WEBVTT

1
00:00:00.080 --> 00:00:05.080
<v Speaker 1>Okay, picture this a world where the objects around you,

2
00:00:05.160 --> 00:00:07.879
<v Speaker 1>maybe like the groceries in your fridge, the equipment in

3
00:00:07.919 --> 00:00:12.439
<v Speaker 1>your office, even that library book, they aren't just inanimate things.

4
00:00:13.279 --> 00:00:17.000
<v Speaker 1>They have a digital identity. They could potentially communicate, tell

5
00:00:17.039 --> 00:00:20.199
<v Speaker 1>you where they've been, what temperature they've hit, maybe even

6
00:00:20.280 --> 00:00:21.640
<v Speaker 1>you know, if they've been pampered with.

7
00:00:21.920 --> 00:00:24.719
<v Speaker 2>It sounds almost like sci fi, doesn't it. But the

8
00:00:24.760 --> 00:00:28.039
<v Speaker 2>core technology is actually here, it's being deployed right now,

9
00:00:28.280 --> 00:00:31.280
<v Speaker 2>and it's well, it's evolving incredibly fast exactly.

10
00:00:31.320 --> 00:00:36.039
<v Speaker 1>And that technology it's called radio frequency identification or RFID.

11
00:00:36.520 --> 00:00:38.759
<v Speaker 1>So welcome to the deep dive today. We are taking

12
00:00:39.039 --> 00:00:42.320
<v Speaker 1>a really serious look at RFID. Yeah, it's time, and

13
00:00:42.359 --> 00:00:46.200
<v Speaker 1>to guide us, we're relying on a frankly really comprehensive source,

14
00:00:46.520 --> 00:00:49.039
<v Speaker 1>a summary report from a workshop. This is put together

15
00:00:49.079 --> 00:00:53.560
<v Speaker 1>by the National Academies, specifically their Computer Science and Telecommunications Board.

16
00:00:53.399 --> 00:00:55.960
<v Speaker 2>And it's great because it's not just one viewpoint. It

17
00:00:56.000 --> 00:01:00.799
<v Speaker 2>brings together experts from well everywhere industry, academia, government, nonprofits too,

18
00:01:00.920 --> 00:01:01.280
<v Speaker 2>give you.

19
00:01:01.200 --> 00:01:03.560
<v Speaker 1>That balanced snapshot right of where the tech is and

20
00:01:03.560 --> 00:01:04.959
<v Speaker 1>the big, big questions around it.

21
00:01:05.120 --> 00:01:05.719
<v Speaker 2>Exactly.

22
00:01:06.239 --> 00:01:08.439
<v Speaker 1>So our mission for this deep dive is basically to

23
00:01:08.560 --> 00:01:12.640
<v Speaker 1>unpack what RFID actually is, how it works, you know,

24
00:01:12.680 --> 00:01:16.400
<v Speaker 1>the fundamentals, where it's currently being used or seriously considered,

25
00:01:16.560 --> 00:01:19.640
<v Speaker 1>and maybe most importantly, dig into some of the really

26
00:01:19.680 --> 00:01:24.079
<v Speaker 1>significant implications it has for society, for culture.

27
00:01:24.280 --> 00:01:27.319
<v Speaker 2>Think of it as a shortcut maybe to understanding this

28
00:01:27.719 --> 00:01:30.799
<v Speaker 2>tech that could become well pretty much everywhere.

29
00:01:30.959 --> 00:01:33.599
<v Speaker 1>Let's kick things off by just breaking down the technology itself,

30
00:01:33.599 --> 00:01:34.439
<v Speaker 1>the nuts and bolts.

31
00:01:34.519 --> 00:01:36.760
<v Speaker 2>Okay, sounds good, So at its.

32
00:01:36.599 --> 00:01:41.400
<v Speaker 1>Heart, RFID is fundamentally about using radio waves right to

33
00:01:41.560 --> 00:01:42.400
<v Speaker 1>identify things.

34
00:01:42.519 --> 00:01:45.959
<v Speaker 2>Yeah, it's like giving objects a unique electronic voice, a

35
00:01:46.079 --> 00:01:47.719
<v Speaker 2>voice that can be heard wirelessly.

36
00:01:47.920 --> 00:01:49.200
<v Speaker 1>Right, And if you think about how we used to

37
00:01:49.280 --> 00:01:53.359
<v Speaker 1>identify things, well, first visually, then barcodes came along, huge step,

38
00:01:53.519 --> 00:01:56.120
<v Speaker 1>huge step. Yeah, but barcodes need that line of sight.

39
00:01:56.159 --> 00:01:58.000
<v Speaker 1>You've got to point the scanner right at the code.

40
00:01:58.040 --> 00:02:02.239
<v Speaker 2>And that's the key difference with RFID, that's the fundamental advantage. Yeah,

41
00:02:02.319 --> 00:02:04.480
<v Speaker 2>it just gets rid of that line of sight requirement,

42
00:02:04.920 --> 00:02:09.120
<v Speaker 2>so you can potentially identify stuff inside boxes or through packaging,

43
00:02:09.199 --> 00:02:11.719
<v Speaker 2>even if it's dirty or blocked.

44
00:02:12.000 --> 00:02:15.039
<v Speaker 1>That non line of sight thing that feels really powerful.

45
00:02:15.319 --> 00:02:18.000
<v Speaker 1>And the report breaks down the basic system into what

46
00:02:18.159 --> 00:02:19.680
<v Speaker 1>three key parts.

47
00:02:19.240 --> 00:02:22.560
<v Speaker 2>Correct, three main components. First, you've got the tag. This

48
00:02:22.639 --> 00:02:25.879
<v Speaker 2>is the little electronic bit that you attached to the

49
00:02:25.919 --> 00:02:28.719
<v Speaker 2>object you want to track. It's got an antenna, usually

50
00:02:28.719 --> 00:02:32.840
<v Speaker 2>a microchip, and it holds information, often just a unique number.

51
00:02:33.039 --> 00:02:35.240
<v Speaker 1>Got it tag on the object? What's second?

52
00:02:35.560 --> 00:02:37.919
<v Speaker 2>Second is the reader. That's the device sending up the

53
00:02:37.960 --> 00:02:41.240
<v Speaker 2>radio signals, talking to the tags and listening for their responses.

54
00:02:41.280 --> 00:02:43.680
<v Speaker 1>And the third part is everything.

55
00:02:43.319 --> 00:02:47.240
<v Speaker 2>Else, pretty much the broader hardware and software environment, everything

56
00:02:47.319 --> 00:02:49.879
<v Speaker 2>needed to make sense of the data coming from the readers,

57
00:02:50.199 --> 00:02:54.120
<v Speaker 2>get it into databases, integrate it with other systems, the infrastructure.

58
00:02:54.240 --> 00:02:56.439
<v Speaker 1>Now here's where you said it gets really interesting. Saying

59
00:02:56.560 --> 00:03:00.599
<v Speaker 1>RFID tag is like saying car. There are loads of different.

60
00:03:00.400 --> 00:03:06.639
<v Speaker 2>Kinds exactly, a whole spectrum, wildly different capabilities, different power sources, everything, And.

61
00:03:06.639 --> 00:03:10.080
<v Speaker 1>The simplest, maybe the most common type, is the passive tag.

62
00:03:10.199 --> 00:03:13.000
<v Speaker 2>Passive tags, Yeah, they're actually quite clever. They don't have

63
00:03:13.039 --> 00:03:13.960
<v Speaker 2>a battery.

64
00:03:14.439 --> 00:03:15.960
<v Speaker 1>At all, So how do they work.

65
00:03:16.319 --> 00:03:20.199
<v Speaker 2>They literally harvest the energy they need from the radio

66
00:03:20.280 --> 00:03:21.560
<v Speaker 2>signal the reader sends.

67
00:03:21.360 --> 00:03:24.599
<v Speaker 1>Out, So reader sends out energy. The tag collects just

68
00:03:24.719 --> 00:03:27.280
<v Speaker 1>enough to power up and ping back its id.

69
00:03:27.520 --> 00:03:30.680
<v Speaker 2>That's the basic idea. But this whole power hardesting thing,

70
00:03:30.800 --> 00:03:33.840
<v Speaker 2>it really limits their range fundamentally.

71
00:03:33.919 --> 00:03:34.960
<v Speaker 1>How limited are we talking?

72
00:03:35.240 --> 00:03:38.159
<v Speaker 2>Well, the range for these purely passive tags is pretty short,

73
00:03:38.479 --> 00:03:42.080
<v Speaker 2>usually near contact like maybe swiping a key card, up

74
00:03:42.159 --> 00:03:46.280
<v Speaker 2>to maybe maybe ten or fifteen meters in perfect lab conditions,

75
00:03:46.400 --> 00:03:48.000
<v Speaker 2>often much less in the real world.

76
00:03:48.400 --> 00:03:50.680
<v Speaker 1>And there's some physics behind why the signal drops off

77
00:03:50.680 --> 00:03:51.080
<v Speaker 1>so fast.

