1
00:00:02,000 --> 00:00:05,500
ARI: Hello, I'm Ari. [CLAUDINE: And I'm Claudine] Welcome to Proving the Negative.

2
00:00:05,500 --> 00:00:09,500
ARI: We're a podcast all about exploring the different sides of cybersecurity,

3
00:00:09,500 --> 00:00:13,790
from political to computer science, international relations to mathematics.

4
00:00:13,790 --> 00:00:16,072
Join us as we talk to our friends about the work they do.

5
00:00:16,072 --> 00:00:21,560
CLAUDINE: Today, we're going to be talking about how hate groups operate online,

6
00:00:21,560 --> 00:00:29,990
how they use social media platforms to coordinate efforts or spread ideologies.

7
00:00:29,990 --> 00:00:32,729
ARI: It's a topic for governments, it's a topic for the public,

8
00:00:32,729 --> 00:00:35,780
and it's of interest, of course, to law enforcement.

9
00:00:35,780 --> 00:00:42,170
CLAUDINE: The way online platforms were used to coordinate the Jan 6th attack on the Capitol,

10
00:00:42,170 --> 00:00:48,290
organised groups can use social media to coordinate offline actions.

11
00:00:48,290 --> 00:00:55,100
Those actions can slip under the radar, so it's a topic I'm keen to learn more about.

12
00:00:55,100 --> 00:01:01,730
FATIMA: My name is Fatima, PhD student at the University of Oxford in cyber security.

13
00:01:01,730 --> 00:01:07,850
My research look into online hate and how it's used across multiple platforms.

14
00:01:07,850 --> 00:01:12,110
ARI: Your elevator pitch - what would that sound like?

15
00:01:12,110 --> 00:01:15,979
We've seen over the past couple of decades how online platforms,

16
00:01:15,979 --> 00:01:21,860
and social media has become an increasingly integral part of communication,

17
00:01:21,860 --> 00:01:26,990
this has especially been highlighted during the ongoing COVID pandemic.

18
00:01:26,990 --> 00:01:31,280
But this of course, had an impact on how malicious actors communicate as well.

19
00:01:31,280 --> 00:01:38,690
Extremist and hate groups make use of platforms to spread propaganda,

20
00:01:38,690 --> 00:01:44,180
recruit individuals, or just form stronger networks with like minded people.

21
00:01:44,180 --> 00:01:47,837
We've seen how this behaviour can lead to catastrophic events,

22
00:01:47,837 --> 00:01:52,400
like the ISIS Paris attacks and the Christchurch mosque attack.

23
00:01:52,400 --> 00:01:56,030
So this isn't just a problem for the online community.

24
00:01:56,030 --> 00:02:02,370
This is something that has a huge impact on the offline world as well.

25
00:02:02,370 --> 00:02:06,394
So this isn't a problem faced just by specific platforms,

26
00:02:06,394 --> 00:02:10,650
it's something that is faced by a majority of online platforms today,

27
00:02:10,650 --> 00:02:15,101
and oftentimes hate groups will make use of multiple platforms,

28
00:02:15,101 --> 00:02:17,040
no matter their audience size.

29
00:02:17,040 --> 00:02:25,530
I argue that we can't look at online hate from the perspective of a single platform.

30
00:02:25,530 --> 00:02:32,160
I look into how we can use a cross-platform perspective,

31
00:02:32,160 --> 00:02:36,600
to see if we can gain any insight into how platforms are used differently,

32
00:02:36,600 --> 00:02:40,920
and whether we can map hateful behaviour across multiple platforms.

33
00:02:40,920 --> 00:02:44,790
So we have a more organised approach to countering it.

34
00:02:44,790 --> 00:02:48,960
CLAUDINE: Could you clarify what you mean by a malicious actor?

35
00:02:48,960 --> 00:02:54,000
FATIMA: What I am considering a malicious actor is...

