1
00:00:12,930 --> 00:00:18,050
I'm going to refresh it again and see if it wakes up.

2
00:00:25,240 --> 00:00:26,760
Plavrdi, are you seeing this?

3
00:00:27,300 --> 00:00:28,740
Ah, there we go.

4
00:00:30,780 --> 00:00:32,420
Welcome, everybody!

5
00:00:32,420 --> 00:00:36,780
I want to thank everybody at DEF CON for coming to watch this Q&A session.

6
00:00:36,940 --> 00:00:41,600
I'm Fallible, we have Plavrdi as our Lagoon, and we are here with Yamila...

7
00:00:41,600 --> 00:00:44,620
Yamila Lavache, I believe.

8
00:00:44,860 --> 00:00:46,680
I totally did that wrong.

9
00:00:46,680 --> 00:00:48,180
She'll have to correct me here in a second.

10
00:00:48,180 --> 00:00:48,960
No, no, no.

11
00:00:48,960 --> 00:00:50,280
You pronounced it right.

12
00:00:50,640 --> 00:00:56,000
The talk is Bypassing Biometric Systems with 3D Printing.

13
00:00:56,000 --> 00:00:58,660
I'm super excited about this one.

14
00:00:59,000 --> 00:01:13,860
Any opportunity you have to talk about the details of how do you take biometric data and make it a physical thing that you can exploit because this is a really cool topic.

15
00:01:14,240 --> 00:01:23,740
I'm excited to see what more you have to say about it, and we're looking forward to questions coming in from the live Q&A audience.

16
00:01:23,960 --> 00:01:31,300
Before we get too far into this, this is, as I understand it, your first DEF CON talk.

17
00:01:31,300 --> 00:01:32,680
Is that correct?

18
00:01:32,860 --> 00:01:37,160
As you well know, there is a tradition here at DEF CON.

19
00:01:37,160 --> 00:01:42,960
For your very first DEF CON talk, we like to shoot the noob.

20
00:01:42,960 --> 00:01:48,280
So I would like you to acquire the beverage that you have prepared.

21
00:01:48,280 --> 00:01:49,780
Excellent.

22
00:01:50,320 --> 00:01:52,540
To tradition.

23
00:01:52,800 --> 00:01:53,880
Cheers.

24
00:01:53,880 --> 00:01:55,080
Cheers.

25
00:01:57,900 --> 00:01:59,420
Awesome.

26
00:01:59,420 --> 00:02:00,560
Wow.

27
00:02:00,560 --> 00:02:03,400
This is my beverage.

28
00:02:03,420 --> 00:02:04,180
Oh.

29
00:02:04,400 --> 00:02:07,260
Ginger liquor from Switzerland.

30
00:02:08,520 --> 00:02:10,300
My whiskey just got to me.

31
00:02:10,300 --> 00:02:11,280
Excuse me.

32
00:02:12,680 --> 00:02:13,780
Okay.

33
00:02:13,780 --> 00:02:16,440
Yeah, you can certainly refill that glass if you want to.

34
00:02:16,440 --> 00:02:17,660
Yeah, you're welcome to.

35
00:02:17,660 --> 00:02:18,380
This is your time.

36
00:02:18,380 --> 00:02:24,480
I also have Fernet because this is the beverage that we bring in Argentina.

37
00:02:24,480 --> 00:02:28,040
It's our beverage, our preferred beverage.

38
00:02:28,040 --> 00:02:29,200
With Coke.

39
00:02:30,080 --> 00:02:32,800
I also have Fernet here.

40
00:02:32,800 --> 00:02:34,360
That's your nice mixer.

41
00:02:34,360 --> 00:02:37,340
You get it in with the Coke and it's the national beverage.

42
00:02:37,340 --> 00:02:38,960
I like that.

43
00:02:39,720 --> 00:02:40,960
Well, excellent.

44
00:02:41,400 --> 00:02:44,040
Let's get you started with this...

45
00:02:44,700 --> 00:02:49,300
Well, you prompted me with a couple of questions, but I want to go a slightly different direction.

46
00:02:49,680 --> 00:02:52,540
Tell us what 3D printers you have.

47
00:02:52,540 --> 00:03:00,040
Because a 3D printer capable of doing the things at the resolution that you're working in is obviously interesting.

