WEBVTT

1
00:00:01.199 --> 00:00:06.200
<v Speaker 1>Welcome to the Sentient Code, where intelligence is engineered, autonomy

2
00:00:06.280 --> 00:00:10.439
<v Speaker 1>is emerging, and a line between human and machine grows thinner.

3
00:00:10.800 --> 00:00:15.359
<v Speaker 1>Each episode, we decode the algorithms, explore the robotics, and

4
00:00:15.439 --> 00:00:19.000
<v Speaker 1>examine the ideas shaping the future of artificial minds.

5
00:00:23.839 --> 00:00:27.280
<v Speaker 2>For all of human history, humanity has treated matter like

6
00:00:27.280 --> 00:00:30.120
<v Speaker 2>a sculptor treats clay. You know, you find a material

7
00:00:30.120 --> 00:00:32.479
<v Speaker 2>in the earth, you extract it, and you just force

8
00:00:32.560 --> 00:00:35.600
<v Speaker 2>it into the shapes you want entirely from the outside.

9
00:00:35.799 --> 00:00:38.159
<v Speaker 3>It's a very top down approach exactly.

10
00:00:38.320 --> 00:00:41.159
<v Speaker 2>I mean, you chisel stone, you heat and hammer iron,

11
00:00:41.560 --> 00:00:45.399
<v Speaker 2>you melt and mold plastic. The physical material itself is

12
00:00:45.520 --> 00:00:47.200
<v Speaker 2>essentially dumb.

13
00:00:47.560 --> 00:00:48.280
<v Speaker 3>It's passive.

14
00:00:48.359 --> 00:00:51.119
<v Speaker 2>It just sits there, an inert substance, waiting for you

15
00:00:51.200 --> 00:00:54.439
<v Speaker 2>to impose your will onto it. But right now there

16
00:00:54.479 --> 00:00:58.640
<v Speaker 2>is this profound mechanical shift happening in material science. We're

17
00:00:58.679 --> 00:01:01.640
<v Speaker 2>no longer just obs deserving or brute forcing matter.

18
00:01:01.799 --> 00:01:02.560
<v Speaker 3>No, we're really not.

19
00:01:02.799 --> 00:01:05.040
<v Speaker 2>We're actually designing it from the ground up to behave

20
00:01:05.079 --> 00:01:09.640
<v Speaker 2>with absolute intentional precision at the bare molecular level. So

21
00:01:09.680 --> 00:01:12.159
<v Speaker 2>our mission for this deep dive today is to figure

22
00:01:12.200 --> 00:01:15.640
<v Speaker 2>out exactly how scientists are turning passive physical matter into

23
00:01:15.719 --> 00:01:17.719
<v Speaker 2>programmable software, moving from a.

24
00:01:17.680 --> 00:01:19.519
<v Speaker 3>World where we build things to a world where things

25
00:01:19.519 --> 00:01:21.079
<v Speaker 3>build themselves exactly.

26
00:01:21.319 --> 00:01:23.319
<v Speaker 2>And I think to really grasp the physics of this,

27
00:01:23.640 --> 00:01:26.239
<v Speaker 2>we need to contrast it with how traditional chemistry is

28
00:01:26.280 --> 00:01:30.280
<v Speaker 2>operated for centuries. Because traditional chemistry, I mean, even at

29
00:01:30.280 --> 00:01:35.879
<v Speaker 2>its most advanced, like synthesizing complex pharmaceuticals, it's still essentially

30
00:01:36.439 --> 00:01:38.680
<v Speaker 2>a game of macroscopic manipulation.

31
00:01:39.079 --> 00:01:43.120
<v Speaker 3>It is. It relies heavily on manipulating external conditions to

32
00:01:43.280 --> 00:01:46.560
<v Speaker 3>force a statistical outcome. Right, if you want a reaction

33
00:01:46.680 --> 00:01:50.040
<v Speaker 3>to happen, you raise the temperature to increase the kinetic energy,

34
00:01:50.319 --> 00:01:51.239
<v Speaker 3>or you increase the.

35
00:01:51.200 --> 00:01:53.400
<v Speaker 2>Pressure, or you drop in a catalyst.

36
00:01:53.159 --> 00:01:55.920
<v Speaker 3>Exactly, you add a heavy metal catalyst to lower the

37
00:01:55.959 --> 00:01:59.879
<v Speaker 3>activation energy. You are violently shoving molecules together in a

38
00:02:00.159 --> 00:02:03.079
<v Speaker 3>giant vat and just relying on the brute force of

39
00:02:03.120 --> 00:02:05.359
<v Speaker 3>probability to get the chemical bonds you want.

40
00:02:05.400 --> 00:02:07.280
<v Speaker 2>It's chaotic, highly chaotic.

41
00:02:07.280 --> 00:02:10.159
<v Speaker 3>It's an incredibly indirect way of guiding outcomes.

42
00:02:10.360 --> 00:02:13.120
<v Speaker 2>So traditional chemistry is sort of like trying to herd

43
00:02:13.159 --> 00:02:17.439
<v Speaker 2>a massive flock of sheep by building rigid fences, waving

44
00:02:17.439 --> 00:02:18.360
<v Speaker 2>your arms and shouting.

45
00:02:18.479 --> 00:02:19.520
<v Speaker 3>That's a great analogy.

46
00:02:19.560 --> 00:02:22.599
<v Speaker 2>You're just creating external boundaries and hoping this sheer chaotic

47
00:02:22.719 --> 00:02:26.400
<v Speaker 2>energy pushes the flock generally where you want it to go.

48
00:02:26.800 --> 00:02:30.120
<v Speaker 2>But programmable chemistry, on the other hand, is like breeding

49
00:02:30.159 --> 00:02:34.199
<v Speaker 2>a new type of sheep that possesses an instinctual genetic

50
00:02:34.319 --> 00:02:37.919
<v Speaker 2>drive to always stand exactly three feet apart from its

51
00:02:37.919 --> 00:02:41.319
<v Speaker 2>neighbor and form a perfect geometric circle. Yes, you don't

52
00:02:41.319 --> 00:02:44.319
<v Speaker 2>need the fences anymore because the behavior is baked right

53
00:02:44.360 --> 00:02:45.439
<v Speaker 2>into the unit itself.

54
00:02:45.599 --> 00:02:49.599
<v Speaker 3>That perfectly captures the shift in thermodynamics we're seeing. You're

55
00:02:49.680 --> 00:02:53.680
<v Speaker 3>moving the burden of organization from the outside environment, the

56
00:02:53.719 --> 00:02:56.439
<v Speaker 3>macroscopic fences, like you said, to the inside of the

57
00:02:56.520 --> 00:03:01.039
<v Speaker 3>entity itself, into the chemistry. Right into the chemistry ammable chemistry.

58
00:03:01.080 --> 00:03:05.479
<v Speaker 3>The molecules possess the architectural instructions embedded directly into their

59
00:03:05.479 --> 00:03:09.639
<v Speaker 3>physical geometry. We are no longer forcing systems into desired

60
00:03:09.680 --> 00:03:13.520
<v Speaker 3>outcomes through sheer heat and pressure, or coding behaviors, exactly

61
00:03:13.599 --> 00:03:17.960
<v Speaker 3>encoding behaviors through localized chemical affinities. Simple instructions at the

62
00:03:18.039 --> 00:03:22.960
<v Speaker 3>nanoscale when they combine, produce highly complex, deterministic behaviors at

63
00:03:22.960 --> 00:03:23.759
<v Speaker 3>the macroscale.

64
00:03:23.840 --> 00:03:26.919
<v Speaker 2>Well, I think when you use the word instructions or code,

65
00:03:27.240 --> 00:03:30.199
<v Speaker 2>most people immediately visualize a computer screen. They think of

66
00:03:30.240 --> 00:03:34.039
<v Speaker 2>Python or C plus plus rs or binary ones and

67
00:03:34.159 --> 00:03:37.039
<v Speaker 2>zeros sitting on a silicon hard drive. So we really

68
00:03:37.120 --> 00:03:40.560
<v Speaker 2>have to clarify what code means when we're talking about

69
00:03:40.560 --> 00:03:41.319
<v Speaker 2>physical matter.

70
00:03:41.479 --> 00:03:43.919
<v Speaker 3>Yeah, because the software of this matter is not written

71
00:03:43.919 --> 00:03:47.120
<v Speaker 3>in lines of texts. It's written using physical chemical bonds,

72
00:03:47.159 --> 00:03:51.879
<v Speaker 3>electrostatic forces, and spatial configurations. The specific angles at which

73
00:03:51.879 --> 00:03:55.639
<v Speaker 3>atoms connect, the geometric shapes they form, the placement of

74
00:03:55.680 --> 00:03:58.919
<v Speaker 3>a positive charge here and a negative charge there. That

75
00:03:59.080 --> 00:04:02.240
<v Speaker 3>geometry is the that stereochemistry is the code. It's all

76
00:04:02.280 --> 00:04:06.479
<v Speaker 3>about steric hindrance and Vanderwohl's forces. The physical shape of

77
00:04:06.479 --> 00:04:10.000
<v Speaker 3>the molecule and its chemical affinity act as the instruction set.

78
00:04:10.199 --> 00:04:12.719
<v Speaker 2>Okay, let's anchor that to something physical for the listener.

79
00:04:13.280 --> 00:04:15.080
<v Speaker 2>Look at the coffee cup on your desk right now.