36
00:02:54,000 --> 00:03:02,344
I'm looking at hate groups that have been identified by law enforcement agencies,

37
00:03:02,344 --> 00:03:09,000
or people who have identified in some way that they are supporters of hateful

38
00:03:09,000 --> 00:03:12,780
ideology that has also been identified by law enforcement agencies.

39
00:03:12,780 --> 00:03:18,660
Various other malicious actors use social media.

40
00:03:18,660 --> 00:03:23,280
ARI: So what kind of malicious actors do we have out there now?

41
00:03:23,280 --> 00:03:24,540
FATIMA: They're not just humans.

42
00:03:24,540 --> 00:03:28,728
You have bots as well who are propagating all kinds of stuff,

43
00:03:28,728 --> 00:03:33,960
whether it's trying to influence the opinion of the online audience,

44
00:03:33,960 --> 00:03:41,000
or whether it's just trying to get people to engage in certain activities online,

45
00:03:41,000 --> 00:03:44,280
in which you can get information from them.

46
00:03:44,280 --> 00:03:51,000
So we saw this with Cambridge Analytica. It wasn't so much malicious,

47
00:03:51,000 --> 00:03:57,490
as it was intrusive and exploiting things that online platforms provide

48
00:03:57,490 --> 00:04:00,960
that weren't brought to the attention of the general public before.

49
00:04:00,960 --> 00:04:09,000
The definition changes all the time - as well as how they make use of platforms.

50
00:04:09,000 --> 00:04:12,960
ARI: The end result is to have this unified approach,

51
00:04:12,960 --> 00:04:18,330
different platforms have one problem in common, particularly the problem you look at.

52
00:04:18,330 --> 00:04:23,040
Why do we have to work together - who needs to work together on this?

53
00:04:23,040 --> 00:04:27,060
FATIMA: It's not so much that we need to have one unified approach.

54
00:04:27,060 --> 00:04:32,640
It's more - we haven't looked at how multiple platforms are used in this space.

55
00:04:32,640 --> 00:04:37,650
Research has seen: "so this is what's going on in Twitter this year, this month..."

56
00:04:37,650 --> 00:04:42,840
"... so that means this is how we need to address online hate".

57
00:04:42,840 --> 00:04:47,970
But then that's specific to Twitter so we can't apply that to multiple platforms.

58
00:04:47,970 --> 00:04:56,550
We need to start looking at all platforms and seeing how hate groups use them.

59
00:04:56,550 --> 00:05:00,360
Do they use them for different things? Are they used together in some way?

60
00:05:00,360 --> 00:05:06,360
Is some stuff posted on one platform, and then other stuff posted on another?

61
00:05:06,360 --> 00:05:13,950
The rules on one platform are more lax than another - a unifying response wouldn't work.

62
00:05:13,950 --> 00:05:20,520
It's more that we haven't looked at the perspectives of different platforms,

63
00:05:20,520 --> 00:05:25,470
So we haven't included that in the counter-approaches we're looking at.

64
00:05:25,470 --> 00:05:30,780
CLAUDINE: I realise it's a work in progress - what have you found so far,

65
00:05:30,780 --> 00:05:33,976
In terms of taking that global perspective of multiple platforms?

66
00:05:33,976 --> 00:05:41,670
FATIMA: I have seen differences in how platforms are used, it can be very subtle.

67
00:05:41,670 --> 00:05:45,030
I was looking at online hate during the US elections.

68
00:05:45,030 --> 00:05:52,110
I was looking at hate groups posting on different platforms - 4chan and Reddit,

69
00:05:52,110 --> 00:05:58,620
Reddit had a much more organised response during the actual election

70
00:05:58,620 --> 00:06:05,760
posting links for fundraising, whether hate groups or Trump-supporting causes.

71
00:06:05,760 --> 00:06:10,230
There was a lot of repeat posting URL links,

72
00:06:10,230 --> 00:06:17,040
but 4chan discussed the results more - not as much of an organised approach.

73
00:06:17,040 --> 00:06:21,390
It's interesting to see how different communities use platforms differently.