48
00:03:00,140 --> 00:03:02,700
So, you have a couple of them, right?

49
00:03:03,040 --> 00:03:05,820
Yes, I have two.

50
00:03:05,820 --> 00:03:10,640
The first one is an FDM, a filament 3D printer.

51
00:03:10,640 --> 00:03:16,380
It's a clone of the Prusa i3 MK2.

52
00:03:16,380 --> 00:03:21,960
I built myself, I assembled the 3D printer myself.

53
00:03:22,180 --> 00:03:24,540
I love it.

54
00:03:25,480 --> 00:03:28,300
Also, it was really interesting too.

55
00:03:28,300 --> 00:03:34,750
It took like a month, but in the end, it's great.

56
00:03:35,120 --> 00:03:39,080
And then, last year, I bought another one.

57
00:03:39,080 --> 00:03:42,500
A resin 3D printer, the Anycubic Photon.

58
00:03:42,500 --> 00:03:53,740
And this is the printer that I use for this research, because I needed the resolution of a resin 3D printer.

59
00:03:53,740 --> 00:03:57,220
I can't do the research with the other printer.

60
00:03:57,400 --> 00:04:02,400
And it sounds like it's doing the job, so that's useful for us to know.

61
00:04:02,400 --> 00:04:05,240
And you might have mentioned that in the talk.

62
00:04:05,240 --> 00:04:08,400
I'm sorry, I missed it when we came through, so thank you for that.

63
00:04:08,400 --> 00:04:09,680
I appreciate that.

64
00:04:09,680 --> 00:04:14,240
So, what got you into this area of research?

65
00:04:16,620 --> 00:04:32,200
Last year, I learned that some people broke the authentication mechanism of Galaxy S10 with just a silicone cover.

66
00:04:32,400 --> 00:04:34,200
That was interesting.

67
00:04:34,200 --> 00:04:41,400
And also, I think it was the Geek Pong cybersecurity competition.

68
00:04:41,400 --> 00:04:50,560
The research team from Tencent broke the authentication mechanism of three different phones.

69
00:04:50,720 --> 00:04:57,700
Taking a photo of a fingerprint, a latent fingerprint in glass.

70
00:04:57,700 --> 00:05:05,920
And then, 20 minutes later, they had a working fake fingerprint.

71
00:05:05,920 --> 00:05:14,440
So, I was thinking that they might have used some type of 3D printing for that.

72
00:05:14,440 --> 00:05:26,660
And that's one of the reasons that I wanted to try and see if I, in my home, with my home resin 3D printer, I can duplicate fingerprints.

73
00:05:26,660 --> 00:05:35,180
Also, the leakage of biometric data from companies.

74
00:05:35,760 --> 00:05:38,100
Say that part again.

75
00:05:38,100 --> 00:05:39,160
The leaked cache?

76
00:05:39,380 --> 00:05:41,140
Oh, no.

77
00:05:41,140 --> 00:05:53,540
The leak of biometric data from several biometric companies also was inspiration, we can say, for this research.

78
00:05:53,540 --> 00:05:58,580
Now, I don't think I've heard of a leak of biometric data from other companies.

79
00:05:58,580 --> 00:06:00,500
Can you talk a little bit more about that?

80
00:06:01,480 --> 00:06:14,500
Yes, it was last year that a leak from a biometric company and all the fingerprints and other data was leaked.

81
00:06:14,500 --> 00:06:27,240
And I was thinking about how, if an attacker can use this information to create fake fingerprints for a person.

82
00:06:27,240 --> 00:06:31,180
Because I can't change my fingerprints.

83
00:06:31,200 --> 00:06:38,220
And if my fingerprints are on a leak and a person can do a duplicate.

84
00:06:38,660 --> 00:06:44,380
So, that was one of the reasons for doing this research also.

85
00:06:44,380 --> 00:06:45,360
That's great.

86
00:06:45,640 --> 00:06:49,540
There was a question I really wanted to ask you.

87
00:06:50,100 --> 00:07:01,100
Is there evidence that there are people who are amassing these people's biometric data and then using that somehow?

88
00:07:01,100 --> 00:07:12,860
I think that the way I was going to put this out there was, what of the methods that you discussed in your talk are things that we can share?