80
00:04:15.400 --> 00:04:18.000
<v Speaker 2>To you, it's a solid, dumb object, just a mug,

81
00:04:18.199 --> 00:04:21.639
<v Speaker 2>just a mug. But at the molecular level, it's this chaotic,

82
00:04:21.800 --> 00:04:25.040
<v Speaker 2>vibrating lattice of atoms. Imagine if you could write a

83
00:04:25.079 --> 00:04:28.600
<v Speaker 2>line of structural code that told those specific atoms to

84
00:04:28.680 --> 00:04:32.279
<v Speaker 2>shift their bonds slightly, altering the thermal conductivity of.

85
00:04:32.240 --> 00:04:35.120
<v Speaker 3>The ceramic, so the cup insulates the coffee better as

86
00:04:35.160 --> 00:04:35.879
<v Speaker 3>it cools down.

87
00:04:36.319 --> 00:04:38.839
<v Speaker 2>Exactly, the geometry dictates the function.

88
00:04:39.079 --> 00:04:42.120
<v Speaker 3>And while the idea of writing code into molecular bonds

89
00:04:42.120 --> 00:04:44.199
<v Speaker 3>to tell matter how to behave sounds like, you know,

90
00:04:44.360 --> 00:04:48.839
<v Speaker 3>speculative science fiction engineering, we actually have a highly successful,

91
00:04:48.879 --> 00:04:51.959
<v Speaker 3>fully functioning prototype of this technology, he do. Yeah, and

92
00:04:52.040 --> 00:04:56.279
<v Speaker 3>it's been running efficiently for about four billion years biology biology. Yeah,

93
00:04:56.639 --> 00:05:01.079
<v Speaker 3>long before material scientists conceived of programmable matter, biology had

94
00:05:01.120 --> 00:05:05.680
<v Speaker 3>already solved the thermodynamic problems of self assembly and molecular coding,

95
00:05:06.040 --> 00:05:07.079
<v Speaker 3>which is wild.

96
00:05:06.800 --> 00:05:10.639
<v Speaker 2>To think about. Yeah, because biological DNA is almost universally

97
00:05:10.680 --> 00:05:12.079
<v Speaker 2>taught as genetic.

98
00:05:11.720 --> 00:05:13.439
<v Speaker 3>Storage, right, like an archive.

99
00:05:13.560 --> 00:05:15.519
<v Speaker 2>Like an archive. Yeah, we think of it as a

100
00:05:15.560 --> 00:05:18.920
<v Speaker 2>passive biological hard drive sitting in the nucleus holding the

101
00:05:18.920 --> 00:05:23.120
<v Speaker 2>blueprints for eye color or height. But DNA is actually

102
00:05:23.439 --> 00:05:27.040
<v Speaker 2>a fully programmable, dynamic physical system.

103
00:05:27.120 --> 00:05:29.639
<v Speaker 3>It doesn't just sit there storing data like a dusty book.

104
00:05:29.800 --> 00:05:33.720
<v Speaker 2>No, it's a physical mechanism that actively directs the assembly

105
00:05:33.800 --> 00:05:37.519
<v Speaker 2>and behavior of living organisms through its structural geometry.

106
00:05:37.720 --> 00:05:40.839
<v Speaker 3>And that's why scientists have stopped looking at biological DNA

107
00:05:41.040 --> 00:05:44.120
<v Speaker 3>merely as a way to understand genetics and heredity. They

108
00:05:44.120 --> 00:05:48.360
<v Speaker 3>are actively hijacking DNA's physical properties to use it as

109
00:05:48.360 --> 00:05:50.439
<v Speaker 3>an engineering building material.

110
00:05:50.319 --> 00:05:51.959
<v Speaker 2>Because it's so predictable, right.

111
00:05:51.920 --> 00:05:55.800
<v Speaker 3>Incredibly predictable. What makes DNA so powerful as a structural

112
00:05:55.839 --> 00:05:59.399
<v Speaker 3>building blog is the Watson Crick pairing rules. The ATC

113
00:05:59.639 --> 00:06:03.639
<v Speaker 3>and G exactly the way the base adenine always binds

114
00:06:03.680 --> 00:06:07.399
<v Speaker 3>with thymine and cytosine always binds with guanine. It creates

115
00:06:07.399 --> 00:06:11.639
<v Speaker 3>an incredibly reliable interlocking foundation. It is a literal physical

116
00:06:11.720 --> 00:06:12.680
<v Speaker 3>language of attraction.

117
00:06:12.920 --> 00:06:16.079
<v Speaker 2>And because of those strict pairing rules, if you synthesize

118
00:06:16.120 --> 00:06:20.800
<v Speaker 2>a specific sequence of ATCNG, you know exactly what complementary

119
00:06:20.800 --> 00:06:23.319
<v Speaker 2>sequence it will aggressively seek out and bind.

120
00:06:23.360 --> 00:06:26.480
<v Speaker 3>Two, you're creating incredibly specific puzzle pieces.

121
00:06:26.639 --> 00:06:29.120
<v Speaker 2>Wait, hold on, let me just I'm struggling to visualize

122
00:06:29.120 --> 00:06:31.800
<v Speaker 2>the actual physical process here. Okay, if you just throw

123
00:06:31.839 --> 00:06:34.800
<v Speaker 2>a billion custom design strands of DNA into a beaker

124
00:06:34.839 --> 00:06:38.079
<v Speaker 2>of water, wouldn't they just constantly crash into each other

125
00:06:38.120 --> 00:06:42.120
<v Speaker 2>and form a massive, useless, tangled knot of genetic spaghetti.

126
00:06:42.199 --> 00:06:44.199
<v Speaker 3>Yeah, if you just dump them in probably.

127
00:06:44.040 --> 00:06:48.160
<v Speaker 2>So how does it actually form a clean, intentional shape.

128
00:06:48.439 --> 00:06:50.399
<v Speaker 3>Well, that gets into the mechanics of what we call

129
00:06:50.600 --> 00:06:52.040
<v Speaker 3>DNA or agami.

130
00:06:51.800 --> 00:06:53.920
<v Speaker 2>Or agami like the paper folding.

131
00:06:53.759 --> 00:06:57.160
<v Speaker 3>Exactly like that, but at the nanoscale. It was pioneered

132
00:06:57.199 --> 00:07:00.160
<v Speaker 3>by researchers like Paul Rothaman. You don't just throw throw

133
00:07:00.199 --> 00:07:02.560
<v Speaker 3>them in at room temperature. You use thermal cycling.

134
00:07:02.600 --> 00:07:03.839
<v Speaker 2>Okay, so temperature control.

135
00:07:03.920 --> 00:07:08.800
<v Speaker 3>Right, You take a very long, single scaffold strand of DNA,

136
00:07:08.959 --> 00:07:12.040
<v Speaker 3>which is often borrowed from a harmless virus actually wow,

137
00:07:12.079 --> 00:07:15.920
<v Speaker 3>and you mix it with hundreds of shorter, custom synthesized

138
00:07:15.959 --> 00:07:20.720
<v Speaker 3>staple strands. Then you heat the mixture up to near boiling.

139
00:07:20.439 --> 00:07:22.800
<v Speaker 2>So you're adding kinetic energy to break everything apart.

140
00:07:23.000 --> 00:07:25.920
<v Speaker 3>Right at that high temperature, the thermal energy is so

141
00:07:26.079 --> 00:07:28.920
<v Speaker 3>high that no bonds can form. The strands are just

142
00:07:29.040 --> 00:07:31.199
<v Speaker 3>wildly vibrating and whipping around.

143
00:07:31.040 --> 00:07:32.680
<v Speaker 2>Just total chaos, total chaos.

144
00:07:33.079 --> 00:07:37.480
<v Speaker 3>But then you slowly meticulously cool the liquid down. This

145
00:07:37.560 --> 00:07:38.800
<v Speaker 3>process is called a kneeling.

146
00:07:38.920 --> 00:07:39.680
<v Speaker 2>Kneeling Okay.

147
00:07:39.959 --> 00:07:42.800
<v Speaker 3>As the temperature drops, the molecules lose kinetic energy and

148
00:07:42.839 --> 00:07:46.240
<v Speaker 3>they start looking for their most thermodynamically stable state.

149
00:07:46.079 --> 00:07:47.240
<v Speaker 2>The lowest energy state.

150
00:07:47.600 --> 00:07:51.439
<v Speaker 3>Exactly. And the short staple strands are programmed with sequences

151
00:07:51.480 --> 00:07:55.199
<v Speaker 3>that perfectly match two distant parts of the long scaffold strand.

152
00:07:55.759 --> 00:07:59.720
<v Speaker 3>As they bind, they physically pinch the long strand, folding it.

153
00:08:00.040 --> 00:08:02.120
<v Speaker 2>Oh. I see. They act like molecular.

154
00:08:01.680 --> 00:08:05.000
<v Speaker 3>Clamps, exactly like clamps. Because point A on the staple

155
00:08:05.079 --> 00:08:07.680
<v Speaker 3>is programmed to bind with point B on the scaffold,

156
00:08:07.959 --> 00:08:11.079
<v Speaker 3>and point C binds with point D, the scaffold strand

157
00:08:11.160 --> 00:08:14.480
<v Speaker 3>naturally bends and folds to make those connections.

158
00:08:13.959 --> 00:08:15.319
<v Speaker 2>And it just forces the shape.

159
00:08:15.439 --> 00:08:18.839
<v Speaker 3>Yeah, and the thermal cooling allows the strands to detach

160
00:08:18.879 --> 00:08:22.279
<v Speaker 3>and reattach if they make a mistake. Constantly seeking the

161
00:08:22.279 --> 00:08:26.040
<v Speaker 3>perfect intended fit. You've essentially programmed the geometry of the

162
00:08:26.079 --> 00:08:29.199
<v Speaker 3>final structure into the chemistry of the initial strands.