74
00:06:21,390 --> 00:06:27,480
That was one thing that I've found so far. CLAUDINE: Could you clarify 'hate group'?

75
00:06:27,480 --> 00:06:35,490
FATIMA: Hate groups have a different definition across different areas, jurisdictions, countries,

76
00:06:35,490 --> 00:06:40,620
each kind of law enforcement agency will have their own definition of a hate group.

77
00:06:40,620 --> 00:06:50,460
Generally, it's when you're going against acceptable social norms.

78
00:06:50,460 --> 00:06:56,340
A lot of the time it's specific to the country that you're in.

79
00:06:56,340 --> 00:06:58,530
This is applicable to social media as well.

80
00:06:58,530 --> 00:07:06,840
Social media platforms generally have their own body to monitor online content.

81
00:07:06,840 --> 00:07:11,100
Each country also has their rules, which social media platforms have to follow.

82
00:07:11,100 --> 00:07:18,780
Similarly, hate groups are going against what is socially acceptable behaviour.

83
00:07:18,780 --> 00:07:22,440
ARI: They say things that mean certain groups of people have a worse time.

84
00:07:22,440 --> 00:07:26,460
People being mistreated - that's a tangible result.

85
00:07:26,460 --> 00:07:32,730
We don't want hate groups because people will suffer in real life.

86
00:07:32,730 --> 00:07:37,980
This becomes a real world harm - these ideas being tossed around in online forums.

87
00:07:37,980 --> 00:07:42,930
These things are bad because people are impacted. It's not a data problem entirely.

88
00:07:42,930 --> 00:07:49,860
Is this a fair assessment of why hate groups are undesirable?

89
00:07:49,860 --> 00:07:58,080
FATIMA: Exactly. It's why most countries now have have listed online hate as a harm,

90
00:07:58,080 --> 00:08:03,150
even though it is a grey area in terms of defining and having one unified response

91
00:08:03,150 --> 00:08:08,250
everywhere and one unified definition of hate across all countries and languages and stuff.

92
00:08:08,250 --> 00:08:16,500
It's universally agreed that hate groups cause harm online, and consequence offline as well.

93
00:08:16,500 --> 00:08:22,830
CLAUDINE: You saw a more organised response for these groups on certain platforms than on others.

94
00:08:22,830 --> 00:08:28,500
Have you seen a change in the relationship between hate groups and the platforms that they use.

95
00:08:28,500 --> 00:08:36,510
If platforms respond to new and emerging threats (posed by hate groups), by reinforcing standards,

96
00:08:36,510 --> 00:08:43,440
how does that affect the way those hate groups use social media platforms and which social media platforms they use?

97
00:08:43,440 --> 00:08:49,830
FATIMA: This is a huge area of research .

98
00:08:49,830 --> 00:08:56,460
The main example that can be used is during ISIS' prime on Twitter,

99
00:08:56,460 --> 00:09:05,970
2014-2016, we saw a lot of pro-ISIS support posted on Twitter.

100
00:09:05,970 --> 00:09:12,300
But after a lot of the attacks that were happening offline,

101
00:09:12,300 --> 00:09:18,510
in particular the Paris attacks, Twitter took a stand and started suspending multiple accounts.

102
00:09:18,510 --> 00:09:26,160
At first it was becoming a badge of honour to these pro-ISIS supporters, where they'd,

103
00:09:26,160 --> 00:09:33,690
create new accounts and say "Oh, I got suspended X many times before", like they'd been doing more for the cause.

104
00:09:33,690 --> 00:09:39,840
But of course, this meant that eventually they were kicked off the platform,

105
00:09:39,840 --> 00:09:45,600
so they had to move to a different platform to spread more extremist content.

106
00:09:45,600 --> 00:09:51,000
We saw them move to more encrypted platforms like Telegram, which of course,

107
00:09:51,000 --> 00:09:57,240
champion themselves as providing encrypted communication.