78
00:03:51.120 --> 00:03:54.360
<v Speaker 2>Oh yeah, it's steep, much steeper than a normal radio signal,

79
00:03:54.439 --> 00:03:56.759
<v Speaker 2>like steeper well a normal signal, it's power drops off

80
00:03:56.800 --> 00:03:58.599
<v Speaker 2>with the square of the distance you know, one over

81
00:03:58.680 --> 00:04:01.599
<v Speaker 2>D squared right, or a passive tag. Think about it.

82
00:04:01.919 --> 00:04:03.599
<v Speaker 2>The signal has to go from the reader to the

83
00:04:03.680 --> 00:04:06.960
<v Speaker 2>tag just to power it up. Then the tag has

84
00:04:07.000 --> 00:04:09.199
<v Speaker 2>to generate its own little signal and send it back

85
00:04:09.199 --> 00:04:12.479
<v Speaker 2>to the reader. So you've got distance impacting the power

86
00:04:12.520 --> 00:04:15.599
<v Speaker 2>going out and distance impacting the signal coming back.

87
00:04:15.719 --> 00:04:18.959
<v Speaker 1>Ah, so it gets hit twice by the distance effect.

88
00:04:18.839 --> 00:04:22.199
<v Speaker 2>Exactly, which means the power of the reader actually gets

89
00:04:22.240 --> 00:04:24.959
<v Speaker 2>back falls off with the fourth power of the distance

90
00:04:25.439 --> 00:04:27.160
<v Speaker 2>one over D to the fourth.

91
00:04:27.000 --> 00:04:30.480
<v Speaker 1>Wow one day. So doubling the distance doesn't cut the

92
00:04:30.519 --> 00:04:32.199
<v Speaker 1>signal to a quarter, it cuts it to a.

93
00:04:32.279 --> 00:04:35.959
<v Speaker 2>Sixteenth precisely that one odo fall off. That's a basic

94
00:04:36.000 --> 00:04:39.800
<v Speaker 2>physical limit. It explains why passive tags are great for, say,

95
00:04:40.199 --> 00:04:42.680
<v Speaker 2>tracking items on a conveyor belt or things within a

96
00:04:42.720 --> 00:04:43.319
<v Speaker 2>few feet in.

97
00:04:43.279 --> 00:04:45.399
<v Speaker 1>A warehouse, but not for long rain stuff.

98
00:04:45.480 --> 00:04:47.560
<v Speaker 2>Nope, not suited for that at all. And the source

99
00:04:47.560 --> 00:04:49.720
<v Speaker 2>mentions other challenges too, right, like interference.

100
00:04:49.800 --> 00:04:53.480
<v Speaker 1>Yeah, and materials like metal and water just absorbing the signal.

101
00:04:53.600 --> 00:04:56.720
<v Speaker 2>Yes, those are huge real world problems trying to tag

102
00:04:56.839 --> 00:05:00.519
<v Speaker 2>say a shopping cart full of groceries reliably. You got metal,

103
00:05:00.519 --> 00:05:03.120
<v Speaker 2>you've got liquids. The signal can get blocked or just

104
00:05:03.360 --> 00:05:05.160
<v Speaker 2>soaked up. Makes it tricky.

105
00:05:05.480 --> 00:05:08.800
<v Speaker 1>But you said there's a spectrum. So are some passive

106
00:05:08.839 --> 00:05:12.480
<v Speaker 1>tags smarter than others? Can they do more than just

107
00:05:12.519 --> 00:05:13.240
<v Speaker 1>send an ID?

108
00:05:13.519 --> 00:05:16.360
<v Speaker 2>That's true? Yeah, The report points out some passive tags

109
00:05:16.399 --> 00:05:19.680
<v Speaker 2>can have like extended memory, so they can store more

110
00:05:19.759 --> 00:05:22.319
<v Speaker 2>data about the object, not just a serial number.

111
00:05:22.519 --> 00:05:22.839
<v Speaker 1>Okay.

112
00:05:22.959 --> 00:05:27.000
<v Speaker 2>Others might even have really simple sensing, like detecting if

113
00:05:27.000 --> 00:05:29.600
<v Speaker 2>a package seal was broken or if it went above

114
00:05:29.639 --> 00:05:32.480
<v Speaker 2>a certain temperature, still powered by the reader, but they

115
00:05:32.519 --> 00:05:34.800
<v Speaker 2>capture and store that extra bit of info.

116
00:05:35.120 --> 00:05:40.560
<v Speaker 1>Okay, So passive tags energy harvesting short range because of

117
00:05:40.560 --> 00:05:43.319
<v Speaker 1>that one journal limit. What's the next level up?

118
00:05:43.519 --> 00:05:47.720
<v Speaker 2>That would be tags with their own power source. Active tags.

119
00:05:48.000 --> 00:05:49.959
<v Speaker 1>Ah, so they have a battery.

120
00:05:49.399 --> 00:05:53.759
<v Speaker 2>Exactly, and that mattery gives them much much longer communication range.

121
00:05:53.800 --> 00:05:57.120
<v Speaker 2>We're talking maybe up to several hundred meters for some

122
00:05:57.199 --> 00:05:58.079
<v Speaker 2>of the high end ones.

123
00:05:58.160 --> 00:06:00.600
<v Speaker 1>And the power fall off is better then, because they're

124
00:06:00.600 --> 00:06:01.600
<v Speaker 1>sending their own signal.

125
00:06:01.759 --> 00:06:04.800
<v Speaker 2>Yes, because they power their own transmission, the signal strength

126
00:06:04.920 --> 00:06:06.959
<v Speaker 2>just falls off as the square of the distance one

127
00:06:07.000 --> 00:06:08.879
<v Speaker 2>d donner like a regular radio signal.

128
00:06:08.920 --> 00:06:10.920
<v Speaker 1>Okay, that makes a big difference for range.

129
00:06:10.879 --> 00:06:14.519
<v Speaker 2>Huge difference makes them suitable for applications needing that longer

130
00:06:14.560 --> 00:06:17.959
<v Speaker 2>reach or maybe more reliable communication and tough environments.

131
00:06:18.120 --> 00:06:20.079
<v Speaker 1>And I guess with a battery they can do more

132
00:06:20.759 --> 00:06:22.079
<v Speaker 1>like sensing over.

133
00:06:21.879 --> 00:06:26.240
<v Speaker 2>Time precisely, Active tags, can you know? Sample sensors continuously

134
00:06:26.519 --> 00:06:29.079
<v Speaker 2>store data logs over time and then transmit that data

135
00:06:29.120 --> 00:06:30.360
<v Speaker 2>when a reader asks for it.

136
00:06:30.639 --> 00:06:33.160
<v Speaker 1>The report mentions some could even talk to each other

137
00:06:33.279 --> 00:06:33.920
<v Speaker 1>peer to peer.

138
00:06:34.160 --> 00:06:37.279
<v Speaker 2>It does mention that possibility. Yeah, with some advanced types

139
00:06:37.279 --> 00:06:40.720
<v Speaker 2>like Class four tags, they could potentially act like nodes

140
00:06:40.720 --> 00:06:44.240
<v Speaker 2>in a sensor network. Though honestly, most r FID setups

141
00:06:44.240 --> 00:06:46.879
<v Speaker 2>you see today still have the tags talking directly back

142
00:06:46.879 --> 00:06:49.279
<v Speaker 2>to a reader, like a star pattern, not a mesh.

143
00:06:49.519 --> 00:06:52.959
<v Speaker 1>Makes sense now thinking about costs, Yeah, the report says

144
00:06:52.959 --> 00:06:54.079
<v Speaker 1>it varies wildly. Oh.

145
00:06:54.079 --> 00:06:57.079
<v Speaker 2>Absolutely. You can get simple passive tags, sometimes just called

146
00:06:57.120 --> 00:07:00.000
<v Speaker 2>inlays for you know, less than a dollar, maybe even

147
00:07:00.160 --> 00:07:02.759
<v Speaker 2>just sends in huge volumes. But then you go up

148
00:07:02.800 --> 00:07:06.720
<v Speaker 2>to sophisticated active tags with batteries, multiple sensors, secure memory.

149
00:07:07.360 --> 00:07:09.319
<v Speaker 2>Those can be hundreds of dollars each.

150
00:07:09.240 --> 00:07:11.879
<v Speaker 1>And the main cost isn't the chip itself anymore.

151
00:07:11.639 --> 00:07:15.279
<v Speaker 2>Not usually no, silicon keeps getting cheaper. The bigger cost

152
00:07:15.360 --> 00:07:18.360
<v Speaker 2>driver is often connecting that tiny chip to the antenna

153
00:07:18.399 --> 00:07:21.240
<v Speaker 2>reliably and then packaging the whole thing so it survives

154
00:07:21.279 --> 00:07:22.000
<v Speaker 2>in its environment.

155
00:07:22.079 --> 00:07:23.319
<v Speaker 1>And they can be tiny, right.