163
00:08:29.319 --> 00:08:33.159
<v Speaker 2>And Rothaman literally used this to fold DNA into nanoscale

164
00:08:33.159 --> 00:08:34.200
<v Speaker 2>smiley faces.

165
00:08:33.919 --> 00:08:37.399
<v Speaker 3>Right, yes, smiley faces stars maps of the Americas, just.

166
00:08:37.360 --> 00:08:40.519
<v Speaker 2>To prove that he had absolute geometric control exactly. And

167
00:08:40.600 --> 00:08:44.799
<v Speaker 2>this is happening entirely without microscopic tweezers. You just synthesize

168
00:08:44.799 --> 00:08:47.639
<v Speaker 2>the liquid heat, it cool it, and the molecules follow

169
00:08:47.679 --> 00:08:50.120
<v Speaker 2>their internal thermodynamic code to fold themselves.

170
00:08:50.279 --> 00:08:54.039
<v Speaker 3>The process is meticulously guided by the encoded information within

171
00:08:54.080 --> 00:08:57.600
<v Speaker 3>the molecules themselves. I mean, you are building structures at

172
00:08:57.600 --> 00:09:00.840
<v Speaker 3>a scale nanometers across that would be absolutely impossible to

173
00:09:00.879 --> 00:09:06.600
<v Speaker 3>construct using any conventional top down manufacturing method like photolithography

174
00:09:06.919 --> 00:09:07.679
<v Speaker 3>or milling.

175
00:09:07.879 --> 00:09:11.320
<v Speaker 2>But DNA has limitations. Right, Oh, definitely, biology prove the

176
00:09:11.320 --> 00:09:15.200
<v Speaker 2>mechanism works. But DNA evolved to function inside a very

177
00:09:15.240 --> 00:09:19.480
<v Speaker 2>specific watery, temperature controlled environment, the living cell.

178
00:09:19.759 --> 00:09:19.879
<v Speaker 3>Right.

179
00:09:20.159 --> 00:09:22.279
<v Speaker 2>If we want to build bridges or car parts or

180
00:09:22.320 --> 00:09:26.080
<v Speaker 2>industrial sensors out of programmable matter, we can't build them

181
00:09:26.080 --> 00:09:28.600
<v Speaker 2>out of fragile genetic material that degrades in sunlight.

182
00:09:28.759 --> 00:09:30.919
<v Speaker 3>No, you definitely can't, which is why the field is

183
00:09:31.039 --> 00:09:35.759
<v Speaker 3>rapidly moving beyond biological molecules into purely synthetic materials. We

184
00:09:35.799 --> 00:09:38.919
<v Speaker 3>are taking the theoretical lessons learned from DNA, the idea

185
00:09:38.960 --> 00:09:42.399
<v Speaker 3>of embedding structural code into a polymer chain and applying

186
00:09:42.399 --> 00:09:45.159
<v Speaker 3>them to custom built synthetic plastics and polymers.

187
00:09:45.320 --> 00:09:47.759
<v Speaker 2>So taking the concept but upgrading the hardware.

188
00:09:47.919 --> 00:09:51.159
<v Speaker 3>Essentially, Yes, we can now design synthetic block cop polymers

189
00:09:51.159 --> 00:09:55.159
<v Speaker 3>with highly specific sequences, completely independent of biological constraints.

190
00:09:55.440 --> 00:09:58.480
<v Speaker 2>And these synthetic materials are engineered from the ground up

191
00:09:58.919 --> 00:10:03.399
<v Speaker 2>to fold, interact, and respond to environmental stimuli in strictly

192
00:10:03.399 --> 00:10:06.960
<v Speaker 2>controlled ways. Right, we're talking about polymers that react to

193
00:10:07.120 --> 00:10:11.639
<v Speaker 2>specific wavelengths of light, to precise temperature shifts, or to

194
00:10:11.720 --> 00:10:14.200
<v Speaker 2>the presence of particular chemical gradients, and.

195
00:10:14.159 --> 00:10:19.039
<v Speaker 3>The mechanisms behind this are just fascinating. Take photo responsive polymers,

196
00:10:19.080 --> 00:10:22.120
<v Speaker 3>for instance, you can synthesize a material containing molecules called

197
00:10:22.240 --> 00:10:26.879
<v Speaker 3>azobenzenes azobenzines. Right, when you hit an azobenzene molecule with

198
00:10:27.039 --> 00:10:31.440
<v Speaker 3>ultraviolet light, it absorbs a photon and instantly undergoes a

199
00:10:31.440 --> 00:10:35.799
<v Speaker 3>conformational change. It physically bends. Wow, it isomerizes from a

200
00:10:35.840 --> 00:10:38.559
<v Speaker 3>straight shape to a bent shape. And when you remove

201
00:10:38.600 --> 00:10:41.120
<v Speaker 3>the light or apply a different wavelength, it snaps back.

202
00:10:41.200 --> 00:10:42.840
<v Speaker 2>So the light isn't just heating it up. The light

203
00:10:42.919 --> 00:10:46.080
<v Speaker 2>is triggering an actual mechanical gear shift at the nanoscale.

204
00:10:46.159 --> 00:10:49.559
<v Speaker 3>Yes, and if you string millions of these molecules together

205
00:10:49.720 --> 00:10:53.960
<v Speaker 3>into a polymer chain, that microscopic bending cascades into a

206
00:10:53.960 --> 00:10:58.039
<v Speaker 3>macroscopic change. You have a synthetic material designed to instantly

207
00:10:58.080 --> 00:11:00.879
<v Speaker 3>become more rigid or to contract like an artificial muscle.

208
00:11:00.960 --> 00:11:02.759
<v Speaker 3>The moment it is exposed to u V.

209
00:11:02.840 --> 00:11:04.440
<v Speaker 2>Light that is insane.

210
00:11:04.759 --> 00:11:09.600
<v Speaker 3>Or consider thermore responsive polymers that undergo a phase transition

211
00:11:09.679 --> 00:11:13.399
<v Speaker 3>at a specific temperature. They shift instantly from being hydrophilic

212
00:11:13.440 --> 00:11:17.159
<v Speaker 3>which means water loving, to hydrophobic, repelling water.

213
00:11:17.399 --> 00:11:20.440
<v Speaker 2>And the behavior isn't because someone is outside flipping a

214
00:11:20.480 --> 00:11:23.600
<v Speaker 2>mechanical switch or adding a dye. The physical reaction is

215
00:11:23.639 --> 00:11:27.720
<v Speaker 2>built directly into the material's structural code. Exactly, the material

216
00:11:27.759 --> 00:11:30.440
<v Speaker 2>itself is making a binary choice based on its environment.

217
00:11:30.559 --> 00:11:32.960
<v Speaker 3>And this naturally leads us deeper into the physics of

218
00:11:32.960 --> 00:11:36.200
<v Speaker 3>self assembly, because when you are dealing with building complex

219
00:11:36.279 --> 00:11:39.519
<v Speaker 3>machines at the nanoscale, the physical tools we rely on

220
00:11:39.600 --> 00:11:41.879
<v Speaker 3>in the macro world completely break down.

221
00:11:42.080 --> 00:11:44.240
<v Speaker 2>The physics just don't work the same way, not at all.

222
00:11:44.279 --> 00:11:47.320
<v Speaker 2>You can't use microscopic tweezers to build a complex system

223
00:11:47.440 --> 00:11:50.879
<v Speaker 2>atom by atom. At that scale, everything is subjected to

224
00:11:50.919 --> 00:11:51.679
<v Speaker 2>Brownie in motion.

225
00:11:51.919 --> 00:11:54.159
<v Speaker 3>Yes, the constant shaking right.

226
00:11:54.159 --> 00:11:58.440
<v Speaker 2>Water molecules are constantly bombarding your materials, shaking them violently.

227
00:11:59.120 --> 00:12:03.159
<v Speaker 2>The scale is overwhelmingly small and the environment is just

228
00:12:03.279 --> 00:12:07.399
<v Speaker 2>pure chaos. If you want a complex functional structure at

229
00:12:07.399 --> 00:12:10.559
<v Speaker 2>the nanoscale, you can't build it You have to design

230
00:12:10.600 --> 00:12:13.639
<v Speaker 2>the pieces so that they want to build themselves exactly.

231
00:12:13.799 --> 00:12:19.159
<v Speaker 3>Self assembly actively leverages that chaotic Brownian motion rather than

232
00:12:19.240 --> 00:12:22.279
<v Speaker 3>fighting it. You aren't building a structure piece by piece

233
00:12:22.320 --> 00:12:26.440
<v Speaker 3>like a bricklayer. The pieces themselves possess the chemical knowledge

234
00:12:26.440 --> 00:12:27.559
<v Speaker 3>of how to aggregate.

235
00:12:27.720 --> 00:12:28.639
<v Speaker 2>They know where they belong.

236
00:12:28.799 --> 00:12:32.080
<v Speaker 3>Right, you provide the environment and the programmed interactions, the

237
00:12:32.159 --> 00:12:36.679
<v Speaker 3>carefully balanced hydrophilic and hydrophobic regions, the electrostatic charges, they

238
00:12:36.720 --> 00:12:39.840
<v Speaker 3>just take over. The constant shaking of the environment actually

239
00:12:39.840 --> 00:12:42.559
<v Speaker 3>helps the pieces jostle around until they lock into their

240
00:12:42.600 --> 00:12:43.840
<v Speaker 3>lowest energy state.