108
00:09:57,240 --> 00:10:02,250
Your communication isn't going to be accessible to law enforcement. No one's going to publish it.

109
00:10:02,250 --> 00:10:08,520
It's going to stick within your groups. We see platform migration a lot in this space,

110
00:10:08,520 --> 00:10:15,960
even over the last year or so - the US elections and BLM resurgence last summer,

111
00:10:15,960 --> 00:10:21,270
we saw that there was a lot more hate on Reddit communities and they were

112
00:10:21,270 --> 00:10:25,770
eventually banned by it because they were increasing their hateful content.

113
00:10:25,770 --> 00:10:34,330
Any time one of these platforms is suspended, they move to another .

114
00:10:34,330 --> 00:10:41,200
That platform will have more lax rules - eventually pressure mounts on them to

115
00:10:41,200 --> 00:10:46,660
monitor their content and to moderate whatever is being posted there.

116
00:10:46,660 --> 00:10:54,460
You get platforms with a lot more reach, the ones with more rules and moderation.

117
00:10:54,460 --> 00:11:04,570
And you have platforms with less reach, but more lax rules. Platforms are used together.

118
00:11:04,570 --> 00:11:12,970
You're still recruiting, but not posting as much extremist or hateful explicit content on mainstream sites.

119
00:11:12,970 --> 00:11:18,970
ARI: They're being used in the way that they've been intended to be used as social media.

120
00:11:18,970 --> 00:11:23,540
Then it's whether you censor. Free speech is not freedom from

121
00:11:23,540 --> 00:11:25,870
consequence, though. FATIMA: This is huge,

122
00:11:25,870 --> 00:11:35,750
especially platforms with more lax rules. Their main ethos is "we're providing freedom of speech".

123
00:11:35,750 --> 00:11:41,110
We won't be monitoring. We won't have as many rules.

124
00:11:41,110 --> 00:11:50,050
You will see that they're used by particular hate groups and not others.

125
00:11:50,050 --> 00:11:56,890
For example, Gab rose to prominence following Trump's election,

126
00:11:56,890 --> 00:12:01,600
a lot of content being posted on Twitter had become identified as hate.

127
00:12:01,600 --> 00:12:05,590
So this other platform (Gab) was saying, Oh, we won't have any rules here.

128
00:12:05,590 --> 00:12:10,540
So - used more by Trump supporting white supremacist type hate.

129
00:12:10,540 --> 00:12:17,590
You won't find as many ISIS supporters there, which is an interesting part of this.

130
00:12:17,590 --> 00:12:20,950
Mostly, it's under the control of platform providers.

131
00:12:20,950 --> 00:12:28,780
Twitter will have their own set of rules.

132
00:12:28,780 --> 00:12:32,710
And so a lot of the time this [hate posts] will just be suspended or deleted.

133
00:12:32,710 --> 00:12:36,490
It's mainly just the platform provider, especially when it comes to removing content.

134
00:12:36,490 --> 00:12:45,190
There can be pressure put on them by governments and even the general public,

135
00:12:45,190 --> 00:12:49,750
if there's enough of a response. It's just so big in terms of perspective.

136
00:12:49,750 --> 00:12:56,800
There are countless platforms, countless hate groups. And so I have to narrow down my scope a little.

137
00:12:56,800 --> 00:13:02,380
I was looking at this from a specific use case, just so we'd get an idea and then

138
00:13:02,380 --> 00:13:07,750
identify hate groups or hate ideologies  present across multiple platforms.

139
00:13:07,750 --> 00:13:14,410
That would help narrow down the platforms and hate ideologies we include.

140
00:13:14,410 --> 00:13:19,150
CLAUDINE: What was the most surprising or unexpected thing that you've come across?

141
00:13:19,150 --> 00:13:26,440
FATIMA: I didn't know whether there would be a difference in how each platform was used,

142
00:13:26,440 --> 00:13:32,000
each of these platforms has their own audience and their own type of content.

