more successful.

302
00:29:53,860 --> 00:30:01,680
And now, let's talk about, I think I've got like five or six target packs here that I'm going to talk through.

303
00:30:02,060 --> 00:30:07,200
Before I start, I just want to have to make a few caveats to the things that we're going to look at.

304
00:30:07,460 --> 00:30:17,140
Certainly, this is not a representation of DEF CON attendees at all, like just like, you know, any subreddit, anyone in the world can join up.

305
00:30:17,440 --> 00:30:19,440
I think it has like 30,000 members.

306
00:30:19,760 --> 00:30:24,320
So it's definitely not a representation of you guys here today.

307
00:30:26,160 --> 00:30:30,040
What were the other... one second, sorry.

308
00:30:32,600 --> 00:30:35,340
Had some very important caveats I wanted to make.

309
00:30:36,180 --> 00:30:36,840
Oh yeah, sorry.

310
00:30:36,840 --> 00:30:45,860
So the people who are up here obviously have their own, make their own decisions, they have their own threat model, their own thresholds of like what they want to reveal online.

311
00:30:46,140 --> 00:30:50,420
They probably have their own understanding of like how much they have revealed.

312
00:30:51,480 --> 00:30:53,940
Oh yeah, and one other thing is on the fly out over here.

313
00:30:53,940 --> 00:30:57,760
I had a lot more information on these slides about what people had posted.

314
00:30:57,840 --> 00:31:07,380
But I had kind of like a moment of clarity where I was like, well, you know, if I put up any one thing that they've revealed, that kind of reveals everything.

315
00:31:07,380 --> 00:31:10,160
So I did a lot, a lot of blurring on the way over here.

316
00:31:10,160 --> 00:31:17,180
So I apologize if it's a little too blurry for the kind of blood that you wanted to see.

317
00:31:17,440 --> 00:31:22,820
But you know, people's privacy also in an area like this is important.

318
00:31:23,000 --> 00:31:24,760
Anyway, let's get on with it.

319
00:31:25,540 --> 00:31:34,820
So the first guy or girl, the AI hit four email addresses containing first name, last name.

320
00:31:35,360 --> 00:31:38,060
I wanted to see talk about any illegal activity.

321
00:31:38,060 --> 00:31:47,240
I asked it to find comments about suffering and also any kind of, you know, device information or models of modems and things like that.

322
00:31:47,240 --> 00:31:50,480
So the first person, you know, quite a lot of hits here.

323
00:31:51,340 --> 00:31:54,280
Router brand model, drug use.

324
00:31:54,700 --> 00:31:57,980
You know, I'm not here to pass judgment on anyone using drugs or whatever.

325
00:31:57,980 --> 00:32:16,700
But you know, if you are going to post that in an online forum, you know, with all kinds of actors looking at you, judging whether you're a target, and you're going to kind of reveal that there are moments of vulnerability about your life, that could be a valuable data point.

326
00:32:17,280 --> 00:32:21,800
This person's right into conspiracies, you know, very easily manipulated, I would imagine.

327
00:32:22,840 --> 00:32:24,320
Crypto bro, yep.

328
00:32:24,800 --> 00:32:38,320
And the other thing was, yeah, so like, one thing if you think about like, malicious actors trying to contact people or trying to like, have a pathway to conversation, they might have to choose between a bunch of like, their hobbies or interests.

329
00:32:38,840 --> 00:32:43,260
You know, a shortcut or a cheat code to that is like, if they're actually seeking a mentor.

330
00:32:43,260 --> 00:32:49,200
So this person, you know, not too much here, but quite a lot of attack pathways.

331
00:32:53,090 --> 00:32:59,710
Yeah, so this one, I actually pulled a lot down because this person was actually posting a fair bit about themselves.

332
00:32:59,710 --> 00:33:03,550
But this one's going along the lines of like, that's the stalker kind of threat.

333
00:33:03,910 --> 00:33:20,870
So the individuals posted, you know, their home location, where they live, but also, you know, their name, age, backstory, quite a painful past, lots of like, experiences of racism, things like that, that an attacker could look kind of used to build rapport.

334
00:33:20,950 --> 00:33:25,210
But the other thing they mentioned was they're traveling alone to DEF CON.

335
00:33:25,490 --> 00:33:45,870
And was, we're going to be at location ABC, you know, and if you're, you know, a female traveling alone, and you're going to reveal everything about yourself and your backstory, as well as, you know, positive ID of what you look like, it's quite a lot of stuff for us quite a lot of sort of opportunities for someone who's trying to,

336
00:33:45,870 --> 00:33:49,450
you know, go after you to to have, you know, in their back pocket.

337
00:33:53,050 --> 00:33:54,510
Yeah, this one's interesting one.