241
00:12:43.679 --> 00:12:46.919
<v Speaker 2>Which you have cleverly designed to be the exact structure

242
00:12:47.000 --> 00:12:50.320
<v Speaker 2>you want exactly, Like putting a bunch of magnetic legos

243
00:12:50.320 --> 00:12:52.879
<v Speaker 2>in a washing machine, right and knowing that because of

244
00:12:52.919 --> 00:12:55.919
<v Speaker 2>the specific placement of the positive and negative magnets on

245
00:12:55.960 --> 00:12:59.679
<v Speaker 2>each brick, after ten minutes of tumbling, you will always

246
00:12:59.679 --> 00:13:02.480
<v Speaker 2>pull out a perfectly formed miniature castle.

247
00:13:02.759 --> 00:13:05.679
<v Speaker 3>That is a perfect way to visualize it. The engineering

248
00:13:05.759 --> 00:13:09.919
<v Speaker 3>challenge is designing those initial energetic interactions perfectly so that

249
00:13:09.960 --> 00:13:14.519
<v Speaker 3>the local castle structure emerges reliably every single time without

250
00:13:14.559 --> 00:13:17.240
<v Speaker 3>getting trapped in a flawed intermediate shape.

251
00:13:17.279 --> 00:13:18.840
<v Speaker 2>Okay, let's pull this out of a lab and into

252
00:13:18.879 --> 00:13:22.159
<v Speaker 2>the real world. Sure, if materials can truly self assemble

253
00:13:22.200 --> 00:13:26.440
<v Speaker 2>and actively respond to stimuli through conformational changes, how does

254
00:13:26.480 --> 00:13:30.320
<v Speaker 2>this actually disrupt industries? I feel like we should look

255
00:13:30.320 --> 00:13:33.159
<v Speaker 2>at medicine first, because the current paradigm of treating disease

256
00:13:33.279 --> 00:13:34.799
<v Speaker 2>is incredibly blunt.

257
00:13:35.320 --> 00:13:39.840
<v Speaker 3>It is very blunt. Traditional medical treatments, specifically oncology and chemotherapy,

258
00:13:40.200 --> 00:13:43.519
<v Speaker 3>rely on flooding the entire biological system with a highly

259
00:13:43.559 --> 00:13:46.840
<v Speaker 3>toxic molecule. You saturate the bloodstream, and rely on the

260
00:13:46.840 --> 00:13:50.200
<v Speaker 3>fact that cancer cells metabolize and divide faster than healthy cells,

261
00:13:50.519 --> 00:13:53.000
<v Speaker 3>meaning they will hopefully absorb more of the poison before

262
00:13:53.000 --> 00:13:56.399
<v Speaker 3>the healthy cells die. It is systemic toxicity.

263
00:13:56.720 --> 00:13:59.399
<v Speaker 2>So traditional chemo is sort of like trying to weed

264
00:13:59.440 --> 00:14:03.200
<v Speaker 2>a fragile arden with the flamethrower. Basically, yes, scorching the

265
00:14:03.200 --> 00:14:06.080
<v Speaker 2>earth and hoping the weeds die slightly faster than the tomatoes.

266
00:14:06.240 --> 00:14:08.840
<v Speaker 3>That's a grim but accurate way to put it.

267
00:14:08.919 --> 00:14:12.559
<v Speaker 2>But programmable medicine that's entirely different. That is like releasing

268
00:14:12.559 --> 00:14:16.240
<v Speaker 2>a swarm of robotic beetles that are chemically programmed to

269
00:14:16.399 --> 00:14:20.639
<v Speaker 2>only eat the leaves of a specific toxic plant completely

270
00:14:20.679 --> 00:14:21.639
<v Speaker 2>ignoring everything else.

271
00:14:21.879 --> 00:14:26.080
<v Speaker 3>And the mechanism that enables that swarm behavior is molecular recognition.

272
00:14:26.720 --> 00:14:31.519
<v Speaker 3>With programmable molecules, you can synthesize a nanoscale carrier like

273
00:14:31.559 --> 00:14:34.519
<v Speaker 3>a little shell exactly, perhaps a hollow shell made of

274
00:14:34.519 --> 00:14:39.240
<v Speaker 3>self assembling polymers, and you trap the toxic chemotherapy drug

275
00:14:39.279 --> 00:14:42.320
<v Speaker 3>inside it. Now, the outside of this shell is covered

276
00:14:42.360 --> 00:14:46.600
<v Speaker 3>in specific chemical ligands or optamers. These are targeting sequences, so.

277
00:14:46.519 --> 00:14:49.879
<v Speaker 2>They are essentially physical keys looking for a very specific lock.

278
00:14:50.240 --> 00:14:54.440
<v Speaker 3>Exactly. As this nanocarrier circulates through the bloodstream, it constantly

279
00:14:54.519 --> 00:14:57.840
<v Speaker 3>bumps into healthy liver cells, heart cells, and lung cells.

280
00:14:58.240 --> 00:15:00.759
<v Speaker 3>But because those cells don't have the core spawning lock,

281
00:15:01.080 --> 00:15:02.919
<v Speaker 3>the carrier just bounces off harmlessly.

282
00:15:03.159 --> 00:15:04.159
<v Speaker 2>It just ignores them.

283
00:15:04.360 --> 00:15:09.679
<v Speaker 3>Right. However, a cancer cell often over expresses unique protein

284
00:15:09.799 --> 00:15:16.000
<v Speaker 3>signatures on its surface, tumor specific antigens. When the nanocarrier

285
00:15:16.080 --> 00:15:19.399
<v Speaker 3>bumps into the cancer cell, the ligands on its surface

286
00:15:19.519 --> 00:15:22.360
<v Speaker 3>perfectly bind to those unique proteins.

287
00:15:22.440 --> 00:15:24.519
<v Speaker 2>It's checking molecular id badges.

288
00:15:24.600 --> 00:15:27.480
<v Speaker 3>But binding isn't enough, Oh really no, Because we can

289
00:15:27.519 --> 00:15:31.080
<v Speaker 3>program the material to require multiple inputs before it releases

290
00:15:31.120 --> 00:15:34.440
<v Speaker 3>the payload. We can engineer the polymer shell so that

291
00:15:34.480 --> 00:15:37.120
<v Speaker 3>it remains tightly sealed at the normal pH of blood,

292
00:15:37.120 --> 00:15:39.759
<v Speaker 3>which is around seven point four. Okay, but the micro

293
00:15:39.840 --> 00:15:44.559
<v Speaker 3>environment immediately surrounding a fast growing tumor is highly acidic because.

294
00:15:44.360 --> 00:15:48.200
<v Speaker 2>Of how tumors metabolize glucose. So you program a polymer

295
00:15:48.240 --> 00:15:50.799
<v Speaker 2>to undergo a structural collapse only when it experiences a

296
00:15:50.799 --> 00:15:52.440
<v Speaker 2>pH of SA six point five.

297
00:15:52.639 --> 00:15:56.399
<v Speaker 3>Yes, the molecule requires the presence of the cancer protein

298
00:15:56.440 --> 00:15:59.399
<v Speaker 3>to bind, and then it requires the acidic pH to

299
00:15:59.440 --> 00:16:03.000
<v Speaker 3>break open. Only when both of those logical conditions of

300
00:16:03.000 --> 00:16:06.559
<v Speaker 3>its programming are satisfied does the shell degrade and deploy

301
00:16:06.639 --> 00:16:10.039
<v Speaker 3>the highly toxic drug directly into the tumor cell. That's incredible,

302
00:16:10.240 --> 00:16:12.799
<v Speaker 3>And once the drug is delivered, the polymer shell is

303
00:16:12.840 --> 00:16:17.279
<v Speaker 3>designed to safely hydrolyze. It just breaks down into harmless

304
00:16:17.279 --> 00:16:19.080
<v Speaker 3>byproducts that the kidneys filter out.

305
00:16:19.240 --> 00:16:22.200
<v Speaker 2>You are turning the drug from a passive chemical that

306
00:16:22.279 --> 00:16:26.519
<v Speaker 2>just poisons whatever it touches into an active decision making

307
00:16:26.559 --> 00:16:29.120
<v Speaker 2>agent that evaluates its environment before deploying.

308
00:16:29.240 --> 00:16:31.480
<v Speaker 3>The precision is staggering.

309
00:16:31.120 --> 00:16:33.480
<v Speaker 2>It really is. And if we apply that same logic

310
00:16:33.519 --> 00:16:38.759
<v Speaker 2>to manufacturing and energy, the implications are just as disruptive.

311
00:16:39.320 --> 00:16:42.320
<v Speaker 2>We're looking at the possibility of fundamentally changing how physical

312
00:16:42.320 --> 00:16:44.120
<v Speaker 2>objects wear down and break.

313
00:16:44.000 --> 00:16:49.240
<v Speaker 3>Absolutely because in current macroscopic manufacturing processes are mostly subtractive

314
00:16:49.320 --> 00:16:53.360
<v Speaker 3>or additive in very energy intensive ways. You cut away material,

315
00:16:53.399 --> 00:16:56.399
<v Speaker 3>which generates a massive waste, or you melt and cast

316
00:16:56.440 --> 00:16:59.960
<v Speaker 3>material under extreme heat. If we move toward self assembling material,

317
00:17:00.240 --> 00:17:03.840
<v Speaker 3>you eliminate those crude fabrication steps. You synthesize the raw

318
00:17:03.960 --> 00:17:07.960
<v Speaker 3>programmed precursors, and the material forms exactly the required internal

319
00:17:08.079 --> 00:17:10.079
<v Speaker 3>architecture with minimal to no waste.