156
00:07:23.199 --> 00:07:26.920
<v Speaker 2>Incredibly small. Sometimes the chip and antenna assembly can weigh

157
00:07:26.920 --> 00:07:30.279
<v Speaker 2>fractions of a gram. The antenna's themselves vary a lot, though,

158
00:07:30.319 --> 00:07:33.959
<v Speaker 2>from submillimeter for very short range up to maybe dozens

159
00:07:33.959 --> 00:07:37.040
<v Speaker 2>of centimeters for longer range. UHF tags, and.

160
00:07:37.000 --> 00:07:38.600
<v Speaker 1>They could be put in different packages.

161
00:07:38.839 --> 00:07:43.959
<v Speaker 2>Yeah, flexible labels like stickers, rigid plastic cases, or even

162
00:07:44.000 --> 00:07:47.959
<v Speaker 2>embedded in epoxy or glass for really harsh environments, like

163
00:07:48.279 --> 00:07:50.920
<v Speaker 2>the tags they inject into pets or livestock.

164
00:07:51.199 --> 00:07:54.759
<v Speaker 1>Right, and the report also puts RFID alongside other.

165
00:07:54.600 --> 00:07:56.639
<v Speaker 2>Tech, doesn't it It does, Yeah, it acknowledges it sits

166
00:07:56.680 --> 00:08:00.000
<v Speaker 2>in an ecosystem. You've got sensor networks doing related things.

167
00:08:00.120 --> 00:08:02.839
<v Speaker 2>Taxless smart cards which are kind of like advanced passive

168
00:08:02.839 --> 00:08:06.079
<v Speaker 2>tags but with more processing power, often used for payments

169
00:08:06.160 --> 00:08:06.839
<v Speaker 2>or secure ID.

170
00:08:07.199 --> 00:08:09.240
<v Speaker 1>And the barcode isn't going away anytime.

171
00:08:08.959 --> 00:08:11.800
<v Speaker 2>Soon, definitely not. Still the cheapest way to put an

172
00:08:11.839 --> 00:08:15.480
<v Speaker 2>identifier on something. The choice always depends on the application,

173
00:08:15.680 --> 00:08:18.079
<v Speaker 2>the value, what you actually need it to do.

174
00:08:18.480 --> 00:08:21.199
<v Speaker 1>Okay, so that covers the tags pretty well. Let's shift

175
00:08:21.240 --> 00:08:25.240
<v Speaker 1>focus to the other key piece, the reader.

176
00:08:25.600 --> 00:08:27.439
<v Speaker 2>Right. The reader it's the active part, the one that

177
00:08:27.439 --> 00:08:31.040
<v Speaker 2>starts the conversation. It sends out the power for passive tags.

178
00:08:31.120 --> 00:08:34.000
<v Speaker 2>It listens for the responses from all types of tags,

179
00:08:34.279 --> 00:08:37.080
<v Speaker 2>and critically, it has to manage when lots of tags

180
00:08:37.080 --> 00:08:37.799
<v Speaker 2>reply at once.

181
00:08:38.080 --> 00:08:41.080
<v Speaker 1>And it's the gateway to the information systems exactly.

182
00:08:41.159 --> 00:08:44.399
<v Speaker 2>It's the bridge getting that raw data off the tag

183
00:08:44.480 --> 00:08:47.200
<v Speaker 2>and into the databases, the software, the networks where it

184
00:08:47.200 --> 00:08:49.759
<v Speaker 2>can actually be used. That's the reader's job.

185
00:08:49.879 --> 00:08:53.919
<v Speaker 1>But readers have limitations too. The report mentions physics regulations.

186
00:08:54.000 --> 00:08:57.759
<v Speaker 2>Absolutely, physics limits things like how sensitive their receivers can

187
00:08:57.799 --> 00:09:00.320
<v Speaker 2>be or how much gain their antennas have based on

188
00:09:00.360 --> 00:09:02.840
<v Speaker 2>cost and size that impacts range.

189
00:09:02.919 --> 00:09:03.919
<v Speaker 1>And the regulations.

190
00:09:04.320 --> 00:09:08.279
<v Speaker 2>Yeah, governments regulate the radio frequencies and the power output

191
00:09:08.320 --> 00:09:12.679
<v Speaker 2>readers can use, especially in those unlicensed bans where RFID

192
00:09:12.799 --> 00:09:16.120
<v Speaker 2>often operates, like the ISM bands. They have to do

193
00:09:16.200 --> 00:09:19.159
<v Speaker 2>that to stop readers interfering with other radio devices.

194
00:09:19.200 --> 00:09:21.440
<v Speaker 1>And that's a big deal globally. Right, the rules aren't

195
00:09:21.440 --> 00:09:22.679
<v Speaker 1>the same everywhere.

196
00:09:22.440 --> 00:09:25.159
<v Speaker 2>Huge deal. A reader optimized for the rules in the US,

197
00:09:25.240 --> 00:09:28.120
<v Speaker 2>say at nine hundred bigahertz, might not perform the same

198
00:09:28.120 --> 00:09:30.519
<v Speaker 2>way or might even be illegal to operate at that

199
00:09:30.600 --> 00:09:33.840
<v Speaker 2>power level in Europe or Japan, where the specific frequency

200
00:09:33.879 --> 00:09:36.080
<v Speaker 2>allocations and power limits are different.

201
00:09:36.240 --> 00:09:38.960
<v Speaker 1>Okay, And how do readers handle that situation where you

202
00:09:39.039 --> 00:09:42.240
<v Speaker 1>might have like a whole palette of tagged goods, hundreds

203
00:09:42.279 --> 00:09:44.120
<v Speaker 1>of tags potentially shutting back at once.

204
00:09:44.519 --> 00:09:48.600
<v Speaker 2>They use something called arbitration protocols or sometimes anti collision protocols.

205
00:09:48.879 --> 00:09:51.120
<v Speaker 2>It's basically a set of rules, like a digital way

206
00:09:51.159 --> 00:09:54.279
<v Speaker 2>for the reader to say okay tags. Starting with a report, now,

207
00:09:54.360 --> 00:09:58.440
<v Speaker 2>then tags starting with B. It interrogates them systematically, one

208
00:09:58.480 --> 00:10:01.559
<v Speaker 2>by one or in small groups until it's identified all

209
00:10:01.559 --> 00:10:02.799
<v Speaker 2>the tags in its range.

210
00:10:02.840 --> 00:10:03.559
<v Speaker 1>But that takes time.

211
00:10:03.600 --> 00:10:06.679
<v Speaker 2>I guess it does, which is why getting super high

212
00:10:06.759 --> 00:10:09.600
<v Speaker 2>read rates reading thousands of items per second in really

213
00:10:09.679 --> 00:10:13.320
<v Speaker 2>dense pag populations is still a technical challenge. And you

214
00:10:13.399 --> 00:10:15.639
<v Speaker 2>also have to worry about readers interfering with each other.

215
00:10:15.679 --> 00:10:17.600
<v Speaker 2>If they're too close, you need coordination.

216
00:10:18.039 --> 00:10:20.799
<v Speaker 1>And if the tag itself is more complex, like it

217
00:10:20.840 --> 00:10:24.799
<v Speaker 1>has memory you can write too, or sensors, does the

218
00:10:24.799 --> 00:10:26.120
<v Speaker 1>reader need to be smarter.

219
00:10:25.919 --> 00:10:28.919
<v Speaker 2>Too, Yes, definitely. If a reader needs to write data

220
00:10:28.960 --> 00:10:33.240
<v Speaker 2>to a tag's memory, it needs security features built in passwords,

221
00:10:33.360 --> 00:10:36.600
<v Speaker 2>access control, something to make sure only authorized systems can

222
00:10:36.679 --> 00:10:39.240
<v Speaker 2>change the data on the tag makes sense. And if

223
00:10:39.279 --> 00:10:41.879
<v Speaker 2>it's reading data from sensor tags, the reader might need

224
00:10:41.919 --> 00:10:45.759
<v Speaker 2>more processing power itself, maybe to filter or aggregate that

225
00:10:45.840 --> 00:10:48.639
<v Speaker 2>sensor data before sending it upstream. It starts acting a

226
00:10:48.639 --> 00:10:50.559
<v Speaker 2>bit like a gateway node and a sensor network.

227
00:10:50.600 --> 00:10:53.679
<v Speaker 1>Okay, and readers for active tags most the batteries.

228
00:10:53.919 --> 00:10:56.720
<v Speaker 2>They generally operate over longer ranges thanks to that one

229
00:10:56.799 --> 00:10:58.879
<v Speaker 2>D dot off fall off we talked about, but they

230
00:10:58.919 --> 00:11:01.240
<v Speaker 2>also tend to be more expens of them passive readers.