338
00:33:54,510 --> 00:34:00,310
So this one, ChachiPT hit for weekend activities, small towns and use their first name.

339
00:34:00,310 --> 00:34:01,650
So it wasn't a lot of things.

340
00:34:01,650 --> 00:34:11,130
But you know, after I kind of was like triaging through, this is kind of like a very efficient and minimalist kind of pathway to an attack.

341
00:34:11,130 --> 00:34:17,670
So if you're going to talk about the city or state that you live in, particularly this one, I think that the population was like under 100,000.

342
00:34:18,010 --> 00:34:22,630
But also talk about a place of interest that you and your family visit on a frequent basis.

343
00:34:22,770 --> 00:34:29,390
And finally, talk about your vehicle description, right down to, you know, decals and stickers on the back of it.

344
00:34:29,390 --> 00:34:35,310
If someone, you know, had that kind of information, they wanted to sort of get close to you or get close to your house.

345
00:34:35,910 --> 00:34:41,750
It's all they would really need to target you from an individual perspective.

346
00:34:42,410 --> 00:34:42,870
Yep.

347
00:34:42,870 --> 00:34:43,890
Breed of pet.

348
00:34:45,010 --> 00:34:45,790
Sure.

349
00:34:45,790 --> 00:34:48,370
Some of the slides have that kind of stuff in there.

350
00:34:48,370 --> 00:34:54,770
Not just talking about, you know, those standard don't do security things like tell people your secret questions.

351
00:34:54,930 --> 00:35:01,090
But if you're going to talk about your breed of pet, I think this is like the French Bulldogs.

352
00:35:01,090 --> 00:35:07,130
Something to do with like every pet I've seen in DEF CON is French Bulldogs.

353
00:35:07,590 --> 00:35:12,990
But if you're going to talk about that, you know, you're opening yourself up to pretty effective spear phishing and things like that.

354
00:35:13,030 --> 00:35:15,610
Person's also very open to online comms.

355
00:35:19,780 --> 00:35:23,040
Yeah, so this guy hit for got a security clearance.

356
00:35:23,040 --> 00:35:25,180
I live in X location and ranks.

357
00:35:25,180 --> 00:35:33,140
So this is a military person, you know, posted to a certain location, revealing their career goals.

358
00:35:33,140 --> 00:35:42,160
You know, if the attacker is thinking about, you know, everyone has something that they want or that they're going after, if you're going to reveal your goals, that's kind of like an easy kind of offering for you.

359
00:35:42,820 --> 00:35:54,320
Talk about their career timeline, grievances at work, military rank, and their workplace and their openness and their kind of like approach to authority.

360
00:35:54,420 --> 00:35:56,820
So quite a lot of stuff here.

361
00:35:57,100 --> 00:36:04,220
I realised in America, you guys have a little bit of different approach to how you conceal security clearances and things like that.

362
00:36:04,640 --> 00:36:09,860
In Australia, it's fairly, it's just very different.

363
00:36:09,860 --> 00:36:13,760
It's a different approach to how you manage your security clearance.

364
00:36:18,860 --> 00:36:21,180
Yeah, so this one had quite a fair bit.

365
00:36:21,180 --> 00:36:30,140
So I was looking for, I found things for gain seniority, specific device specifications, and things that they were struggling with.

366
00:36:30,140 --> 00:36:42,220
So, you know, the AI found a quite a lot of things across the post from this person, particularly the pet name, breed, became a senior manager quite quickly.

367
00:36:42,500 --> 00:36:56,040
If anyone here is, you know, just to expand on that a little bit, I imagine that probably people here who work for companies that have, you know, significant IP that they want to protect, or maybe you want to work one day for companies that have IP they want to protect,

368
00:36:56,040 --> 00:37:05,820
you know, moving up in management in those companies is probably something that an attacker would be looking for if they're trying to assess your ability, what kind of stuff you know about that company.

369
00:37:07,360 --> 00:37:09,100
Yeah, revealing router brand model.

370
00:37:09,100 --> 00:37:10,840
I'm not 100% sure how bad that is.

371
00:37:10,840 --> 00:37:12,200
You guys would probably tell me.

372
00:37:13,040 --> 00:37:16,380
Personal demons, where they live, and also painful past.

373
00:37:16,380 --> 00:37:23,520
The person experienced quite a lot of bullying, and particularly the number of posts with regards to this was pretty high.

374
00:37:23,520 --> 00:37:34,680
So, you know, in terms of like building rapport and kind of like getting alongside this person, I would imagine that, you know, an advanced attacker would probably have a pretty good chance from that angle.

375
00:37:36,420 --> 00:37:39,580
And last but not least, yeah, this is probably one of my favourites.