143
00:13:32,000 --> 00:13:39,850
There was more explicit content on 4chan,

144
00:13:39,850 --> 00:13:44,290
people weren't afraid to link pictures or graphics and stuff.

145
00:13:44,290 --> 00:13:53,750
Reddit was more specific, because Reddit has more rules and moderation.

146
00:13:53,750 --> 00:13:57,460
That could have had an impact - "we don't want to get kicked off this platform..."

147
00:13:57,460 --> 00:14:01,630
"... so we have to stick by the rules". 4chan was less organised,

148
00:14:01,630 --> 00:14:06,910
you could post anything and there were a lot more agreements

149
00:14:06,910 --> 00:14:10,990
and disagreements. ARI: If you had all the resources and you had all the money in the world.

150
00:14:10,990 --> 00:14:15,520
What big question, would you explore? FATIMA: For this particular area of research,

151
00:14:15,520 --> 00:14:22,660
there are so many platforms online, there's so many hate groups.

152
00:14:22,660 --> 00:14:30,190
Hate is posted in different languages, cultures... there are so many definitions.

153
00:14:30,190 --> 00:14:33,970
There hasn't been a lot of comparison in this area.

154
00:14:33,970 --> 00:14:38,260
Whether it's comparing hateful content in one language to English,

155
00:14:38,260 --> 00:14:42,830
because most research in this area focuses on English hate,

156
00:14:42,830 --> 00:14:49,930
we haven't had comparison across different countries, languages, cultures, platforms.

157
00:14:49,930 --> 00:14:57,640
This is something we need to do to better understand of how hate operates online.

158
00:14:57,640 --> 00:15:01,780
So if I had an unlimited amount of time and money,

159
00:15:01,780 --> 00:15:10,500
I would look at hate from different languages, seeing how it compares,

160
00:15:10,500 --> 00:15:15,100
whether there's cause for concern in other countries as well,

161
00:15:15,100 --> 00:15:18,860
whether that has an impact on online hate in other places.

162
00:15:18,860 --> 00:15:24,500
We saw with ISIS' attacks, a lot of the time it was organised in Arabic.

163
00:15:24,500 --> 00:15:30,310
But then actual attacks happen in non-Arabic speaking countries.

164
00:15:30,310 --> 00:15:33,940
ARI: Is there a word we can use instead of hate? FATIMA: This is interesting,

165
00:15:33,940 --> 00:15:40,850
when I was starting, I was focussing on extremism and extremist groups.

166
00:15:40,850 --> 00:15:49,900
There were less groups [referred to by] law enforcement under this particular term.

167
00:15:49,900 --> 00:15:55,570
They would have a list of hate groups because, I guess for them a lot of,

168
00:15:55,570 --> 00:16:01,930
extremist groups are hate groups that have been linked to attacks offline.

169
00:16:01,930 --> 00:16:07,120
There's a definite impact, from the online to offline.

170
00:16:07,120 --> 00:16:12,070
ISIS is clearly an extremist group.

171
00:16:12,070 --> 00:16:20,020
Then you have individual extremists like the Christchurch shooter who isn't operating under a group.

172
00:16:20,020 --> 00:16:27,140
But still very much an extremist. This Christchurch shooter only became an

173
00:16:27,140 --> 00:16:33,460
extremist after he did the attack. He wouldn't have been considered an extremist while posting online.

174
00:16:33,460 --> 00:16:38,350
It would have just been online hate, because he was attacking certain groups.

175
00:16:38,350 --> 00:16:41,570
It wouldn't have been extremism, per se.

176
00:16:41,570 --> 00:16:49,060
I went towards hate in my research is because it was a slightly broader.

177
00:16:49,060 --> 00:16:53,170
It had a slightly broader definition than something specific like extremism.

178
00:16:53,170 --> 00:16:59,780
Currently, with how legislation across most countries is being proposed,

179
00:16:59,780 --> 00:17:05,780
'hate' is what is generally used. You will find more resources or information,

180
00:17:05,780 --> 00:17:09,640
on the type of groups or ideologies that are counted as hate.