320
00:17:10.240 --> 00:17:13.240
<v Speaker 2>I mean, think about the static objects you interact with constantly.

321
00:17:14.000 --> 00:17:17.200
<v Speaker 2>A concrete bridge, the casing of your phone, the paint

322
00:17:17.240 --> 00:17:19.559
<v Speaker 2>on your car. They are all caught in a one

323
00:17:19.559 --> 00:17:20.359
<v Speaker 2>way street.

324
00:17:20.119 --> 00:17:22.839
<v Speaker 3>Of entropy, always breaking down exactly.

325
00:17:22.759 --> 00:17:27.039
<v Speaker 2>From the moment they are manufactured. They slowly degrade, accumulate microfractures,

326
00:17:27.079 --> 00:17:31.359
<v Speaker 2>and eventually fail. Imagine if those materials were designed with

327
00:17:31.480 --> 00:17:33.640
<v Speaker 2>dynamic reversible chemical.

328
00:17:33.240 --> 00:17:37.359
<v Speaker 3>Bonds you're referring to intrinsic self healing materials. Instead of

329
00:17:37.440 --> 00:17:41.319
<v Speaker 3>using traditional permanent covalent bonds that snap under stress and

330
00:17:41.359 --> 00:17:46.039
<v Speaker 3>stay broken, you build the material using dynamic covalent chemistry

331
00:17:46.160 --> 00:17:50.400
<v Speaker 3>or supermolecular networks. These are networks that can continuously break

332
00:17:50.440 --> 00:17:52.079
<v Speaker 3>and reform their crosslingks.

333
00:17:52.559 --> 00:17:54.119
<v Speaker 2>So if you get a scratch on your car door,

334
00:17:54.480 --> 00:17:57.559
<v Speaker 2>the kinetic energy of the impact breaks the local bonds, right,

335
00:17:57.720 --> 00:17:59.920
<v Speaker 2>But because the material is programmed to seek its low

336
00:18:00.279 --> 00:18:04.079
<v Speaker 2>energy state and those broken bonds are highly reactive, the

337
00:18:04.119 --> 00:18:07.200
<v Speaker 2>molecular coat of the paint essentially initiates a sequence to

338
00:18:07.400 --> 00:18:08.680
<v Speaker 2>rebind across the gap.

339
00:18:08.839 --> 00:18:11.960
<v Speaker 3>Yes, the polymer chain slowly diffuse across the scratch and

340
00:18:12.039 --> 00:18:12.920
<v Speaker 3>tangle back together.

341
00:18:13.000 --> 00:18:15.200
<v Speaker 2>So the scratch literally heals itself just sitting in the

342
00:18:15.279 --> 00:18:16.039
<v Speaker 2>driveway to.

343
00:18:16.160 --> 00:18:18.519
<v Speaker 3>Sit in there. Yeah. And we also see this massive

344
00:18:18.519 --> 00:18:22.039
<v Speaker 3>potential in the energy sector, specifically regarding efficiency and adaptation.

345
00:18:22.079 --> 00:18:26.559
<v Speaker 3>Well like solar panels exactly, our current energy capture technologies

346
00:18:26.599 --> 00:18:30.119
<v Speaker 3>are highly rigid. A silicon solar panel sits on a roof.

347
00:18:30.519 --> 00:18:34.240
<v Speaker 3>It operates optimally at a very specific temperature and a

348
00:18:34.319 --> 00:18:36.119
<v Speaker 3>specific angle of light, but.

349
00:18:36.119 --> 00:18:39.680
<v Speaker 2>The environment is constantly changing. The sun moves, clouds, roll

350
00:18:39.759 --> 00:18:41.119
<v Speaker 2>in the temperature spikes.

351
00:18:41.759 --> 00:18:46.440
<v Speaker 3>If we integrate adaptive programmable polymers into the photovoltaic matrix,

352
00:18:46.880 --> 00:18:50.920
<v Speaker 3>the material could autonomously optimize itself. We can have coatings

353
00:18:50.960 --> 00:18:55.799
<v Speaker 3>that dynamically reconfigure their surface nanostructure to alter their refractive

354
00:18:55.799 --> 00:18:56.839
<v Speaker 3>index throughout.

355
00:18:56.519 --> 00:18:59.480
<v Speaker 2>The day to maximize light absorption depending on the angle

356
00:18:59.519 --> 00:19:01.279
<v Speaker 2>of the income photons exactly.

357
00:19:01.519 --> 00:19:04.640
<v Speaker 3>Or think about battery technology. The biggest issue with lithium

358
00:19:04.680 --> 00:19:08.920
<v Speaker 3>ion cells is that repeated charging cycles cause physical microfractures

359
00:19:08.920 --> 00:19:10.359
<v Speaker 3>in the electrodes.

360
00:19:09.839 --> 00:19:14.000
<v Speaker 2>Right, which eventually kills the battery's capacity. The battery essentially

361
00:19:14.039 --> 00:19:16.799
<v Speaker 2>crushes itself from the inside out every time you charge.

362
00:19:16.519 --> 00:19:20.640
<v Speaker 3>It, exactly. But if we build the internal separators and

363
00:19:20.680 --> 00:19:25.359
<v Speaker 3>electrodes out of programmable self healing polymers, the battery could

364
00:19:25.400 --> 00:19:30.400
<v Speaker 3>continuously repair those microfractures in real time. Wow, the material

365
00:19:30.519 --> 00:19:34.640
<v Speaker 3>senses the mechanical stress and autonomously patches the structural defect,

366
00:19:34.960 --> 00:19:38.240
<v Speaker 3>extending the life span of the energy storage system exponentially.

367
00:19:38.559 --> 00:19:40.000
<v Speaker 2>But you know, to pull back and look at the

368
00:19:40.079 --> 00:19:43.839
<v Speaker 2>underlying logic of all these examples. For a polymer shell

369
00:19:44.640 --> 00:19:47.160
<v Speaker 2>to quote unquote know it has found a cancer cell,

370
00:19:47.559 --> 00:19:49.400
<v Speaker 2>or for a solar panel coding to know it needs

371
00:19:49.400 --> 00:19:51.200
<v Speaker 2>to alter its refractive index, or.

372
00:19:51.160 --> 00:19:52.599
<v Speaker 3>For a battery to trigger a.

373
00:19:52.559 --> 00:19:56.559
<v Speaker 2>Repair exactly, the material has to process information. It's quite

374
00:19:56.559 --> 00:19:59.319
<v Speaker 2>little and making a choice based on environmental inputs, and

375
00:19:59.359 --> 00:20:02.319
<v Speaker 2>that brings us I think the most conceptually difficult part

376
00:20:02.359 --> 00:20:02.599
<v Speaker 2>of this.

377
00:20:02.599 --> 00:20:07.200
<v Speaker 3>Shift molecular computation. Molecular computation, Yeah, it requires a complete

378
00:20:07.240 --> 00:20:11.279
<v Speaker 3>reimagining of what computation actually is. We are heavily biased

379
00:20:11.279 --> 00:20:14.559
<v Speaker 3>toward electronic computation. We think of logic as electricity flowing

380
00:20:14.559 --> 00:20:18.759
<v Speaker 3>through silicon pathways, governed by physical transistors acting as gits.

381
00:20:19.039 --> 00:20:24.119
<v Speaker 2>Right, a motherboard copper traces electrons, That is a computer. Yeah,

382
00:20:24.160 --> 00:20:27.400
<v Speaker 2>so how does a beaker full of liquid chemicals perform

383
00:20:27.480 --> 00:20:28.440
<v Speaker 2>logic operations?

384
00:20:28.440 --> 00:20:31.039
<v Speaker 3>Well, a logic gait fundamentally is just a system that

385
00:20:31.079 --> 00:20:34.839
<v Speaker 3>takes one or more inputs, processes them according to a rule,

386
00:20:35.319 --> 00:20:39.440
<v Speaker 3>and produces a definitive output. An electronic transistor does this

387
00:20:39.480 --> 00:20:43.039
<v Speaker 3>by blocking or allowing the flow of electrons, But molecules

388
00:20:43.079 --> 00:20:46.400
<v Speaker 3>can do this by blocking or allowing chemical reactions. Interesting

389
00:20:46.440 --> 00:20:50.079
<v Speaker 3>in programmable chemistry, the molecules themselves perform the computations through

390
00:20:50.079 --> 00:20:54.119
<v Speaker 3>their physical interactions. We can design chemical reaction networks to

391
00:20:54.240 --> 00:20:57.839
<v Speaker 3>flawlessly mimic the exact same boollion logic operations.

392
00:20:57.880 --> 00:21:01.519
<v Speaker 2>Your laptop uses like eighty gas or gates.

393
00:21:01.279 --> 00:21:03.279
<v Speaker 3>Sandy gates or gates, not gates all of them.

394
00:21:03.400 --> 00:21:05.359
<v Speaker 2>Let's actually map the physics of an A and D

395
00:21:05.480 --> 00:21:07.279
<v Speaker 2>gate for the listener, because I think this is crucial.

396
00:21:07.400 --> 00:21:07.720
<v Speaker 3>Okay.

397
00:21:07.799 --> 00:21:10.359
<v Speaker 2>In a silicon computer, an A and D gate means

398
00:21:10.640 --> 00:21:14.039
<v Speaker 2>if electrical signal A is present A and D electrical

399
00:21:14.079 --> 00:21:17.640
<v Speaker 2>signal B is present, then send electrical signal C. Right.