89
00:07:12,860 --> 00:07:19,040
That I could digitize and create a database of other people's biometric data?

90
00:07:19,760 --> 00:07:23,360
Yes, I think this technique can be used.

91
00:07:23,360 --> 00:07:32,240
As you can see in the paper and in the video, it's not easy, all the process to duplicate the fingerprint.

92
00:07:32,560 --> 00:07:40,860
It takes time and experience, but I think that it can be done.

93
00:07:40,860 --> 00:07:47,440
And also with other equipment, because my research was a very low budget one.

94
00:07:47,440 --> 00:07:51,060
I only have my 3D printer.

95
00:07:51,360 --> 00:08:01,440
For example, I don't have a digital microscope to measure the reach height and something like this.

96
00:08:01,440 --> 00:08:13,700
But I think with more resources, it could be possible to do it in a bigger scale, not just one.

97
00:08:14,860 --> 00:08:19,040
So with that, what was your failure rate?

98
00:08:19,800 --> 00:08:26,060
How often did you have to reprint the fingerprints that you were creating?

99
00:08:26,060 --> 00:08:33,740
And each of your examples, I know that you had your, hey, it worked on this version, it didn't work on this version.

100
00:08:33,740 --> 00:08:39,620
But I didn't see where you were discussing how many times you had to try in order to get one that actually worked.

101
00:08:39,620 --> 00:08:43,620
Yes, I have here some of the failure.

102
00:08:43,620 --> 00:08:45,680
I don't know if you can see.

103
00:08:45,680 --> 00:08:46,900
A lot of failure.

104
00:08:49,600 --> 00:08:55,480
So I have here the molds and some of the fake fingerprints.

105
00:08:56,600 --> 00:09:02,460
The most important thing with the duplication of the fingerprint is the size.

106
00:09:02,460 --> 00:09:16,820
If you can configure the correct size and specifically the reach height, because we can't measure the reach height without a specific technology like the microscope.

107
00:09:16,820 --> 00:09:18,520
I don't have one.

108
00:09:19,780 --> 00:09:37,920
So for that reason, and also the configuration of the printer, according to the resin and this type, I print at the end, I print like 10 batches.

109
00:09:38,400 --> 00:09:44,060
And in each one, I print several fingerprints, not just one.

110
00:09:44,060 --> 00:09:54,520
And also I try with different, for example, I print this in a horizontal way and it works better in a vertical way.

111
00:09:54,520 --> 00:10:01,240
You can see here that it has the support because I needed to print in vertical.

112
00:10:01,240 --> 00:10:05,960
The resolution was better in vertical than in the horizontal.

113
00:10:05,960 --> 00:10:12,320
So that's information that you discover when you are testing and experimenting.

114
00:10:13,300 --> 00:10:16,680
I wouldn't think of it before.

115
00:10:16,820 --> 00:10:18,420
Absolutely.

116
00:10:18,640 --> 00:10:24,980
And also I don't have experience in criminalistic or nothing.

117
00:10:24,980 --> 00:10:35,060
It's just I wanted to try if a regular user with a 3D printer can duplicate fingerprints.

118
00:10:35,060 --> 00:10:38,080
That was the idea of the research.

119
00:10:39,020 --> 00:10:42,160
It's very fun to try all the other methods also.

120
00:10:42,160 --> 00:10:45,500
Not when I burn myself with hot glue.

121
00:10:48,000 --> 00:10:54,080
Yeah, I think that a lot of makers are going to be familiar with that particular one.

122
00:10:54,400 --> 00:10:55,600
Oh, yes.

123
00:10:56,820 --> 00:11:04,900
I learned about this experience and now I do it with a lot of care.

124
00:11:04,900 --> 00:11:05,860
Live and learn.

125
00:11:05,860 --> 00:11:06,640
I like that.

126
00:11:06,640 --> 00:11:09,950
So did you try these attacks on other devices?

127
00:11:11,820 --> 00:11:19,060
I tried only on the devices that are in the paper and in the talk.

128
00:11:19,060 --> 00:11:24,940
But also the people from Cisco, from Talos, they made very similar research.

129
00:11:24,940 --> 00:11:31,060
I was talking from one to one of them with Paul and he used the same printer.