181
00:17:09,640 --> 00:17:19,840
CLAUDINE: Have you noticed any overarching similarities between different hate groups,

182
00:17:19,840 --> 00:17:21,940
obviously not in terms of ideology necessarily,

183
00:17:21,940 --> 00:17:30,880
in terms of the approaches they use to market themselves, recruit or organise?

184
00:17:30,880 --> 00:17:46,750
FATIMA: A lot of the time there wasn't enough content for me to compare.

185
00:17:46,750 --> 00:17:53,500
I tended to just group my hate groups into one overarching hate ideology,

186
00:17:53,500 --> 00:17:56,800
and compare them across multiple platforms.

187
00:17:56,800 --> 00:17:59,260
That being said, while I was doing my research,

188
00:17:59,260 --> 00:18:08,080
say one hate group is propagating hate through selling merchandise (t-shirts, music...).

189
00:18:08,080 --> 00:18:13,930
This would be promoted by other hate groups, "this is a link, help our cause".

190
00:18:13,930 --> 00:18:16,420
So there was collaboration between them.

191
00:18:16,420 --> 00:18:24,070
It wasn't part of my research, but it's something that I found interesting,

192
00:18:24,070 --> 00:18:32,560
Y'know? Is this one hate group?

193
00:18:32,560 --> 00:18:39,070
Is this two hate groups? They have different names, organisational bodies and websites,

194
00:18:39,070 --> 00:18:42,880
there was a lot of overlap between them online.

195
00:18:42,880 --> 00:18:47,030
Many people will celebrate across the groups,

196
00:18:47,030 --> 00:18:56,970
whether it's politicians that are very radically right, or certain public figures,

197
00:18:56,970 --> 00:19:01,960
a lot of the time, it's public figures who have come to light in some way in this area.

198
00:19:01,960 --> 00:19:10,660
So they're celebrated across groups as some kind of idol.

199
00:19:10,660 --> 00:19:13,990
So there's a lot of overlap between them. ARI: Let's talk more about you.

200
00:19:13,990 --> 00:19:20,440
What are the frustrations/things you enjoy about the work that you do?

201
00:19:20,440 --> 00:19:29,080
It's the one space where I'm constantly learning, I don't know if I would get as much in another.

202
00:19:29,080 --> 00:19:36,310
One thing I found difficult was finding an area where I could provide insight.

203
00:19:36,310 --> 00:19:43,870
That's something that I struggled with. There's so much research,

204
00:19:43,870 --> 00:19:48,460
you realise "wow, there's so much research going on. How do I find the gap here?"

205
00:19:48,460 --> 00:19:51,360
That's one thing that I found difficult.

206
00:19:51,360 --> 00:19:58,810
Oftentimes you don't realise how long it can take just to identify that gap.

207
00:19:58,810 --> 00:20:04,780
Just to find this novel area of space that you can work in.

208
00:20:04,780 --> 00:20:11,830
I would say, have one question that you're constantly answering,

209
00:20:11,830 --> 00:20:15,400
no matter what kind of analysis that you're doing.

210
00:20:15,400 --> 00:20:20,030
That can also be a little difficult in your research.

211
00:20:20,030 --> 00:20:27,730
But once you have these clearly outlined , it becomes so interesting and then

212
00:20:27,730 --> 00:20:33,340
you start enjoying the whole process of actually learning from your analysis.

213
00:20:33,340 --> 00:20:39,850
There are going to be frustrating days where nothing's working!

214
00:20:39,850 --> 00:20:49,060
CLAUDINE: What are you working on now? FATIMA: As a result of a summer school...

215
00:20:49,060 --> 00:20:53,560
This is a group project with different researchers across the UK and Europe,

216
00:20:53,560 --> 00:20:59,950
we're using similar methods (text/social media analysis) where

217
00:20:59,950 --> 00:21:07,510
we're looking at how sentiment across online communities changes during crypto hypes.