400
00:21:17.759 --> 00:21:20.440
<v Speaker 2>How do molecules physically perform that exact same map.

401
00:21:20.559 --> 00:21:23.240
<v Speaker 3>We use a mechanism called a loft street or allosteric

402
00:21:23.319 --> 00:21:28.519
<v Speaker 3>regulation lost. Right, Imagine three different custom designed molecules in

403
00:21:28.559 --> 00:21:31.799
<v Speaker 3>a solution. Let's call them input A, input B, and

404
00:21:31.839 --> 00:21:35.880
<v Speaker 3>the processing unit molecule C. Molecule C is designed with

405
00:21:35.960 --> 00:21:40.799
<v Speaker 3>a very specific tightly folded geometry. In its default state,

406
00:21:40.920 --> 00:21:44.559
<v Speaker 3>its active site, the part that can create the output reaction,

407
00:21:44.759 --> 00:21:49.200
<v Speaker 3>is hidden hysterically injured. Exactly, it physically cannot react with

408
00:21:49.319 --> 00:21:52.240
<v Speaker 3>anything in the environment because the active site is buried

409
00:21:52.279 --> 00:21:53.440
<v Speaker 3>inside the folded structure.

410
00:21:53.519 --> 00:21:54.440
<v Speaker 2>It's locked from the inside.

411
00:21:54.519 --> 00:21:58.359
<v Speaker 3>It's locked now. Molecule C has two specific binding pockets

412
00:21:58.359 --> 00:22:01.200
<v Speaker 3>on its exterior. If on the input A is present

413
00:22:01.240 --> 00:22:03.799
<v Speaker 3>in the liquid, it binds to the first pocket. The

414
00:22:03.839 --> 00:22:07.960
<v Speaker 3>molecule shifts slightly, but the active site remains hidden. No output.

415
00:22:08.480 --> 00:22:10.599
<v Speaker 3>If only input B is present, it binds to the

416
00:22:10.640 --> 00:22:13.519
<v Speaker 3>second pocket. Another slight shift, but still.

417
00:22:13.240 --> 00:22:15.480
<v Speaker 2>No output, so the logic gate remains closed.

418
00:22:15.519 --> 00:22:18.519
<v Speaker 3>But if input A and input B are present simultaneously,

419
00:22:18.839 --> 00:22:22.440
<v Speaker 3>they both bind to the exterior pockets. The combined mechanical

420
00:22:22.480 --> 00:22:26.240
<v Speaker 3>stress of both inputs binding causes a massive conformational change

421
00:22:26.519 --> 00:22:27.519
<v Speaker 3>in molecule See.

422
00:22:27.400 --> 00:22:28.359
<v Speaker 2>It shifts shape.

423
00:22:28.599 --> 00:22:32.640
<v Speaker 3>It drastically changes its physical shape, unzipping and exposing the

424
00:22:32.680 --> 00:22:37.359
<v Speaker 3>hidden active site. Now and only now, molecule C can

425
00:22:37.400 --> 00:22:38.920
<v Speaker 3>catalyze the final reaction.

426
00:22:39.200 --> 00:22:39.599
<v Speaker 2>Wow.

427
00:22:39.720 --> 00:22:44.079
<v Speaker 3>That chemical reaction is the computation. The system has processed

428
00:22:44.079 --> 00:22:48.039
<v Speaker 3>two distinct pieces of information applied to logical rule and

429
00:22:48.160 --> 00:22:49.160
<v Speaker 3>arrived at an output.

430
00:22:49.400 --> 00:22:53.480
<v Speaker 2>Is literally doing math with physical shapes. The stereochemistry is

431
00:22:53.519 --> 00:22:57.319
<v Speaker 2>the logic gait. You've basically replaced a silicon transistor with

432
00:22:57.319 --> 00:22:58.960
<v Speaker 2>a shape shifting protein or polymer.

433
00:22:59.119 --> 00:22:59.960
<v Speaker 3>Yes, exactly.

434
00:23:00.319 --> 00:23:02.000
<v Speaker 2>And once you realize, you can build an A and

435
00:23:02.079 --> 00:23:04.079
<v Speaker 2>D gate, you can build an our gate. You can

436
00:23:04.119 --> 00:23:07.079
<v Speaker 2>string them together into complex cascades where the output of

437
00:23:07.119 --> 00:23:09.279
<v Speaker 2>one chemical gait becomes the input for the next.

438
00:23:09.519 --> 00:23:12.519
<v Speaker 3>Leonard Adelman actually proved this back in nineteen ninety four.

439
00:23:12.799 --> 00:23:15.880
<v Speaker 3>He used custom strands of DNA to computationally solve the

440
00:23:15.880 --> 00:23:17.160
<v Speaker 3>traveling salesman.

441
00:23:16.759 --> 00:23:20.559
<v Speaker 2>Problem, which is a notoriously complex mathematical routing problem.

442
00:23:20.279 --> 00:23:23.400
<v Speaker 3>Very complex. He encoded the cities and the flight paths

443
00:23:23.400 --> 00:23:26.359
<v Speaker 3>into DNA sequences, mix them in a test tube, and

444
00:23:26.440 --> 00:23:29.640
<v Speaker 3>let the sheer parallel processing power of trillions of molecules

445
00:23:29.720 --> 00:23:32.319
<v Speaker 3>naturally binding together compute the most efficient route.

446
00:23:32.440 --> 00:23:36.480
<v Speaker 2>But I mean, why go through the trouble. My laptop

447
00:23:36.599 --> 00:23:39.440
<v Speaker 2>can solve the traveling salesman problem in a fraction of

448
00:23:39.480 --> 00:23:43.720
<v Speaker 2>a millisecond. Why painstakingly build a computer out of liquid

449
00:23:43.720 --> 00:23:48.799
<v Speaker 2>polymers or DNA when silicon is so unimaginably fast.

450
00:23:48.599 --> 00:23:51.720
<v Speaker 3>Because of the environment. Silicon computers are incredibly fast, but

451
00:23:51.759 --> 00:23:54.599
<v Speaker 3>they are incredibly fragile and entirely macroscopic.

452
00:23:54.799 --> 00:23:55.279
<v Speaker 2>That's true.

453
00:23:55.440 --> 00:23:58.519
<v Speaker 3>Cannot insert a microchip inside a living human cell to

454
00:23:58.599 --> 00:24:02.240
<v Speaker 3>monitor its metabolism. You can drop a silicon motherboard into

455
00:24:02.279 --> 00:24:06.480
<v Speaker 3>a highly corrosive, toxic chemical spill to analyze the specific

456
00:24:06.519 --> 00:24:07.440
<v Speaker 3>isotopes present.

457
00:24:07.519 --> 00:24:09.359
<v Speaker 2>No silicon shorts out in water.

458
00:24:09.279 --> 00:24:12.240
<v Speaker 3>It melts under extreme heat, and it triggers immune responses

459
00:24:12.279 --> 00:24:13.440
<v Speaker 3>in biological systems.

460
00:24:13.480 --> 00:24:17.039
<v Speaker 2>So molecular computers are built out of the very fabric

461
00:24:17.079 --> 00:24:18.640
<v Speaker 2>of those extreme environments.

462
00:24:18.759 --> 00:24:22.240
<v Speaker 3>They operate natively in the wet, messy, chemical world. They

463
00:24:22.279 --> 00:24:25.440
<v Speaker 3>open the door to processing complex information precisely in the

464
00:24:25.440 --> 00:24:29.119
<v Speaker 3>places where electronics are entirely useless. You could inject a

465
00:24:29.200 --> 00:24:33.400
<v Speaker 3>molecular computer into an aquifer to map heavy metal contamination

466
00:24:33.559 --> 00:24:38.000
<v Speaker 3>molecule by molecule, processing the data locally and neutralizing the

467
00:24:38.079 --> 00:24:42.359
<v Speaker 3>toxins on site. You are decentralizing processing power, moving it

468
00:24:42.440 --> 00:24:44.640
<v Speaker 3>directly to the microscopic site of the.

469
00:24:44.559 --> 00:24:47.799
<v Speaker 2>Problem, which honestly forces us to step back and look

470
00:24:47.799 --> 00:24:49.880
<v Speaker 2>at the philosophical crater this leaves behind.

471
00:24:50.000 --> 00:24:50.680
<v Speaker 3>It really does.

472
00:24:50.839 --> 00:24:53.960
<v Speaker 2>If we have purely synthetic systems that can compute logic,

473
00:24:54.519 --> 00:24:57.720
<v Speaker 2>that can make independent decisions based on their environment, that

474
00:24:57.799 --> 00:25:01.079
<v Speaker 2>can self organize from chaos into order and self repair

475
00:25:01.119 --> 00:25:04.920
<v Speaker 2>when damage. Yeah, we're backing ourselves into a corner. Regarding

476
00:25:04.960 --> 00:25:05.720
<v Speaker 2>the definition of.

477
00:25:05.680 --> 00:25:09.240
<v Speaker 3>Life, the boundary completely dissolves. This actually goes back to

478
00:25:09.480 --> 00:25:12.839
<v Speaker 3>Erwin Schrodinger's famous nineteen forty four book What Is Life Right,

479
00:25:13.079 --> 00:25:15.559
<v Speaker 3>where he proposed that life at its core is driven

480
00:25:15.599 --> 00:25:17.000
<v Speaker 3>by an a periodic.

481
00:25:16.480 --> 00:25:18.960
<v Speaker 2>Crystal, the physical structure that contains information.