130
00:11:31,060 --> 00:11:35,960
And he tested in different, totally different devices.

131
00:11:35,960 --> 00:11:42,760
Like computers and padlocks and I think other phones.

132
00:11:42,760 --> 00:11:47,160
So also you can make data results.

133
00:11:47,160 --> 00:11:54,440
Because it's similar but they use the same printer and other software, different software.

134
00:11:54,440 --> 00:11:57,220
And they test on other devices.

135
00:11:57,220 --> 00:12:01,400
So I think it can be used to see.

136
00:12:01,400 --> 00:12:05,080
But they succeed in most of the devices.

137
00:12:05,080 --> 00:12:07,880
So it's an attack that is feasible.

138
00:12:08,060 --> 00:12:09,320
Excellent, that makes sense.

139
00:12:09,320 --> 00:12:16,520
And it's something that you can, if a person wanted to, this could be expanded into a pile of other devices.

140
00:12:16,660 --> 00:12:24,720
So beyond just other devices, what other types of biometric attacks would you be interested in researching in the future?

141
00:12:26,580 --> 00:12:30,960
I am very interested in facial recognition.

142
00:12:31,180 --> 00:12:34,940
In two different aspects of facial recognition.

143
00:12:34,940 --> 00:12:38,260
The first one is avoid facial recognition.

144
00:12:38,260 --> 00:12:41,780
I think it's an interesting topic.

145
00:12:41,780 --> 00:12:46,660
And the other one is using fake biometric trace.

146
00:12:46,660 --> 00:12:51,200
For example, a mask or a 3D printed head.

147
00:12:51,320 --> 00:12:54,360
I am working on that.

148
00:12:54,360 --> 00:12:59,980
And I am testing this type of attacks in facial recognition systems.

149
00:12:59,980 --> 00:13:04,960
I don't have conclusive results yet, but I am trying.

150
00:13:04,960 --> 00:13:09,840
And I'm having a lot of fun because I love doing this type of research.

151
00:13:09,840 --> 00:13:11,540
That's fantastic.

152
00:13:11,740 --> 00:13:13,960
That's the right way to take that.

153
00:13:14,560 --> 00:13:18,860
Let's check, do we have any good questions coming in?

154
00:13:20,200 --> 00:13:23,060
Not seeing too many questions coming in.

155
00:13:23,060 --> 00:13:28,340
A lot of people are talking about some of the data dumps that have been coming through.

156
00:13:28,840 --> 00:13:33,420
Many thousands of fingerprints getting released.

157
00:13:34,100 --> 00:13:38,800
That could be a problem with government and federal contractors.

158
00:13:39,320 --> 00:13:42,680
Fingerprints getting released to the public as well.

159
00:13:42,680 --> 00:13:47,340
Which is a really interesting thing to direct the conversation towards.

160
00:13:47,340 --> 00:13:50,680
As more and more people are...

161
00:13:52,120 --> 00:13:54,900
As more of this data is out there and available.

162
00:13:55,420 --> 00:14:01,460
The type of research you are doing is probably more important for some of these companies.

163
00:14:01,460 --> 00:14:05,300
It's that old statement, I think you even said it in your talk.

164
00:14:05,300 --> 00:14:09,700
That once your biometric data is out there in the world.

165
00:14:09,700 --> 00:14:13,500
You only have 10 fingers.

166
00:14:13,620 --> 00:14:14,620
You can't reset it.

167
00:14:14,620 --> 00:14:16,660
You can't reset those.

168
00:14:17,000 --> 00:14:23,960
I think that my ocular scan is owned by Microsoft somewhere.

169
00:14:24,480 --> 00:14:31,100
There is only a limited number of options that you have.

170
00:14:31,820 --> 00:14:41,940
You are definitely going in the direction of the physical manifestation of these rebuilding things physically.

171
00:14:43,200 --> 00:14:45,100
What would be...

172
00:14:46,740 --> 00:14:53,300
You already mentioned that having a good microscope would help you approach the next...

173
00:14:53,820 --> 00:14:56,120
Whatever the next milestone is on this.

174
00:14:56,120 --> 00:15:00,360
If you can't really measure the ridge height, then you are having a difficult time.