231
00:11:01.480 --> 00:11:04.279
<v Speaker 2>The report gives a rough estimate maybe one thousand to

232
00:11:04.360 --> 00:11:07.720
<v Speaker 2>two thousand dollars for a long range active reader versus

233
00:11:07.759 --> 00:11:10.000
<v Speaker 2>maybe a few hundred up to one thousand for a

234
00:11:10.000 --> 00:11:13.440
<v Speaker 2>typical passive reader, and ten is also vary depending on

235
00:11:13.519 --> 00:11:15.240
<v Speaker 2>frequency and coverage needed.

236
00:11:15.759 --> 00:11:18.399
<v Speaker 1>So connecting that reader data to the rest of the

237
00:11:18.399 --> 00:11:21.840
<v Speaker 1>world of the internet company databases, that's a critical step.

238
00:11:21.600 --> 00:11:25.000
<v Speaker 2>Absolutely crucial and honestly often the most complex part of

239
00:11:25.080 --> 00:11:28.879
<v Speaker 2>rolling out a big RFID system. Getting that data reliably

240
00:11:28.919 --> 00:11:33.000
<v Speaker 2>from the physical tag through the reader across potentially different networks,

241
00:11:33.039 --> 00:11:36.200
<v Speaker 2>maybe across different companies in a supply chain that needs

242
00:11:36.240 --> 00:11:38.240
<v Speaker 2>serious software and integration work.

243
00:11:38.480 --> 00:11:41.720
<v Speaker 1>So putting it all together, tags, readers, software, you really

244
00:11:41.759 --> 00:11:42.559
<v Speaker 1>have to think about.

245
00:11:42.320 --> 00:11:45.480
<v Speaker 2>The whole system precisely. You choose the right tag based

246
00:11:45.480 --> 00:11:48.840
<v Speaker 2>on the assets value, right barcode for cheap stuff, maybe

247
00:11:48.879 --> 00:11:51.840
<v Speaker 2>a simple passive tag for a case, a tougher passive

248
00:11:51.879 --> 00:11:55.120
<v Speaker 2>tag for a Pallette, active tag, maybe with GPS for

249
00:11:55.279 --> 00:11:56.559
<v Speaker 2>a truck or container.

250
00:11:56.720 --> 00:11:58.480
<v Speaker 1>You have to consider the range you need, how the

251
00:11:58.519 --> 00:12:00.159
<v Speaker 1>tag might be oriented.

252
00:12:00.039 --> 00:12:04.399
<v Speaker 2>Exactly, reader density, how you'll cover the area, and critically

253
00:12:04.639 --> 00:12:07.360
<v Speaker 2>what happens to those tags after they leave your control,

254
00:12:07.679 --> 00:12:09.519
<v Speaker 2>like when a customer buys the item.

255
00:12:09.759 --> 00:12:12.399
<v Speaker 1>Yeah, that brings up those privacy questions again. The report

256
00:12:12.440 --> 00:12:14.799
<v Speaker 1>mentions tags might need to be killed at the point

257
00:12:14.799 --> 00:12:15.159
<v Speaker 1>of sale.

258
00:12:15.320 --> 00:12:19.840
<v Speaker 2>That's one approach discussed, Yeah, especially for consumer goods to

259
00:12:19.879 --> 00:12:24.080
<v Speaker 2>address privacy concerns. And location tracking comes into play too,

260
00:12:24.240 --> 00:12:27.679
<v Speaker 2>maybe using GPS on active tags, or even using things

261
00:12:27.720 --> 00:12:31.480
<v Speaker 2>like Wi Fi access points as impromptu readers for certain

262
00:12:31.480 --> 00:12:32.240
<v Speaker 2>types of tags.

263
00:12:32.360 --> 00:12:36.080
<v Speaker 1>Building the business case, then it means weighing all these costs.

264
00:12:35.799 --> 00:12:41.039
<v Speaker 2>Tags, readers, software integration, network, all of that against the

265
00:12:41.039 --> 00:12:42.279
<v Speaker 2>benefits you hope to get.

266
00:12:42.639 --> 00:12:45.559
<v Speaker 1>It requires careful thought, and that brings us to standards,

267
00:12:45.559 --> 00:12:48.879
<v Speaker 1>which you mentioned are vital for making this all work together, especially.

268
00:12:48.639 --> 00:12:52.519
<v Speaker 2>Globally, absolutely vital because, as we said, those radio regulations

269
00:12:52.519 --> 00:12:55.559
<v Speaker 2>for the unlicensed bands used by RFID, like the ISM

270
00:12:55.639 --> 00:12:58.519
<v Speaker 2>or SRD bands, they vary a lot around the world.

271
00:12:58.279 --> 00:13:00.240
<v Speaker 1>So a KAG and reader that work great in one

272
00:13:00.240 --> 00:13:02.720
<v Speaker 1>country might not work the same or even be legal

273
00:13:02.879 --> 00:13:03.360
<v Speaker 1>in another.

274
00:13:03.799 --> 00:13:07.639
<v Speaker 2>Exactly read raids might differ, ranges might change. For instance,

275
00:13:07.679 --> 00:13:10.360
<v Speaker 2>the nine hundred megahertz band has different rules in the

276
00:13:10.480 --> 00:13:11.240
<v Speaker 2>US compared to.

277
00:13:11.200 --> 00:13:15.399
<v Speaker 1>Europe, so standardization groups are key ISO, EPC Global right.

278
00:13:15.480 --> 00:13:18.360
<v Speaker 2>EPC Global grew out of the Autoid Center at MIT

279
00:13:19.000 --> 00:13:22.000
<v Speaker 2>and has been really influential. They developed protocols like Class

280
00:13:22.039 --> 00:13:25.519
<v Speaker 2>zero and Class one tags, and importantly the Gen two.

281
00:13:25.519 --> 00:13:29.639
<v Speaker 1>Standard and Gen two is becoming an international isostandard.

282
00:13:28.879 --> 00:13:32.679
<v Speaker 2>Yes, which is a huge step towards global compatibility, especially

283
00:13:32.720 --> 00:13:36.039
<v Speaker 2>for that nine hundred megahertz UHF frequency band.

284
00:13:36.279 --> 00:13:39.080
<v Speaker 1>Why did nine hundred megahurts become sort of the preferred

285
00:13:39.159 --> 00:13:42.000
<v Speaker 1>choice for supply chains despite those issues you mentioned with

286
00:13:42.080 --> 00:13:42.519
<v Speaker 1>metal and.

287
00:13:42.480 --> 00:13:45.080
<v Speaker 2>Liquids, it really hit a sweet spot offering a better

288
00:13:45.120 --> 00:13:48.120
<v Speaker 2>balance of read range and data speed compared to lower

289
00:13:48.159 --> 00:13:51.480
<v Speaker 2>frequencies like thirteen point five to six minute hurts. Even

290
00:13:51.480 --> 00:13:54.679
<v Speaker 2>with the challenges signal fading from multipath signals getting absorbed,

291
00:13:54.720 --> 00:13:57.879
<v Speaker 2>that potential for longer range reads and faster processing made

292
00:13:57.919 --> 00:14:00.919
<v Speaker 2>it attractive for tracking cases and palettes and big warehouses

293
00:14:00.960 --> 00:14:02.039
<v Speaker 2>and distribution centers.

294
00:14:02.080 --> 00:14:04.639
<v Speaker 1>And the mandates from big players really pushed it forward.

295
00:14:04.879 --> 00:14:09.159
<v Speaker 2>Oh absolutely, when companies like Walmart, Target, and the US

296
00:14:09.200 --> 00:14:12.159
<v Speaker 2>Department of Defense said to their suppliers, you need to

297
00:14:12.200 --> 00:14:16.279
<v Speaker 2>put RFID tags on your shipments, and often specified nine

298
00:14:16.360 --> 00:14:20.320
<v Speaker 2>hundred megahertz Gen two that created an enormous momentum for

299
00:14:20.399 --> 00:14:23.039
<v Speaker 2>adoption and standardization in that whole sector.

300
00:14:23.519 --> 00:14:26.440
<v Speaker 1>The report touches on US regulations from the SCC.

301
00:14:26.200 --> 00:14:29.159
<v Speaker 2>Too, Yeah, just outlining how they regulate that nine two

302
00:14:29.320 --> 00:14:33.279
<v Speaker 2>nine hundred eight megohurtz ban for unlicensed use, setting limits

303
00:14:33.320 --> 00:14:36.480
<v Speaker 2>on power emissions, how devices can operate to try and

304
00:14:36.559 --> 00:14:40.399
<v Speaker 2>minimize interference. Though managing interference when you have lots of

305
00:14:40.440 --> 00:14:43.840
<v Speaker 2>readers working close together is still a significant technical challenge.

306
00:14:43.919 --> 00:14:47.159
<v Speaker 1>So overall, standards are moving forward, but it's still maybe

307
00:14:47.159 --> 00:14:49.360
<v Speaker 1>a bit of a mixed bag, some vendor specific stuff

308
00:14:49.440 --> 00:14:50.039
<v Speaker 1>still out there.