218
00:21:07,510 --> 00:21:14,170
For people who don't know, this is when the crypto currency values rise and fall.

219
00:21:14,170 --> 00:21:21,670
We're looking at online crypto communities where this is discussed,

220
00:21:21,670 --> 00:21:28,450
whether there's a community involved when prices change, etc.

221
00:21:28,450 --> 00:21:31,450
It's not specifically feelings [we look at],

222
00:21:31,450 --> 00:21:37,840
it's the overall group opinion.

223
00:21:37,840 --> 00:21:43,540
We did find a few interesting findings.

224
00:21:43,540 --> 00:21:54,820
When prices of Bitcoin increased,

225
00:21:54,820 --> 00:22:03,490
then the online community became more individual and independent.

226
00:22:03,490 --> 00:22:10,660
So they will talk about "I bought/sold this stock in bitcoin".

227
00:22:10,660 --> 00:22:16,500
Next, I want to look at how these platforms are used within networks of hate,

228
00:22:16,500 --> 00:22:20,470
and how activity maps across platforms,

229
00:22:20,470 --> 00:22:28,600
whether there is a larger network of hate that encompasses multiple platforms.

230
00:22:28,600 --> 00:22:34,600
It's something I haven't done before - hopefully I can get something out of it!

231
00:22:34,600 --> 00:22:37,570
So it's more about using network analysis approaches.

232
00:22:37,570 --> 00:22:46,690
CLAUDINE: If people are interested in learning more, what resources would you suggest?

233
00:22:46,690 --> 00:22:55,480
FATIMA: This conference called ICWSM, general research in online social media platforms.

234
00:22:55,480 --> 00:23:01,150
There's a lot of research going on across loads of different topics, not just online hate,

235
00:23:01,150 --> 00:23:10,150
it's from - looking at pandemic sentiment around vaccinations.

236
00:23:10,150 --> 00:23:14,140
Whether you're looking at online hate or not,

237
00:23:14,140 --> 00:23:21,490
this conference will have a lot of the methods applicable to your research.

238
00:23:21,490 --> 00:23:28,780
This is a resource I found really useful. It also publishes data sets,

239
00:23:28,780 --> 00:23:36,980
loads of different types of datasets as well that are really useful.

240
00:23:36,980 --> 00:23:42,500
Research project VOX - Pol have a lot of ideas and a lot of articles,

241
00:23:42,500 --> 00:23:47,160
whether it's fully-fledged research publications, articles...

242
00:23:47,160 --> 00:23:53,240
"We've seen X happening online", So that's been a starting point.

243
00:23:53,240 --> 00:24:00,950
With how much technology and online resources that we're using day to day,

244
00:24:00,950 --> 00:24:06,003
at home or at work or what have you, realising how these can be misused,

245
00:24:06,003 --> 00:24:09,590
and how to use these technologies in the proper way...

246
00:24:09,590 --> 00:24:16,970
[that's] My understanding of cybersecurity - such a broad area.

247
00:24:16,970 --> 00:24:22,760
I found way more definitions of cyber security than I had anticipated.

248
00:24:22,760 --> 00:24:26,000
CLAUDINE: That was our interview with Fatima. Join us next week

249
00:24:26,000 --> 00:24:27,830
for another fascinating conversation.

250
00:24:27,830 --> 00:24:32,240
In the meantime, you can tweet at us @HelloPTNPod

251
00:24:32,240 --> 00:24:36,500
and you can subscribe on Apple Podcasts or wherever you listen to podcasts.

252
00:24:36,500 --> 00:24:42,330
The title there is PTNPod. See you next week. ARI: Bye!

253
00:24:42,330 --> 00:24:47,520
CLAUDINE: This has been a podcast from the Centre for Doctoral Training in Cybersecurity, University of Oxford.

254
00:24:47,520 --> 00:24:52,411
Funded by the Engineering and Physical Sciences Research Council.