175
00:15:01,300 --> 00:15:03,980
Determining which ones are going to work or not.

176
00:15:04,300 --> 00:15:09,340
What other pieces of technology do you think you'd want in order to move your research further?

177
00:15:11,220 --> 00:15:16,540
I think for the resolution of the printing.

178
00:15:16,540 --> 00:15:21,780
I think with my 3D printer it's okay.

179
00:15:22,340 --> 00:15:26,980
Perhaps with a better 3D printer.

180
00:15:26,980 --> 00:15:33,880
We will succeed in less tries.

181
00:15:33,880 --> 00:15:37,720
But also the microscope will be good.

182
00:15:37,720 --> 00:15:41,440
Because the worst problem was the ridge height.

183
00:15:41,440 --> 00:15:43,060
I needed to try.

184
00:15:43,060 --> 00:15:47,000
I know the normal human ridge height.

185
00:15:47,000 --> 00:15:54,840
But I need to try a lot to find out the exact one to duplicate the fingerprint.

186
00:15:54,840 --> 00:15:58,540
So that was a lot of tries.

187
00:15:58,640 --> 00:16:00,120
That makes sense.

188
00:16:00,300 --> 00:16:02,380
We did have one question come in.

189
00:16:03,520 --> 00:16:09,340
Lefebunachi wants to know, do you think you'll be back next year to discuss facial recognition?

190
00:16:09,460 --> 00:16:17,080
And what might help you to be able to do enough research to be able to present on that next year?

191
00:16:18,300 --> 00:16:19,040
No.

192
00:16:21,080 --> 00:16:22,280
Yes, I don't know.

193
00:16:22,280 --> 00:16:23,900
I am trying.

194
00:16:23,900 --> 00:16:29,700
Now we are working with the masks and the 3D printed head.

195
00:16:30,000 --> 00:16:31,680
We need a scanner.

196
00:16:31,680 --> 00:16:33,540
It's another technology.

197
00:16:33,540 --> 00:16:37,340
So we are trying to achieve that.

198
00:16:37,340 --> 00:16:39,680
But I think it could be possible.

199
00:16:39,680 --> 00:16:42,400
With this research or with others.

200
00:16:42,940 --> 00:16:48,600
It was always my dream to contribute with DEF CON in some sort of way.

201
00:16:48,600 --> 00:16:52,500
And I'm very happy to present a talk this year.

202
00:16:53,860 --> 00:16:57,500
Well, you definitely hit the ground running with this one.

203
00:16:57,680 --> 00:17:00,880
I really like your research and the way you put this together.

204
00:17:01,600 --> 00:17:06,780
I love that Palavrity just came up with this idea to share with me.

205
00:17:07,140 --> 00:17:18,480
And I think you actually talked about this a bit in your talk of creating fake fingerprints for yourself so you could be somebody else to an effect.

206
00:17:18,600 --> 00:17:24,800
Would you talk a little bit about the TSA agents in some other country?

207
00:17:24,800 --> 00:17:32,780
Was it Argentinian TSA agents or somebody that got fired for using fake fingers?

208
00:17:33,600 --> 00:17:34,200
Yes.

209
00:17:34,200 --> 00:17:42,640
Here in Argentina, the local airline fired, I think it was six people.

210
00:17:42,640 --> 00:17:45,180
Because they were using silicone.

211
00:17:45,180 --> 00:17:48,940
I am going to show... I have now the silicone.

212
00:17:49,240 --> 00:17:54,700
This is just a silicone fingerprint.

213
00:17:54,700 --> 00:18:02,320
But they use a silicone finger to impersonate the different users.

214
00:18:02,320 --> 00:18:09,880
To try to... I don't know how to say the word.

215
00:18:09,880 --> 00:18:16,580
But they try to falsify their entry to work.

216
00:18:17,120 --> 00:18:20,340
The different people.

217
00:18:20,340 --> 00:18:22,320
So , only one of them...

218
00:18:23,400 --> 00:18:25,560
If show time is six o'clock.

219
00:18:25,560 --> 00:18:27,300
Show to people.

220
00:18:27,300 --> 00:18:29,320
Show to the work, sorry.

221
00:18:29,320 --> 00:18:33,600
And with the other five people, silicone fingers.