309
00:14:50.120 --> 00:14:52.679
<v Speaker 2>That's the picture the report paints, Yeah, which can make

310
00:14:52.720 --> 00:14:54.720
<v Speaker 2>it harder for companies trying to figure out the costs

311
00:14:54.759 --> 00:14:57.320
<v Speaker 2>and benefits because you might be comparing systems that aren't

312
00:14:57.320 --> 00:14:59.679
<v Speaker 2>fully interchangeable. Yet it's definitely still evolving.

313
00:15:00.080 --> 00:15:03.279
<v Speaker 1>Okay, speaking of costs and benefits, let's dig into the applications.

314
00:15:03.480 --> 00:15:06.600
<v Speaker 1>Where is RFID actually being used and what are the

315
00:15:06.639 --> 00:15:08.960
<v Speaker 1>business cases. It seems like it's popping up all over.

316
00:15:09.279 --> 00:15:13.759
<v Speaker 2>The report really emphasizes the core power here. URFID makes

317
00:15:13.799 --> 00:15:17.960
<v Speaker 2>it relatively easy to add identification and communication to objects

318
00:15:18.279 --> 00:15:21.080
<v Speaker 2>to give them a digital presence, and that's valuable in

319
00:15:21.159 --> 00:15:22.639
<v Speaker 2>just so many different areas.

320
00:15:22.759 --> 00:15:26.279
<v Speaker 1>Supply chain management seems to be the absolute giant right now,

321
00:15:26.519 --> 00:15:28.360
<v Speaker 1>driven by those mandates.

322
00:15:28.080 --> 00:15:31.360
<v Speaker 2>Definitely the biggest driver. The potential benefits are just huge.

323
00:15:31.879 --> 00:15:35.240
<v Speaker 2>Better visibility of where your stuff is, reducing loss or

324
00:15:35.279 --> 00:15:38.559
<v Speaker 2>theft shrinkage. They call it, making receiving and shipping way

325
00:15:38.600 --> 00:15:42.440
<v Speaker 2>more efficient, getting goods through warehouses faster, improving the flow

326
00:15:42.480 --> 00:15:46.840
<v Speaker 2>of information, tracking perishables with sensor tags, getting automated proof

327
00:15:46.840 --> 00:15:47.440
<v Speaker 2>of delivery.

328
00:15:47.559 --> 00:15:50.600
<v Speaker 1>It's all about efficiency, transparency, automation exactly.

329
00:15:50.679 --> 00:15:53.080
<v Speaker 2>Building a smarter, faster logistics system.

330
00:15:53.320 --> 00:15:56.600
<v Speaker 1>Lots of pilots happening. Yeah, but that key question seems

331
00:15:56.639 --> 00:16:00.039
<v Speaker 1>to be tag individual items or just the cases.

332
00:15:59.759 --> 00:16:03.320
<v Speaker 2>In that's a critical decision. Yeah, and cost is usually

333
00:16:03.360 --> 00:16:06.159
<v Speaker 2>the deciding factor for item level tagging. It's just hard

334
00:16:06.200 --> 00:16:08.519
<v Speaker 2>to justify putting even a five or ten cent tag

335
00:16:08.559 --> 00:16:10.279
<v Speaker 2>on something that only sells for a dollar or two.

336
00:16:10.519 --> 00:16:13.919
<v Speaker 1>So high value items might get tagged individually.

337
00:16:13.639 --> 00:16:16.720
<v Speaker 2>Right, But for most things a hybrid approach is likely

338
00:16:16.759 --> 00:16:20.200
<v Speaker 2>for the foreseeable future. You know, bar codes on the

339
00:16:20.240 --> 00:16:23.360
<v Speaker 2>individual items, RFID on the cases and pallets. They come

340
00:16:23.360 --> 00:16:27.240
<v Speaker 2>in that dream of having perfect real time item level

341
00:16:27.240 --> 00:16:31.279
<v Speaker 2>inventory everywhere. Still mostly a dream for now, but.

342
00:16:31.240 --> 00:16:34.840
<v Speaker 1>The applications go way beyond just tracking boxes moving forward

343
00:16:34.840 --> 00:16:35.639
<v Speaker 1>in the supply chain.

344
00:16:35.759 --> 00:16:39.639
<v Speaker 2>Right, oh yeah, The report lists a whole bunch managing

345
00:16:39.679 --> 00:16:44.039
<v Speaker 2>returns so reverse logistics, improving quality control by having a

346
00:16:44.080 --> 00:16:47.600
<v Speaker 2>better history of the product, getting more accurate inventory accounts

347
00:16:47.639 --> 00:16:52.159
<v Speaker 2>for accounting, streamlining warranty claims, optimizing how you use assets

348
00:16:52.200 --> 00:16:56.519
<v Speaker 2>like vehicles or emergency gear, helping with product recalls, even

349
00:16:56.600 --> 00:16:57.919
<v Speaker 2>managing recycling debtor.

350
00:16:58.120 --> 00:17:01.679
<v Speaker 1>Healthcare seems like a really strong fit. Tracking is so important.

351
00:17:01.320 --> 00:17:05.200
<v Speaker 2>There, absolutely tagging patients for safety checks, tracking expensive medical

352
00:17:05.200 --> 00:17:07.920
<v Speaker 2>equipment so it doesn't get lost yep. Tagging pharmaceuticals to

353
00:17:07.920 --> 00:17:11.759
<v Speaker 2>fight counterfeiting, managery calls, check expiration dates, maybe even automatically

354
00:17:11.839 --> 00:17:14.200
<v Speaker 2>check for drug interactions when medicine is administered.

355
00:17:14.279 --> 00:17:18.319
<v Speaker 1>And security uses access control theft prevention yep.

356
00:17:18.440 --> 00:17:21.319
<v Speaker 2>Those two. Though the report is careful to note that

357
00:17:21.480 --> 00:17:24.839
<v Speaker 2>just slapping a basic RFID tag on something isn't high

358
00:17:24.839 --> 00:17:28.640
<v Speaker 2>security by itself. You'd need more sophisticated tags and systems

359
00:17:28.680 --> 00:17:32.720
<v Speaker 2>to prevent determined attackers from spoofing or bypassing them. The

360
00:17:32.839 --> 00:17:36.400
<v Speaker 2>key takeaway is really matching the specific business need to

361
00:17:36.480 --> 00:17:38.519
<v Speaker 2>the right type of RFID.

362
00:17:38.880 --> 00:17:42.279
<v Speaker 1>What about using RFID right where consumers are like in

363
00:17:42.319 --> 00:17:45.480
<v Speaker 1>retail stores. The report seemed a bit more cautious about

364
00:17:45.480 --> 00:17:46.519
<v Speaker 1>the business case there.

365
00:17:46.920 --> 00:17:49.400
<v Speaker 2>Yeah, it suggests it's been a tougher cell compared to

366
00:17:49.440 --> 00:17:52.599
<v Speaker 2>the back end supply chain stuff. Achieving those big efficiency

367
00:17:52.599 --> 00:17:55.720
<v Speaker 2>gains is harder and often more expensive in the more

368
00:17:55.839 --> 00:17:57.480
<v Speaker 2>chaotic retail environment.

369
00:17:57.160 --> 00:17:59.240
<v Speaker 1>And ideas like automated checkout.

370
00:17:58.880 --> 00:18:01.920
<v Speaker 2>Super appealing, right, but the cost of tagging every single

371
00:18:01.920 --> 00:18:05.279
<v Speaker 2>item is still mostly prohibitive for general merchandise. Plus it

372
00:18:05.319 --> 00:18:08.519
<v Speaker 2>immediately brings up major privacy and security questions for shoppers.

373
00:18:08.519 --> 00:18:11.720
<v Speaker 1>There's that interesting story about the upscale clothing store experiment.

374
00:18:12.200 --> 00:18:15.759
<v Speaker 2>Right, They tagged fancy clothes so staff could get info

375
00:18:15.920 --> 00:18:19.839
<v Speaker 2>quickly at Kiosks. The staff liked having the info, but

376
00:18:19.880 --> 00:18:23.160
<v Speaker 2>found the system itself a bit clunky or distracting sometimes,

377
00:18:23.799 --> 00:18:26.799
<v Speaker 2>and critically, the store hadn't really built the back end

378
00:18:26.839 --> 00:18:29.240
<v Speaker 2>systems needed to use all the data the tags we're

379
00:18:29.279 --> 00:18:32.720
<v Speaker 2>generating for things like optimizing inventory and real time.

380
00:18:33.279 --> 00:18:36.039
<v Speaker 1>That really points to the risk of information overload. The

381
00:18:36.079 --> 00:18:37.880
<v Speaker 1>report mentions, it really does.