376
00:37:40,020 --> 00:37:45,800
This guy posted one time in the entire 18 years of the subreddit.

377
00:37:46,140 --> 00:37:51,260
Posted one time, first name, last name at whatever.com.

378
00:37:51,800 --> 00:37:53,560
And I just googled that.

379
00:37:53,780 --> 00:37:58,020
And straight up, pitches himself at work, ID card.

380
00:37:58,460 --> 00:38:05,000
You know, I'm not assuming that this is any kind of like high security area, and that this is a massive security breach.

381
00:38:05,000 --> 00:38:15,800
But you know, if you only just go over to the to the physical village, you can probably you can get some information about like, just having the picture of the ID card usually will get you like you can create an imitation ID card.

382
00:38:15,800 --> 00:38:23,340
That's like a pretty good bit of information to have if you're trying to get into a facility.

383
00:38:23,620 --> 00:38:26,740
So yeah, don't post pictures of your ID card.

384
00:38:31,240 --> 00:38:43,380
Yeah, so those individual target packs only revealed small things individually, what five or six different things.

385
00:38:43,880 --> 00:38:51,200
But I kind of encourage you to kind of think about your own social media, not just your Reddit posts, but every other social media that you use.

386
00:38:51,440 --> 00:39:08,720
And think about whether, you know, you had more than or accumulated measures, accumulated version of those target packs, and how much information a person would have if they were to just use something like AI, spend a bit of time and create, you know,

387
00:39:08,720 --> 00:39:14,120
create a target pack surrounding you or, you know, individuals that you love or colleagues.

388
00:39:15,980 --> 00:39:18,720
This isn't a real person, that's an AI guy.

389
00:39:19,520 --> 00:39:22,280
But yeah, that's the end of the speech.

390
00:39:22,280 --> 00:39:23,760
Does anyone have any questions?

391
00:39:35,930 --> 00:39:41,310
So did you validate the data that you get it from chatJPT?

392
00:39:41,310 --> 00:39:42,730
So is it real?

393
00:39:42,730 --> 00:39:47,530
Is the data in the data set or chatJPT just imagined it?

394
00:39:49,650 --> 00:39:51,470
Sorry, I think I understood your question.

395
00:39:51,470 --> 00:40:07,010
You're saying, did I validate whether it was accurate, what it was finding versus... yeah, so as you sort of saw in some of those slides, had like what the output of the search was in like the black terminal kind of section.

396
00:40:07,010 --> 00:40:14,050
But then I went back on to Reddit to kind of grade screenshots, make it a bit more user-friendly in terms of a speech.

397
00:40:14,050 --> 00:40:23,430
I did go back, had a look at what the comment was in context, and that kind of, but you know, it just added to the amount of time that I had to spend doing it.

398
00:40:23,430 --> 00:40:26,970
But yeah, that required human interaction.

399
00:40:27,570 --> 00:40:32,850
It was not really possible to just have it completely automated.

400
00:40:33,650 --> 00:40:35,450
Does that answer your question?

401
00:40:39,680 --> 00:40:40,820
Any others?

402
00:40:49,310 --> 00:40:49,910
Okay, yeah.

403
00:40:49,910 --> 00:40:54,390
So the question was, what was the most common leak that you saw over and over again?

404
00:40:57,260 --> 00:41:07,380
Yeah, so these are like, I chose these because they're like the best ones up there for kind of like a broad spectrum of things that people were leaking.

405
00:41:09,940 --> 00:41:12,700
Didn't see a lot of like personal names.

406
00:41:12,820 --> 00:41:17,100
Like I was saying at the start, the DEF CON subreddit was actually pretty hard to do.

407
00:41:17,140 --> 00:41:22,600
Like some of the other ones that I was looking at, like r slash cringe and r slash NFL.

408
00:41:23,080 --> 00:41:29,060
Like I was just sort of like wanting to see how bad this subreddit was versus others, and it was really good.

409
00:41:29,060 --> 00:41:33,420
So it was very difficult to find much information in there.

410
00:41:33,420 --> 00:41:44,900
I know it looks like those few slides, there was lots, but I probably had something in the realm of like maybe 20 or 30, which I thought would be a good starting point.

411
00:41:44,900 --> 00:41:51,720
And probably, yeah, wasn't a lot of like first names, a lot of last names, almost zero like addresses, actual addresses.

412
00:41:53,300 --> 00:41:57,660
There was just lots, and I think that the PII thing was probably fairly covered by people.

413
00:41:57,660 --> 00:42:02,720
I think most people understand and not put those kind of like actual facts about yourself in there.