222
00:18:33,940 --> 00:18:34,820
Wow.

223
00:18:35,940 --> 00:18:37,380
That's fascinating.

224
00:18:37,460 --> 00:18:43,500
Because you always heard of times when one person would go into work and just pull the punch cards for everybody and put those in.

225
00:18:43,500 --> 00:18:46,380
So they thought they solved it by forcing the fingerprint.

226
00:18:46,380 --> 00:18:51,460
And it sounds as though they were like, yeah, we'll just do silicone reprints of everyone's fingerprints.

227
00:18:51,780 --> 00:18:52,460
Sign them in.

228
00:18:52,460 --> 00:19:05,560
Another interesting side of this story is when I was showing this research to friends and people, a lot of them asked me to reproduce their finger for this type of...

229
00:19:05,560 --> 00:19:07,400
Of course, I don't.

230
00:19:07,400 --> 00:19:09,220
Of course, I don't.

231
00:19:09,220 --> 00:19:20,280
But a lot of people asked me, especially in optical sensors, asked me for a silicone copy of their...

232
00:19:20,280 --> 00:19:24,420
But no, don't DM me asking for that.

233
00:19:24,420 --> 00:19:26,840
Well, I mean, that is the next question, right?

234
00:19:26,840 --> 00:19:32,340
If you've done a bunch of these different systems, are you talking about an optical fingerprint scanner?

235
00:19:32,340 --> 00:19:34,820
Or are you talking about the retina scanner?

236
00:19:34,980 --> 00:19:39,900
Because I would really like to know how you would attack something like a retina scanner.

237
00:19:40,860 --> 00:19:47,400
Yes, we are not attacking iris yet.

238
00:19:47,400 --> 00:19:48,020
Not yet.

239
00:19:48,020 --> 00:19:53,000
And also the finger vein.

240
00:19:53,000 --> 00:20:04,510
I was trying to get one of these.

241
00:20:05,540 --> 00:20:11,300
I'm talking to the vendor, but not yet.

242
00:20:11,300 --> 00:20:12,640
You haven't managed it yet.

243
00:20:12,640 --> 00:20:20,520
Well, maybe someone who has some access to that is listening now and will go, Oh, well, we need to make sure you get that.

244
00:20:20,520 --> 00:20:27,580
So hopefully that's something that will come of your DEF CON talk time.

245
00:20:29,080 --> 00:20:42,620
So, let's see, people are doing more commentary in the chat channel about some of the relative comfort or how often some of these different biometric systems are in use out in the world.

246
00:20:42,640 --> 00:20:44,980
That's kind of interesting.

247
00:20:45,520 --> 00:20:49,240
So, let's see, you have your collection over here.

248
00:20:49,240 --> 00:20:51,220
You have your silicone fingers.

249
00:20:51,220 --> 00:20:55,140
Yes, I can show some of these.

250
00:20:55,140 --> 00:21:01,800
I have here the alginate molds.

251
00:21:01,800 --> 00:21:06,740
It's like the same alginate that they use for dental molds.

252
00:21:07,740 --> 00:21:20,440
Also, I have to buy the alginate in a dental office, so it was complicated.

253
00:21:20,440 --> 00:21:28,900
Also, the fingerprint powder, I have problems to get that.

254
00:21:29,580 --> 00:21:33,980
You probably have to answer some really interesting questions, I'm assuming.

255
00:21:33,980 --> 00:21:35,220
Yes.

256
00:21:35,860 --> 00:21:44,540
Also, I have here, the best molds are the hot glue molds, because they are really, really detailed.

257
00:21:44,540 --> 00:21:50,820
Also, the alginate molds are very good, but the hot glue molds are better.

258
00:21:50,820 --> 00:22:01,280
And for the fake fingerprints, material that worked better for us was latex.

259
00:22:01,280 --> 00:22:06,300
I use liquid latex, you can see here, it's really thin, really, really thin.

260
00:22:07,580 --> 00:22:10,120
But you can use a school glue also.

261
00:22:10,120 --> 00:22:16,280
I use liquid latex, the same one that they use for special effects.

262
00:22:16,640 --> 00:22:22,380
I think it will be better, but I think with the school glue, it's enough.