382
00:18:38.079 --> 00:18:41.279
<v Speaker 2>Just collecting tons of data isn't the goal. You need

383
00:18:41.319 --> 00:18:44.039
<v Speaker 2>a clear purpose for it and the systems to process

384
00:18:44.039 --> 00:18:46.680
<v Speaker 2>it and turn it into useful action. You know, maybe

385
00:18:46.720 --> 00:18:48.480
<v Speaker 2>knowing someone picked up a sweater and put it back

386
00:18:48.519 --> 00:18:51.359
<v Speaker 2>is interesting, but what's the actual business value and how

387
00:18:51.359 --> 00:18:54.480
<v Speaker 2>do you implement a system to leverage that effectively? It's

388
00:18:54.519 --> 00:18:55.640
<v Speaker 2>not always obvious yet.

389
00:18:56.000 --> 00:18:58.559
<v Speaker 1>The report then just lists a whole stream of other

390
00:18:58.599 --> 00:19:01.759
<v Speaker 1>places RFID is being user look that it's quite diverse.

391
00:19:01.960 --> 00:19:04.799
<v Speaker 2>It really is. Things like automated toll roads. Easy pass

392
00:19:04.839 --> 00:19:10.440
<v Speaker 2>is a classic example, contactless payment cards, pet identification chips, passports,

393
00:19:10.440 --> 00:19:13.920
<v Speaker 2>though that one sparked huge privacy debates, leading to requirements

394
00:19:13.920 --> 00:19:15.400
<v Speaker 2>for shielding and encryption.

395
00:19:15.200 --> 00:19:18.519
<v Speaker 1>Employee badges for access, even implants.

396
00:19:18.200 --> 00:19:21.720
<v Speaker 2>Yeah, voluntary implants for access to secure areas, or even

397
00:19:21.720 --> 00:19:24.039
<v Speaker 2>for convenience in places like nightclubs. Believe it or Not

398
00:19:24.119 --> 00:19:27.839
<v Speaker 2>shows the range also tracking medicines down to the single pill,

399
00:19:28.200 --> 00:19:32.400
<v Speaker 2>tracking systems in schools, managing reservations or finding family members

400
00:19:32.400 --> 00:19:34.559
<v Speaker 2>in theme parks, timing runners and.

401
00:19:34.599 --> 00:19:37.119
<v Speaker 1>Marathons, museum artifact tracking.

402
00:19:37.000 --> 00:19:41.400
<v Speaker 2>Right managing collections, maybe even enhancing visitor experiences. And it's

403
00:19:41.440 --> 00:19:45.119
<v Speaker 2>a tool in ubiquitous computing research, tracking objects and people

404
00:19:45.240 --> 00:19:48.519
<v Speaker 2>to create smarter environments, maybe for elder care monitoring or

405
00:19:48.559 --> 00:19:49.240
<v Speaker 2>just reminders.

406
00:19:49.359 --> 00:19:52.079
<v Speaker 1>Luggage tracking is mentioned, but sounds like it's still tough.

407
00:19:52.000 --> 00:19:56.079
<v Speaker 2>Still challenging. Yeah, tag orientation reading reliably in that kind

408
00:19:56.119 --> 00:19:59.839
<v Speaker 2>of messy airport environment, and clothing tagging comes up again,

409
00:20:00.160 --> 00:20:04.200
<v Speaker 2>remembering the backlash Beneton faced over privacy fears, but also

410
00:20:04.279 --> 00:20:08.160
<v Speaker 2>the potential upsides, easier returns, no need for receipts. Maybe

411
00:20:08.200 --> 00:20:11.440
<v Speaker 2>your washing machine reads the tag. But again, those privacy

412
00:20:11.440 --> 00:20:13.039
<v Speaker 2>concerns are right there, and that.

413
00:20:12.960 --> 00:20:17.960
<v Speaker 1>Pivots us perfectly into the really big picture society and culture.

414
00:20:18.279 --> 00:20:20.319
<v Speaker 1>This is where the tech really touches all of us,

415
00:20:20.319 --> 00:20:22.440
<v Speaker 1>and the report gets into some serious questions.

416
00:20:22.640 --> 00:20:26.079
<v Speaker 2>Yeah, this vision of an Internet of things where computers

417
00:20:26.079 --> 00:20:31.839
<v Speaker 2>can instantly identify anything anywhere, it has this huge promise right, efficiency, safety, convenience,

418
00:20:32.640 --> 00:20:35.720
<v Speaker 2>But as the report really hammers home, it also comes

419
00:20:35.720 --> 00:20:39.680
<v Speaker 2>with significant risks misuse, unintended consequences.

420
00:20:39.759 --> 00:20:43.279
<v Speaker 1>And this isn't just academic worry. The report notes real

421
00:20:43.319 --> 00:20:44.160
<v Speaker 1>pushback happen.

422
00:20:44.240 --> 00:20:48.400
<v Speaker 2>Oh yeah, it mentions over forty consumer privacy and civil

423
00:20:48.440 --> 00:20:51.920
<v Speaker 2>liberties groups actually called for a moratorium on using our

424
00:20:52.079 --> 00:20:56.680
<v Speaker 2>FID in consumer products. That's significant. It was enough pressure

425
00:20:56.720 --> 00:20:59.640
<v Speaker 2>to make companies like Benetton and even Walmart rethink or

426
00:20:59.680 --> 00:21:03.039
<v Speaker 2>modify their initial plans. There were even strikes and boycotts mentioned.

427
00:21:03.240 --> 00:21:06.319
<v Speaker 1>What were the main fears? The report listed five potential threats.

428
00:21:06.480 --> 00:21:10.000
<v Speaker 2>Yeah, The core concerns raised were one tags being hidden

429
00:21:10.160 --> 00:21:13.240
<v Speaker 2>placed inside products where you can't easily see or remove them.

430
00:21:13.640 --> 00:21:17.079
<v Speaker 2>Two every single object potentially having a unique idea that

431
00:21:17.119 --> 00:21:20.119
<v Speaker 2>could be tracked over its lifetime. Three the massive aggregation

432
00:21:20.200 --> 00:21:23.640
<v Speaker 2>of data from all these tags, potentially creating detailed.

433
00:21:23.200 --> 00:21:24.599
<v Speaker 1>Profiles ok three.

434
00:21:24.640 --> 00:21:27.839
<v Speaker 2>Four the possibility of hidden readers being placed in public

435
00:21:27.960 --> 00:21:31.799
<v Speaker 2>or private spaces scanning tags without people's knowledge or consent.

436
00:21:32.519 --> 00:21:35.680
<v Speaker 2>And five the end result of all that, the potential

437
00:21:35.720 --> 00:21:39.079
<v Speaker 2>for pervasive tracking and profiling of individuals based on the

438
00:21:39.119 --> 00:21:40.599
<v Speaker 2>items they possess or carry.

439
00:21:41.000 --> 00:21:43.599
<v Speaker 1>It seems like the very nature of the tech, the

440
00:21:43.680 --> 00:21:47.960
<v Speaker 1>tags being small, the readers wireless, enables these issues.

441
00:21:48.119 --> 00:21:50.599
<v Speaker 2>That's a key insight from the report. Yes, tags can

442
00:21:50.640 --> 00:21:54.319
<v Speaker 2>be tiny, nearly invisible, they move around with people and objects.

443
00:21:54.599 --> 00:22:00.000
<v Speaker 2>Readers operate wirelessly passively listening. Sometimes it creates this asymmetry

444
00:22:00.000 --> 00:22:04.839
<v Speaker 2>tree powerful readers interrogating these widespread, often simple tags, potentially

445
00:22:04.880 --> 00:22:06.519
<v Speaker 2>without the tagholder even knowing.

446
00:22:06.359 --> 00:22:08.599
<v Speaker 1>And the risks grow as the tags get smarter.

447
00:22:08.759 --> 00:22:12.319
<v Speaker 2>Absolutely, the report traces this progression. A simple read only

448
00:22:12.400 --> 00:22:15.839
<v Speaker 2>ID tag that raises certain privacy questions, but a tag

449
00:22:15.880 --> 00:22:18.359
<v Speaker 2>you can write new data onto that raises questions about

450
00:22:18.359 --> 00:22:19.720
<v Speaker 2>who controls the information on.

451
00:22:19.680 --> 00:22:21.720
<v Speaker 1>Your belongings and active tags.

452
00:22:21.680 --> 00:22:26.680
<v Speaker 2>Active tags with sensors, maybe networking capabilities that opens the

453
00:22:26.720 --> 00:22:31.720
<v Speaker 2>door to persistent tracking collecting data. Over time, the questions

454
00:22:31.880 --> 00:22:35.519
<v Speaker 2>shift from what is this object? To what has this

455
00:22:35.599 --> 00:22:39.039
<v Speaker 2>object experienced? Or even can someone tell this object what

456
00:22:39.079 --> 00:22:39.359
<v Speaker 2>to do?

457
00:22:39.640 --> 00:22:43.640
<v Speaker 1>There's also this concern about path determinism mentioned what's that about?