482
00:25:19.160 --> 00:25:21.880
<v Speaker 3>Exactly when you look at biology through the lens of

483
00:25:21.960 --> 00:25:26.400
<v Speaker 3>programmable matter, life itself loses some of its mystique. It

484
00:25:26.440 --> 00:25:29.440
<v Speaker 3>can be viewed as just a highly sophisticated, highly evolved

485
00:25:29.480 --> 00:25:31.079
<v Speaker 3>form of programmable chemistry.

486
00:25:31.440 --> 00:25:34.079
<v Speaker 2>Like we look at a white blood cell hunting down

487
00:25:34.119 --> 00:25:37.119
<v Speaker 2>a bacterial pathogen, chasing it through the bloodstream, and it

488
00:25:37.160 --> 00:25:40.240
<v Speaker 2>looks like a conscious living action. It looks like agency.

489
00:25:40.359 --> 00:25:42.640
<v Speaker 2>It does look like that, but fundamentally we know it

490
00:25:42.720 --> 00:25:46.559
<v Speaker 2>is a complex cascade of chemical logic gaits opening and closing.

491
00:25:46.839 --> 00:25:50.720
<v Speaker 2>It's following a gradient of chemical inputs executing in internal code.

492
00:25:50.799 --> 00:25:54.480
<v Speaker 3>Exactly. A biological cell is just a localized pocket of

493
00:25:54.559 --> 00:25:59.680
<v Speaker 3>billions of molecules interacting according to complex evolutionary rules to

494
00:26:00.000 --> 00:26:04.559
<v Speaker 3>reduce emergent behavior that we intuitively label as living. By

495
00:26:04.640 --> 00:26:09.799
<v Speaker 3>creating purely synthetic, engineered systems that exhibit these exact same properties,

496
00:26:10.519 --> 00:26:16.200
<v Speaker 3>self organization, adaptation to stimuli, rudimentary computational decision making, scientists

497
00:26:16.519 --> 00:26:19.599
<v Speaker 3>are proving that the mechanics of life are not exclusive

498
00:26:19.640 --> 00:26:20.359
<v Speaker 3>to biology.

499
00:26:20.799 --> 00:26:22.680
<v Speaker 2>It feels like we are scripping away the magic of

500
00:26:22.720 --> 00:26:26.480
<v Speaker 2>biology and replacing it with pure thermodynamics and engineering. Yeah,

501
00:26:26.519 --> 00:26:28.759
<v Speaker 2>in a way, which is awe inspiring because it means

502
00:26:28.759 --> 00:26:30.799
<v Speaker 2>we can build with it. Yeah, But it also means

503
00:26:30.799 --> 00:26:33.880
<v Speaker 2>that the line between a biological, living organism and a

504
00:26:33.960 --> 00:26:37.640
<v Speaker 2>highly programmed piece of synthetic plastic is getting dangerously thin.

505
00:26:37.799 --> 00:26:38.599
<v Speaker 3>It's very thin.

506
00:26:38.799 --> 00:26:42.440
<v Speaker 2>Are we just redefining life to make our engineering sound cooler,

507
00:26:43.039 --> 00:26:46.880
<v Speaker 2>or are these synthetic systems actually approaching something fundamentally alive.

508
00:26:47.640 --> 00:26:51.200
<v Speaker 3>It is a profound philosophical shift in how we categorize

509
00:26:51.200 --> 00:26:55.640
<v Speaker 3>the universe. We have historically separated matter into the animate

510
00:26:55.720 --> 00:26:59.440
<v Speaker 3>and the inanimate, the living and the dead. But programmable

511
00:26:59.480 --> 00:27:03.319
<v Speaker 3>material exist in a gray zone. They challenge our understanding

512
00:27:03.359 --> 00:27:06.839
<v Speaker 3>of matter itself. Matter is no longer a passive substance.

513
00:27:07.160 --> 00:27:10.759
<v Speaker 3>It is an active participant in processes that closely resemble

514
00:27:10.839 --> 00:27:14.960
<v Speaker 3>thought and decision making. We are proving that agency, the

515
00:27:15.000 --> 00:27:17.480
<v Speaker 3>ability to react and adapt, is just a function of

516
00:27:17.519 --> 00:27:18.799
<v Speaker 3>structural complexity.

517
00:27:19.240 --> 00:27:22.559
<v Speaker 2>But as incredible and world changing as all of this sounds,

518
00:27:23.000 --> 00:27:26.200
<v Speaker 2>we do need to ground this in the immediate reality. Definitely,

519
00:27:26.240 --> 00:27:29.559
<v Speaker 2>we aren't quite living in this fully programmable, self assembling

520
00:27:29.640 --> 00:27:32.079
<v Speaker 2>utopia yet. I mean, I still have to charge my

521
00:27:32.079 --> 00:27:34.960
<v Speaker 2>phone and my car still get scratched. Designing matter from

522
00:27:35.000 --> 00:27:38.680
<v Speaker 2>the ground up comes with massive fundamental engineering hurdles. Huge

523
00:27:38.759 --> 00:27:42.000
<v Speaker 2>hurdles and the biggest one being the challenge of nonlinear complexity.

524
00:27:42.079 --> 00:27:45.279
<v Speaker 3>Yeah, that is the core difficulty of this entire endeavor,

525
00:27:45.319 --> 00:27:48.519
<v Speaker 3>and it really comes down to the battle between kinetics

526
00:27:48.519 --> 00:27:53.480
<v Speaker 3>and thermodynamics. Okay, designing these molecular systems to behave predictably

527
00:27:53.599 --> 00:27:58.680
<v Speaker 3>in complex, messy, real world environments is agonizingly difficult. In

528
00:27:58.720 --> 00:28:02.920
<v Speaker 3>a tightly controlled sterile lab experiment with cure distilled water

529
00:28:03.039 --> 00:28:06.720
<v Speaker 3>and exact temperature controls, you can make a targeted nanocarrier

530
00:28:06.839 --> 00:28:11.039
<v Speaker 3>work flawlessly, sure, But the real world, the human bloodstream,

531
00:28:11.240 --> 00:28:14.880
<v Speaker 3>the ocean, an industrial vat is full of fluctuating temperatures,

532
00:28:15.160 --> 00:28:19.519
<v Speaker 3>chaotic fluid dynamics, impurities, and unexpected chemical interactions, and.

533
00:28:19.599 --> 00:28:23.160
<v Speaker 2>Molecular interactions are highly nonlinear. If a single line of

534
00:28:23.200 --> 00:28:26.599
<v Speaker 2>code in the smartphone app is written wrong, the app crashes,

535
00:28:26.680 --> 00:28:29.240
<v Speaker 2>It throws an error code. It's annoying, but the errors

536
00:28:29.240 --> 00:28:31.799
<v Speaker 2>contained the phone doesn't turn into a toaster. But if

537
00:28:31.799 --> 00:28:34.799
<v Speaker 2>a chemical bond goes rogue in a self assembling material,

538
00:28:35.039 --> 00:28:36.880
<v Speaker 2>the entire architecture can mutate.

539
00:28:37.200 --> 00:28:40.839
<v Speaker 3>Because of the nonlinearity, A tiny change doesn't just create

540
00:28:40.880 --> 00:28:44.400
<v Speaker 3>a proportional error, it can cascade into a completely different

541
00:28:44.480 --> 00:28:48.359
<v Speaker 3>physical outcome. A slight shift in the ambient pH or

542
00:28:48.359 --> 00:28:51.759
<v Speaker 3>the presence of a stray enzyme might alter the binding

543
00:28:51.799 --> 00:28:55.240
<v Speaker 3>affinity of your polymer just enough that instead of assembling

544
00:28:55.240 --> 00:28:59.200
<v Speaker 3>into a neat, hollow shell, the molecules fold inside out,

545
00:28:59.519 --> 00:29:02.480
<v Speaker 3>or they actgate into a massive, useless clump.

546
00:29:02.599 --> 00:29:05.559
<v Speaker 2>It's like trying to choreograph an intricate ballet for a

547
00:29:05.720 --> 00:29:09.200
<v Speaker 2>trillion microscopic dancers who are all blindfolded, and if one

548
00:29:09.319 --> 00:29:12.640
<v Speaker 2>dancer trips, the kinetic energy knocks over the entire continent.

549
00:29:12.759 --> 00:29:15.279
<v Speaker 3>That's a great way to put it. And this nonlinearity

550
00:29:15.440 --> 00:29:18.839
<v Speaker 3>makes the physical scaling of these systems a massive bottleneck.

551
00:29:19.279 --> 00:29:24.279
<v Speaker 3>The unpredictability scales exponentially as you increase the volume. Synthesizing

552
00:29:24.319 --> 00:29:28.000
<v Speaker 3>a microgram of a programmable polymer in a microphlitic chip

553
00:29:28.119 --> 00:29:30.519
<v Speaker 3>is one thing. Producing a ton of it in a

554
00:29:30.559 --> 00:29:33.279
<v Speaker 3>giant industrial vat where the temperature at the center of

555
00:29:33.319 --> 00:29:35.799
<v Speaker 3>the vat is slightly different than the temperature near the

556
00:29:35.799 --> 00:29:40.279
<v Speaker 3>cooling jacket, introduces thermodynamic gradients that completely ruin the self

557
00:29:40.279 --> 00:29:41.200
<v Speaker 3>assembly process.