414
00:42:02,820 --> 00:42:12,160
But the thing I did see posting over and over again was like, yeah, personal, like sort of like own feelings, own mental state.

415
00:42:12,160 --> 00:42:25,600
Last thing I wanted to mention actually, like the mental health slide, the mental health, I think it was like one of the first or second slides had someone was discussing their like mental status.

416
00:42:25,600 --> 00:42:29,500
It was kind of like a really , that was a difficult one for me.

417
00:42:29,500 --> 00:42:39,100
I was a bit conflicted about whether to put that up there because a lot of people use talking about that kind of stuff in a forum to kind of like help them get through it.

418
00:42:40,020 --> 00:42:59,480
But you know, particularly if you're thinking, if you believe you have a high threat model, you know, revealing that kind of stuff to the entire world that you are potentially compromised in different areas, you know, it does open you up to being manipulated in that way or being attacked in that way.

419
00:43:00,100 --> 00:43:04,260
But yeah, I can't think off the top of my head, apologize , what like the most frequent one was.

420
00:43:04,260 --> 00:43:10,980
Maybe that's a good thing to do next, kind of like work out where the where the worst, where the worst leaks are happening.

421
00:43:10,980 --> 00:43:11,920
Thanks, sir.

422
00:43:15,110 --> 00:43:17,690
I had a question about like fact checking.

423
00:43:17,750 --> 00:43:21,590
So especially on Reddit, sometimes people like to role play as different things in different subreddits.

424
00:43:21,590 --> 00:43:28,250
And you'll see people on one forum saying, Oh, I'm a software engineer, in a different thread, they'll say, Oh, I'm an auto mechanic.

425
00:43:28,250 --> 00:43:31,870
And they might not be either, but they want to participate in these threads.

426
00:43:31,910 --> 00:43:38,890
Is there a way that you've kind of vetted information sharing as face fake or true information?

427
00:43:39,790 --> 00:43:40,950
Yeah, that's a good question.

428
00:43:41,270 --> 00:43:54,530
So sort of manually, like manually, what people will do if they're like, if they're calling bullshit on a person, they'll just sort of go to the user and look at all the other, look at all the other posts that are in and go through.

429
00:43:56,210 --> 00:43:59,190
Do it using this to do it, I imagine it'd be pretty easy.

430
00:43:59,190 --> 00:44:02,910
Like, you know, I was only distilled down to that DEF CON subreddit.

431
00:44:02,910 --> 00:44:09,450
So the sorry, the prompt was only focused directly on that text file.

432
00:44:10,850 --> 00:44:22,530
But if you were to compile a bunch of, you know, when you download that torrent, it was just like, which, which, you know, it's like one and a half terabyte, I wasn't gonna download it all, just downloaded the ones that I wanted to .

433
00:44:22,550 --> 00:44:25,790
But if you, I'd imagine this would be how I would go about it.

434
00:44:26,050 --> 00:44:28,110
Again, not a big, not a data guy.

435
00:44:28,110 --> 00:44:29,270
But this is how I would do it.

436
00:44:29,350 --> 00:44:31,750
Go to the person, find all the ones that they're in.

437
00:44:32,250 --> 00:44:41,050
And, yeah, go get that text file, put it all together, and then ask it questions about things like job, things like experience.

438
00:44:41,230 --> 00:44:48,590
And, you know, from the results I got, it is fairly capable in doing that and looking through and finding examples of that kind of stuff.

439
00:44:48,690 --> 00:44:51,090
But yeah, not I didn't, I didn't actually do it.

440
00:44:51,090 --> 00:44:51,650
But thanks.

441
00:44:51,650 --> 00:44:52,170
Yeah.

442
00:44:52,990 --> 00:45:03,950
I was wondering, when you found yourself moving from Reddit to other platforms, could you rank the platforms, which ones people share the most information about themselves on?

443
00:45:04,610 --> 00:45:05,190
Okay, yeah.

444
00:45:05,190 --> 00:45:11,950
So I didn't search through Discord or the forums, just because Discord hates being scraped.

445
00:45:11,950 --> 00:45:17,910
And the forums, I didn't actually find that much, you know, actual information in there.

446
00:45:19,070 --> 00:45:20,190
So yeah.

447
00:45:20,670 --> 00:45:21,570
No.

448
00:45:21,570 --> 00:45:21,910
Yeah.

449
00:45:21,910 --> 00:45:22,750
Thanks.

450
00:45:25,010 --> 00:45:25,890
All right.

451
00:45:25,890 --> 00:45:29,890
I've been given a couple of minutes warning wrap up if there's any one or two questions left.

452
00:45:29,910 --> 00:45:31,630
Otherwise, thanks very much.

453
00:45:31,630 --> 00:45:32,810
Been an awesome audience.