458
00:22:43.960 --> 00:22:46.720
<v Speaker 2>Yeah, that's the worry that early design choices may be

459
00:22:46.799 --> 00:22:50.680
<v Speaker 2>made to optimize for cost, in say, supply chains could

460
00:22:50.720 --> 00:22:53.359
<v Speaker 2>accidentally lock us into a path where it becomes really

461
00:22:53.359 --> 00:22:57.079
<v Speaker 2>hard to add strong privacy protections later on, especially as

462
00:22:57.079 --> 00:22:59.960
<v Speaker 2>the text spreads into more personal uses, like you can't

463
00:23:00.200 --> 00:23:02.720
<v Speaker 2>easily bolt on privacy after the foundation is built the

464
00:23:02.759 --> 00:23:03.599
<v Speaker 2>wrong way and.

465
00:23:03.559 --> 00:23:06.440
<v Speaker 1>Where the tags get read makes a huge difference, socially

466
00:23:06.640 --> 00:23:07.559
<v Speaker 1>massive difference.

467
00:23:07.880 --> 00:23:12.039
<v Speaker 2>Reading tags inside a closed warehouse relatively few societal issues.

468
00:23:12.079 --> 00:23:14.920
<v Speaker 2>Reading them inside a retail store starts raising questions about

469
00:23:14.920 --> 00:23:18.039
<v Speaker 2>tracking shopper behavior. Reading tags on items you've bought and

470
00:23:18.079 --> 00:23:20.559
<v Speaker 2>are carrying down the street. That's where the biggest privacy

471
00:23:20.559 --> 00:23:21.400
<v Speaker 2>alarms go off.

472
00:23:21.519 --> 00:23:24.240
<v Speaker 1>The report really stresses the linking of tag data to

473
00:23:24.279 --> 00:23:25.240
<v Speaker 1>personal identity.

474
00:23:25.559 --> 00:23:27.880
<v Speaker 2>Right even if the tag itself doesn't have your name

475
00:23:27.960 --> 00:23:30.480
<v Speaker 2>encoded on it, the data from that tag where it

476
00:23:30.559 --> 00:23:33.680
<v Speaker 2>was read, when maybe combined with other data like security

477
00:23:33.680 --> 00:23:36.960
<v Speaker 2>camera footage or payment records, can potentially be cross referenced.

478
00:23:37.240 --> 00:23:40.240
<v Speaker 2>It could be used to identify you or infer sensitive

479
00:23:40.279 --> 00:23:42.799
<v Speaker 2>things about your habits, your health, your associations.

480
00:23:43.319 --> 00:23:46.319
<v Speaker 1>That example with the RFID tags and car tires was

481
00:23:46.359 --> 00:23:46.960
<v Speaker 1>pretty stark.

482
00:23:47.200 --> 00:23:49.880
<v Speaker 2>It really illustrates the point. Even if the tire tag

483
00:23:49.920 --> 00:23:52.160
<v Speaker 2>isn't directly linked to you when you buy it, a

484
00:23:52.200 --> 00:23:56.079
<v Speaker 2>network of readers could potentially track vehicles by their unique

485
00:23:56.119 --> 00:23:59.440
<v Speaker 2>tire IDs. Maybe notice a car repeatedly park near a

486
00:23:59.440 --> 00:24:02.279
<v Speaker 2>specific place polytical HQ or a certain clinic or a

487
00:24:02.319 --> 00:24:06.039
<v Speaker 2>religious center. It highlights that big concern about unintended or

488
00:24:06.079 --> 00:24:09.400
<v Speaker 2>unauthorized uses tracking happening without people even being aware.

489
00:24:09.480 --> 00:24:11.960
<v Speaker 1>It's possible, and it's not just privacy, is it. The

490
00:24:12.039 --> 00:24:14.359
<v Speaker 1>report lists other concerns too, Yeah, a.

491
00:24:14.359 --> 00:24:18.000
<v Speaker 2>Whole cluster of related societal values trust. Can we trust

492
00:24:18.079 --> 00:24:21.039
<v Speaker 2>the systems and the companies using them. Safety, are there

493
00:24:21.039 --> 00:24:25.839
<v Speaker 2>any health risks though generally low for RFID power levels? Security?

494
00:24:25.920 --> 00:24:30.119
<v Speaker 2>How vulnerable are tags and systems to hacking or spoofing? Fairness?

495
00:24:30.160 --> 00:24:33.559
<v Speaker 2>Will the benefits and risks be distributed fairly? Accountability, who's

496
00:24:33.599 --> 00:24:38.480
<v Speaker 2>responsible if data is misused? Accessibility, reliability, and that fundamental

497
00:24:38.480 --> 00:24:40.079
<v Speaker 2>idea of informed consent.

498
00:24:40.359 --> 00:24:43.839
<v Speaker 1>It sets up that kind of utopian versus dystopian tension

499
00:24:43.880 --> 00:24:44.839
<v Speaker 1>around the technology.

500
00:24:44.960 --> 00:24:48.079
<v Speaker 2>It really does, and the report strongly argues that addressing

501
00:24:48.119 --> 00:24:51.480
<v Speaker 2>these concerns before the technology becomes totally widespread and locked

502
00:24:51.519 --> 00:24:54.200
<v Speaker 2>in is crucial for building public trust and avoiding a

503
00:24:54.200 --> 00:24:57.960
<v Speaker 2>major backlash that could stop even beneficial uses. Designing with

504
00:24:58.039 --> 00:24:59.319
<v Speaker 2>social norms in mind.

505
00:24:59.119 --> 00:25:02.519
<v Speaker 1>From the start, but policy always seems to lag behind technology.

506
00:25:02.599 --> 00:25:05.559
<v Speaker 2>That's the perennial challenge, and you have function creep tech

507
00:25:05.640 --> 00:25:09.519
<v Speaker 2>design for supply chains suddenly gets eyed for surveillance uses

508
00:25:09.559 --> 00:25:13.200
<v Speaker 2>the original designers never envisioned. So the report suggests we

509
00:25:13.240 --> 00:25:16.559
<v Speaker 2>basically have two main levers, regulation setting rules of the

510
00:25:16.640 --> 00:25:19.119
<v Speaker 2>road and the design of the technology itself trying to

511
00:25:19.119 --> 00:25:20.079
<v Speaker 2>build in safeguards.

512
00:25:20.440 --> 00:25:24.039
<v Speaker 1>The report also notes that privacy itself is slippery, hard

513
00:25:24.039 --> 00:25:25.400
<v Speaker 1>to define very true.

514
00:25:25.519 --> 00:25:28.519
<v Speaker 2>It's not one single thing. It depends heavily on context.

515
00:25:28.839 --> 00:25:32.079
<v Speaker 2>What's private at home isn't private in public. You often

516
00:25:32.119 --> 00:25:35.319
<v Speaker 2>don't know when it's been violated. The feedback loop is weak,

517
00:25:35.880 --> 00:25:38.200
<v Speaker 2>and it's not just about hiding data. It's also tied

518
00:25:38.240 --> 00:25:40.519
<v Speaker 2>up with autonomy, dignity, and trust.

519
00:25:41.119 --> 00:25:44.559
<v Speaker 1>Did the report mention any technical fixes features built into

520
00:25:44.559 --> 00:25:45.599
<v Speaker 1>standards like Gen two?

521
00:25:45.720 --> 00:25:48.599
<v Speaker 2>Yes it does. Gen two, for example, includes a basic

522
00:25:48.680 --> 00:25:52.880
<v Speaker 2>kill command, is password protected, meant to permanently disable the tag,

523
00:25:53.079 --> 00:25:55.559
<v Speaker 2>and it can even send back a confirmation that it's dead.

524
00:25:55.839 --> 00:25:58.640
<v Speaker 2>There's also access control for writing data to tags.

525
00:25:58.839 --> 00:25:59.720
<v Speaker 1>But isn't that enough?

526
00:25:59.799 --> 00:26:03.279
<v Speaker 2>Well, that's the debate The report highlights. Are these features sufficient,

527
00:26:03.599 --> 00:26:06.240
<v Speaker 2>How easy are they for a regular person to actually use?

528
00:26:06.559 --> 00:26:08.640
<v Speaker 2>How do you verify the tag is really dead? Do

529
00:26:08.680 --> 00:26:11.400
<v Speaker 2>you need home readers or kiosks in stores? And how

530
00:26:11.440 --> 00:26:15.359
<v Speaker 2>secure are those passwords and access controls. Other ideas mentioned

531
00:26:15.359 --> 00:26:18.680
<v Speaker 2>include things like blocker tags, personal devices that can maybe

532
00:26:18.799 --> 00:26:21.960
<v Speaker 2>jam or shield RFID signals in your immediate vicinity.

533
00:26:22.039 --> 00:26:25.240
<v Speaker 1>The court principle that came up was notice and awareness.

534
00:26:26.079 --> 00:26:28.039
<v Speaker 1>Transparency absolutely key.