263
00:22:23,500 --> 00:22:32,740
Silicone works great in the optical scanners, but not in capacitive, because silicone is insulating, so no.

264
00:22:33,160 --> 00:22:38,580
For the capacitive scanners, we need to use latex.

265
00:22:38,580 --> 00:22:47,160
And with your real finger behind, the conductivity is enough to fool the scanner.

266
00:22:47,160 --> 00:22:48,820
Interesting.

267
00:22:48,820 --> 00:22:53,900
Yes, that part is really interesting.

268
00:22:58,470 --> 00:23:00,920
Ah, wood glue.

269
00:23:01,040 --> 00:23:04,660
I also use wood glue, it's very thin also.

270
00:23:04,660 --> 00:23:05,680
Which kind of glue?

271
00:23:05,680 --> 00:23:07,440
Wood glue.

272
00:23:07,440 --> 00:23:20,840
Wood glue works great, but you can use it one or two times and then it gets rigid and it won't work anymore.

273
00:23:20,840 --> 00:23:26,020
But latex, you can use it several times for a lot of time.

274
00:23:26,020 --> 00:23:27,580
And silicone also.

275
00:23:27,580 --> 00:23:38,940
Okay, so it looks like a lot of what you've just gone over are things that you need the assistance of the person who has the fingerprint in order to build this.

276
00:23:38,940 --> 00:23:40,200
Is that correct?

277
00:23:40,200 --> 00:23:48,900
How many of these methods can you, if you lift a fingerprint off of a glass or a picture like you were mentioning earlier?

278
00:23:50,620 --> 00:23:58,960
Yes, the 3D printing part was only on a non-cooperative way.

279
00:23:58,960 --> 00:24:13,620
We obtained the fingerprints in glass or an inked fingerprint, one that you can obtain in the police, that type of inked fingerprint.

280
00:24:13,620 --> 00:24:19,700
So we used that to create the 3D printed mold.

281
00:24:20,440 --> 00:24:25,080
We created a mold, a negative and a positive.

282
00:24:25,360 --> 00:24:32,980
The positive, because the resin is hard, you can see here that I can't bend it.

283
00:24:34,020 --> 00:24:46,120
So the positive worked only on one of the optical scanners and in the ultrasonic, but it was not good.

284
00:24:46,120 --> 00:24:53,440
But the mold was great and we cast the 3D printed mold with liquid latex.

285
00:24:53,440 --> 00:25:04,660
This work on all the type of sensors that we try in the capacitive, in the optical and also in the ultrasonic.

286
00:25:05,440 --> 00:25:07,340
Got it, all of them.

287
00:25:07,600 --> 00:25:13,420
How many different manufacturers are there of these different fingerprint scanners?

288
00:25:13,420 --> 00:25:16,720
And how many of them have you had an opportunity to test on?

289
00:25:18,060 --> 00:25:23,960
I tested only four, but I think there are a lot.

290
00:25:25,860 --> 00:25:33,500
Where do you intend to take this portion of your experimentation as you go forward?

291
00:25:33,500 --> 00:25:41,660
I know I asked a little bit of this question before, but we've already covered there's a few gaps that would be interesting to fill.

292
00:25:41,660 --> 00:25:43,200
What is next?

293
00:25:43,200 --> 00:25:44,500
Where do you go?

294
00:25:48,380 --> 00:25:49,380
I've lost you.

295
00:25:49,380 --> 00:26:08,800
Another attack vector that I didn't try is obtaining a fingerprint from, for example, a liquid biometric database and from the sensor, directly from the sensor.

296
00:26:08,800 --> 00:26:16,910
I think that in this way, perhaps we don't need to enhance the fingerprint.

297
00:26:17,700 --> 00:26:29,140
And we can't, because for all of that, we need an enhancement process because the obtained fingerprints were not directly actionable.

298
00:26:29,140 --> 00:26:34,140
We need to enhance it in order to do the mold.

299
00:26:34,140 --> 00:26:42,060
But in that case, I think it could be possible to make a mold without the enhancement process.

300
00:26:42,060 --> 00:26:44,780
So I would like to try that.

301
00:26:45,620 --> 00:26:46,200
That's good.

302
00:26:46,280 --> 00:26:47,940
Yeah, I like that.