535
00:26:28.119 --> 00:26:31.559
<v Speaker 2>According to privacy advocates cited in the report, people should

536
00:26:31.559 --> 00:26:34.559
<v Speaker 2>know when tags are present, when they're being read, what

537
00:26:34.680 --> 00:26:38.319
<v Speaker 2>data is collected, how it's used, things like clear labels

538
00:26:38.319 --> 00:26:42.359
<v Speaker 2>on tagged products, no secret reading, no hidden databases. The

539
00:26:42.440 --> 00:26:45.839
<v Speaker 2>report connects this to established ideas like the Fair Information

540
00:26:45.920 --> 00:26:50.160
<v Speaker 2>Practice Principles or fit pps fIF pps like what principles

541
00:26:50.279 --> 00:26:55.400
<v Speaker 2>like purpose specification, be clear why you're collecting the data, collection, limitation,

542
00:26:55.559 --> 00:26:59.240
<v Speaker 2>only collect what's necessary. Accountability, someone has to be responsible.

543
00:26:59.440 --> 00:27:02.039
<v Speaker 2>Securities safeguards protect the data, things like that.

544
00:27:02.240 --> 00:27:05.400
<v Speaker 1>What about consumer choice? Can't people just opt out?

545
00:27:05.759 --> 00:27:08.920
<v Speaker 2>It's complicated. The report points out that when RFID is

546
00:27:08.960 --> 00:27:12.200
<v Speaker 2>tied to convenience or benefits like easy pass saving you time,

547
00:27:12.680 --> 00:27:16.000
<v Speaker 2>or loyalty cards giving discounts, it becomes harder not to participate.

548
00:27:16.480 --> 00:27:18.839
<v Speaker 2>You might opt in for one reason without realizing you're

549
00:27:18.839 --> 00:27:21.480
<v Speaker 2>also enabling other kinds of tracking or profiling down the

550
00:27:21.519 --> 00:27:25.720
<v Speaker 2>line that infrastructure built for efficiency could potentially be repurposed.

551
00:27:25.880 --> 00:27:28.359
<v Speaker 1>So it really comes back to building public trust that

552
00:27:28.519 --> 00:27:30.319
<v Speaker 1>seems paramount absolutely.

553
00:27:30.480 --> 00:27:34.680
<v Speaker 2>The report concludes that trust requires real safeguards like those

554
00:27:34.759 --> 00:27:38.839
<v Speaker 2>FIT based principles, but also meaningful ways for different groups,

555
00:27:38.880 --> 00:27:43.240
<v Speaker 2>including consumer advocates, to participate in shaping how the technology

556
00:27:43.319 --> 00:27:44.759
<v Speaker 2>is developed. And used that.

557
00:27:44.799 --> 00:27:47.960
<v Speaker 1>Idea of killing tags at the point of sale. It

558
00:27:48.039 --> 00:27:51.400
<v Speaker 1>seems like a straightforward fix for many consumer privacy warriors.

559
00:27:51.519 --> 00:27:54.559
<v Speaker 2>It does address a lot of the post purchase tracking concerns. Yes,

560
00:27:55.119 --> 00:27:58.319
<v Speaker 2>but the report cleverly raises the counter argument, what if

561
00:27:58.319 --> 00:28:01.279
<v Speaker 2>the consumer wants the tag to stay at maybe for

562
00:28:01.359 --> 00:28:05.799
<v Speaker 2>tracking food expiration, getting recall ERT's automatically easier recycling info.

563
00:28:06.119 --> 00:28:09.759
<v Speaker 1>Ah, so killing it removes potentially useful features too exactly.

564
00:28:09.960 --> 00:28:12.279
<v Speaker 2>It shows that the right answer depends on the specific

565
00:28:12.319 --> 00:28:15.960
<v Speaker 2>application and the user's own values. In context, a universal

566
00:28:16.039 --> 00:28:19.000
<v Speaker 2>kill switch might not be ideal for everyone or every situation.

567
00:28:19.160 --> 00:28:20.759
<v Speaker 1>Any other concerns mentioned.

568
00:28:20.480 --> 00:28:23.759
<v Speaker 2>Briefly, Yeah, the idea of publicity, how does this tech

569
00:28:23.839 --> 00:28:26.680
<v Speaker 2>contribute to or detract from the shared public good? And

570
00:28:26.839 --> 00:28:31.000
<v Speaker 2>environmental sustainability. What happens when we have billions, maybe trillions,

571
00:28:31.440 --> 00:28:34.400
<v Speaker 2>of these tiny tags needing disposal or recycling.

572
00:28:34.720 --> 00:28:38.920
<v Speaker 1>It sounds that trust might become almost a negotiation between

573
00:28:38.960 --> 00:28:41.839
<v Speaker 1>the company using the tech and the person being tagged.

574
00:28:42.000 --> 00:28:45.680
<v Speaker 2>The report suggests something like that, Yeah, a provider's reputation

575
00:28:45.759 --> 00:28:49.400
<v Speaker 2>for handling data responsibly could become really important, And it

576
00:28:49.400 --> 00:28:53.599
<v Speaker 2>acknowledges that people's attitudes vary, citing those nightclub implants as

577
00:28:53.640 --> 00:28:56.160
<v Speaker 2>an example where some people will choose to be tagged

578
00:28:56.160 --> 00:28:58.559
<v Speaker 2>in specific controlled environments.

579
00:28:58.039 --> 00:29:00.039
<v Speaker 1>But ignoring the broader concerns.

580
00:28:59.599 --> 00:29:03.160
<v Speaker 2>Is risk, very risky. It could lead to confrontation, regulation

581
00:29:03.559 --> 00:29:07.759
<v Speaker 2>and ultimately prevent even the really positive constructive uses of

582
00:29:07.759 --> 00:29:09.200
<v Speaker 2>the technology from taking root.

583
00:29:09.359 --> 00:29:13.880
<v Speaker 1>And focusing only on individual privacy might miss collective benefits.

584
00:29:13.920 --> 00:29:15.960
<v Speaker 2>That's the other side of the coin. Yeah, there might

585
00:29:16.000 --> 00:29:19.200
<v Speaker 2>be societal gains in safety or efficiency that requires some

586
00:29:19.440 --> 00:29:23.240
<v Speaker 2>level of data sharing, and focusing only unlocking everything down

587
00:29:23.640 --> 00:29:26.279
<v Speaker 2>might prevent those. It's a really tough balancing act.

588
00:29:26.359 --> 00:29:28.880
<v Speaker 1>Okay, So wrapping up this deep dive, then, yeah, it's

589
00:29:28.960 --> 00:29:32.839
<v Speaker 1>clear RFID is well. It's a fascinating technology, lots of

590
00:29:32.839 --> 00:29:36.799
<v Speaker 1>different types, from simple passive ones to complex active sensors, and.

591
00:29:36.799 --> 00:29:40.160
<v Speaker 2>It's already moving fast, especially in logistics and supply chains,

592
00:29:40.680 --> 00:29:44.119
<v Speaker 2>but being explored just about everywhere else too, from hospitals

593
00:29:44.160 --> 00:29:45.079
<v Speaker 2>to museums.

594
00:29:45.119 --> 00:29:47.440
<v Speaker 1>But all that potential comes hand in hand with these

595
00:29:47.559 --> 00:29:53.960
<v Speaker 1>really deep social and cultural questions privacy, transparency, trust, They're central.

596
00:29:54.119 --> 00:29:56.759
<v Speaker 2>Yeah, this isn't just a technical discussion. It forces us

597
00:29:56.799 --> 00:30:00.799
<v Speaker 2>to think about design, standards, deployment, How do we balance

598
00:30:00.839 --> 00:30:04.559
<v Speaker 2>the power of knowing where everything is with fundamental human values.

599
00:30:04.640 --> 00:30:10.160
<v Speaker 1>It's a technology that demands careful font balancing efficiency with control, transparency,

600
00:30:10.480 --> 00:30:11.960
<v Speaker 1>and that crucial public trust.

601
00:30:12.119 --> 00:30:14.599
<v Speaker 2>And this National Academy's report makes it really clear these

602
00:30:14.599 --> 00:30:17.279
<v Speaker 2>are complex conversations we need to be having now as

603
00:30:17.359 --> 00:30:18.319
<v Speaker 2>the tech evolves.

604
00:30:18.559 --> 00:30:21.200
<v Speaker 1>Absolutely so here's maybe the final thought for you, the listener.

605
00:30:21.559 --> 00:30:24.680
<v Speaker 1>As RFID potentially gets woven deeper into the world around us,

606
00:30:24.680 --> 00:30:27.519
<v Speaker 1>into everyday objects, how do we make sure the benefits

607
00:30:27.519 --> 00:30:31.079
<v Speaker 1>of knowing what everything is everywhere actually serve everyone. How

608
00:30:31.079 --> 00:30:33.559
<v Speaker 1>do we avoid building systems they could track and profile

609
00:30:33.680 --> 00:30:36.640
<v Speaker 1>people without their real consent or knowledge. It's a huge

610
00:30:36.720 --> 00:30:41.599
<v Speaker 1>challenge needs engineers, businesses, policymakers, advocates, really all of us.

611
00:30:41.799 --> 00:30:44.319
<v Speaker 1>What part of this whole picture stood out most to you.