303
00:26:48,100 --> 00:26:52,120
Well, we are rapidly approaching the end of our allotted time.

304
00:26:52,760 --> 00:26:58,020
Plarity, did you see any extra questions in the chat that we should hit?

305
00:26:58,960 --> 00:27:01,940
Yeah, I think there's probably two that have come in.

306
00:27:01,940 --> 00:27:08,820
One from the ironically named Dexterous said that I've seen biometric hand scanners.

307
00:27:08,820 --> 00:27:14,540
Not sure how they work, but I've noticed that when a person loses a lot of weight, they have to get the handprint redone.

308
00:27:14,540 --> 00:27:16,400
Does that happen with fingerprints?

309
00:27:16,520 --> 00:27:21,020
And if it does, could you just resize a copy of the fingerprint to work correctly?

310
00:27:23,340 --> 00:27:25,000
I didn't understand.

311
00:27:25,000 --> 00:27:32,860
I understand that it's a hand, like all the fingerprints are scanned at the same time?

312
00:27:33,060 --> 00:27:39,080
Yes, they're saying if they do the whole hand and then somebody loses a lot of weight, they have to redo it.

313
00:27:39,080 --> 00:27:42,500
And do you think it would be the same thing with just a finger?

314
00:27:43,300 --> 00:27:56,740
It could be because the reach height, sorry, the reach width varies with the health.

315
00:27:56,740 --> 00:28:04,860
And I can see complexity of the people and it could be the same.

316
00:28:04,860 --> 00:28:06,560
It's interesting to try.

317
00:28:06,560 --> 00:28:09,860
I will love to try that.

318
00:28:09,860 --> 00:28:14,920
And have you also looked at how fingerprint data is stored in the devices?

319
00:28:14,920 --> 00:28:22,900
Is it just stored with numbers or actually, how do devices store fingerprint data?

320
00:28:22,920 --> 00:28:28,840
The devices that I analyze, they are images.

321
00:28:29,380 --> 00:28:33,080
A biometric database with images.

322
00:28:33,080 --> 00:28:35,480
So that's what's the interesting part.

323
00:28:35,480 --> 00:28:42,600
I want to try if you can obtain in some way the database.

324
00:28:42,780 --> 00:28:50,640
I would like to try and see if you can reproduce the fingerprint from that data.

325
00:28:50,880 --> 00:28:53,140
I think it's interesting too.

326
00:28:53,380 --> 00:28:55,940
That sounds like a wonderful place to aim next.

327
00:28:55,940 --> 00:28:57,640
So we are right at the end.

328
00:28:57,640 --> 00:29:04,140
I really appreciate your willingness to come and give the presentation and spend some extra time with us in this Q&A.

329
00:29:04,140 --> 00:29:20,140
If you are willing to do so, we like to have presenters drop into the track one at the end and give contact information, Twitter or websites or anything if somebody wants to reach out and address you afterwards.

330
00:29:20,560 --> 00:29:29,800
And then I would really like to ask you one final question of what would be the big takeaway that you would like us to have to grab from your talk?

331
00:29:30,040 --> 00:29:32,540
What's the big punchline here?

332
00:29:33,060 --> 00:29:51,140
I think the big takeaway is geometric systems in a special fingerprint are not secure enough for using it as an authentication method.

333
00:29:51,140 --> 00:30:09,060
I know that for regular people, perhaps it's not a problem, but with different treat models, other people, more exposed people, it's just not a good way to authenticate to the devices.

334
00:30:09,060 --> 00:30:22,080
Because if a printer in my home can duplicate the fingerprints with more resources, then it could...

335
00:30:23,080 --> 00:30:27,120
It's not secure enough for depending upon your threat model.

336
00:30:27,120 --> 00:30:28,040
Okay, excellent.

337
00:30:28,040 --> 00:30:29,840
Well, thank you so much.

338
00:30:29,840 --> 00:30:37,580
If anyone has any additional questions, they are welcome to reach out and there will be some contact information hitting track one soon.

339
00:30:37,580 --> 00:30:40,980
Otherwise, have a great rest of your convention.

340
00:30:41,200 --> 00:30:44,420
I wish you the best and hope to see more from you later.

341
00:30:46,240 --> 00:30:47,360
Thank you.