558
00:29:41.480 --> 00:29:43.400
<v Speaker 2>You can't just scale up the beaker to a swimming

559
00:29:43.440 --> 00:29:46.359
<v Speaker 2>pool now, you really can't. The physics of mass transfer

560
00:29:46.400 --> 00:29:51.440
<v Speaker 2>and heat dissipation completely change and fixing that scaling issue

561
00:29:51.640 --> 00:29:55.359
<v Speaker 2>requires a completely new paradigm of interdisciplinary science.

562
00:29:55.440 --> 00:29:59.559
<v Speaker 3>It does. Progress in programmable matter cannot happen in a

563
00:29:59.559 --> 00:30:04.400
<v Speaker 3>tradition academic silo. It requires a seamless, almost unnatural convergence

564
00:30:04.440 --> 00:30:08.279
<v Speaker 3>of fields because it touches everything exactly. You need organic

565
00:30:08.359 --> 00:30:11.640
<v Speaker 3>chemists who inherently think like software engineers. You need computer

566
00:30:11.720 --> 00:30:16.559
<v Speaker 3>scientists who deeply understand quantum thermodynamics and fluid dynamics. We

567
00:30:16.640 --> 00:30:19.759
<v Speaker 3>are having to invent new physical tools just to observe

568
00:30:19.799 --> 00:30:23.240
<v Speaker 3>these systems operating in real time, let alone manipulate them.

569
00:30:23.359 --> 00:30:23.799
<v Speaker 2>Wow.

570
00:30:24.200 --> 00:30:27.000
<v Speaker 3>Our theoretical vision of what these molecules can do is

571
00:30:27.079 --> 00:30:29.799
<v Speaker 3>currently running slightly ahead of our practical tool set to

572
00:30:29.880 --> 00:30:31.319
<v Speaker 3>reliably manufacture them.

573
00:30:31.720 --> 00:30:34.759
<v Speaker 2>And beyond the sheer technical hurdles of scaling, there is

574
00:30:34.759 --> 00:30:38.680
<v Speaker 2>the undeniable engineering challenge of control and safety. Oh absolutely,

575
00:30:38.960 --> 00:30:42.039
<v Speaker 2>When you begin manufacturing synthetic materials that have the physical

576
00:30:42.039 --> 00:30:46.279
<v Speaker 2>capacity to adapt to environments, process local information, and potentially

577
00:30:46.279 --> 00:30:49.599
<v Speaker 2>self catalyze or self replicate, you're introducing a completely new

578
00:30:49.720 --> 00:30:52.599
<v Speaker 2>mechanical risk vector into the physical world.

579
00:30:52.799 --> 00:30:55.880
<v Speaker 3>The control problem is paramount I mean, how do you

580
00:30:56.000 --> 00:30:59.960
<v Speaker 3>engineer reliable constraints into materials that have their own internal

581
00:31:00.079 --> 00:31:05.759
<v Speaker 3>programmatic agency. A self assembling polymer system designed to process

582
00:31:05.880 --> 00:31:08.839
<v Speaker 3>toxins and an oil spill is operating in a highly

583
00:31:08.920 --> 00:31:12.480
<v Speaker 3>chaotic environment. If it encounters a novel chemical stressor in

584
00:31:12.519 --> 00:31:15.039
<v Speaker 3>the ocean that we didn't model in the lab, its

585
00:31:15.119 --> 00:31:19.480
<v Speaker 3>dynamic bonds might reconfigure in ways that alter its primary function.

586
00:31:19.599 --> 00:31:23.400
<v Speaker 2>It's the mechanical reality of unintended consequences. If a synthetic

587
00:31:23.440 --> 00:31:27.039
<v Speaker 2>polymer begins catalyzing reactions out of control, you can't just

588
00:31:27.160 --> 00:31:29.480
<v Speaker 2>hit a macroscopic power button to turn it off. No,

589
00:31:29.799 --> 00:31:32.799
<v Speaker 2>it doesn't plug into the wall. The process is entirely

590
00:31:32.799 --> 00:31:34.200
<v Speaker 2>localized in chemical.

591
00:31:33.920 --> 00:31:36.519
<v Speaker 3>Which is why a massive portion of the current engineering

592
00:31:36.519 --> 00:31:40.680
<v Speaker 3>effort is focused purely on designing robust containment mechanisms and

593
00:31:40.720 --> 00:31:41.799
<v Speaker 3>physical kill switches.

594
00:31:41.839 --> 00:31:43.559
<v Speaker 2>How do you put a kill switch on a molecule?

595
00:31:43.799 --> 00:31:49.279
<v Speaker 3>By engineering synthetic oxotrophy osodrophy? You design the self assembling

596
00:31:49.279 --> 00:31:52.960
<v Speaker 3>material so that its structural integrity is entirely dependent on

597
00:31:53.000 --> 00:31:57.039
<v Speaker 3>a highly specific, rare synthetic molecule that does not exist

598
00:31:57.119 --> 00:32:00.799
<v Speaker 3>in nature. You must constantly feed the system this artificial

599
00:32:00.839 --> 00:32:02.039
<v Speaker 3>stabilizer oh ICEE.

600
00:32:02.039 --> 00:32:05.279
<v Speaker 2>So if the material escapes the controlled environment or starts

601
00:32:05.319 --> 00:32:09.359
<v Speaker 2>replicating unpredictably. It rapidly exhausts its local supply of the

602
00:32:09.359 --> 00:32:13.759
<v Speaker 2>stabilizer exactly and without it, the thermodynamic bonds become unstable

603
00:32:14.039 --> 00:32:17.920
<v Speaker 2>and the entire structure rapidly diplemerizes back into inner dust.

604
00:32:18.039 --> 00:32:21.440
<v Speaker 3>Yes, you engine your fragility into the code. Ensuring the

605
00:32:21.480 --> 00:32:25.799
<v Speaker 3>safe deployment of programmable chemistry requires that these containment strategies,

606
00:32:25.880 --> 00:32:29.920
<v Speaker 3>these physical failsafs, are integrated into the molecular architecture from

607
00:32:29.960 --> 00:32:33.279
<v Speaker 3>step one. They have to be heavily scrutinized by cross

608
00:32:33.319 --> 00:32:36.759
<v Speaker 3>disciplinary regulatory frameworks before they ever leave a contained lap.

609
00:32:36.920 --> 00:32:39.599
<v Speaker 2>So to bring all these threads together, the trajectory of

610
00:32:39.640 --> 00:32:44.079
<v Speaker 2>the materials science is undeniably clear. The hard historical lines

611
00:32:44.079 --> 00:32:47.960
<v Speaker 2>we've drawn for centuries between chemistry, biology, and computer science

612
00:32:48.240 --> 00:32:49.400
<v Speaker 2>are blurring entirely.

613
00:32:49.440 --> 00:32:50.119
<v Speaker 3>They're merging.

614
00:32:50.279 --> 00:32:55.240
<v Speaker 2>They're merging into a single unified discipline of programmatic molecular design.

615
00:32:56.000 --> 00:32:59.759
<v Speaker 2>Information and physical matter are becoming so deeply intertwined that

616
00:32:59.799 --> 00:33:03.319
<v Speaker 2>you can barely separate the two concepts anymore. We are

617
00:33:03.359 --> 00:33:07.160
<v Speaker 2>completing the shift from a world of macroscopic external control,

618
00:33:07.240 --> 00:33:09.839
<v Speaker 2>beating the dumb clay into shape, to a world of

619
00:33:09.880 --> 00:33:14.680
<v Speaker 2>microscopic internal instruction where the clay structurally knows what it

620
00:33:14.720 --> 00:33:15.240
<v Speaker 2>wants to be.

621
00:33:15.440 --> 00:33:17.880
<v Speaker 3>It is a complete paradigm shift in our relationship with

622
00:33:17.920 --> 00:33:21.480
<v Speaker 3>the physical universe. It moves our focus from static, dead

623
00:33:21.559 --> 00:33:25.720
<v Speaker 3>objects to dynamic responsive systems. We are no longer limited

624
00:33:25.759 --> 00:33:28.720
<v Speaker 3>by what materials inherently are based on their extraction from

625
00:33:28.759 --> 00:33:32.000
<v Speaker 3>the Earth. We're only limited by what we can mathematically

626
00:33:32.079 --> 00:33:34.400
<v Speaker 3>program them to do. We are opening up a space

627
00:33:34.440 --> 00:33:38.440
<v Speaker 3>of engineering possibilities limited only by our ability to imagine

628
00:33:38.480 --> 00:33:40.839
<v Speaker 3>and compute new molecular architectures.

629
00:33:40.920 --> 00:33:42.319
<v Speaker 2>And as we move into that space, I'm going to

630
00:33:42.400 --> 00:33:45.519
<v Speaker 2>leave you with a completely new implication to Mollover. We've

631
00:33:45.559 --> 00:33:48.240
<v Speaker 2>talked a lot today about how this technology is erasing

632
00:33:48.279 --> 00:33:52.079
<v Speaker 2>the distinction between hardware and software. Think about your smartphone.

633
00:33:52.119 --> 00:33:54.400
<v Speaker 2>When you want your phone to do something new, you

634
00:33:54.440 --> 00:33:56.920
<v Speaker 2>don't throw away the physical glass in silicon and buy

635
00:33:56.920 --> 00:34:00.160
<v Speaker 2>a new device. You connect to a network, download to

636
00:34:00.240 --> 00:34:03.079
<v Speaker 2>software update, and the internal behavior of the phone changes.

637
00:34:03.839 --> 00:34:06.240
<v Speaker 2>If we apply that same logic to the physical world,

638
00:34:06.759 --> 00:34:09.239
<v Speaker 2>if the walls of your house, the frame of your car,

639
00:34:09.400 --> 00:34:11.679
<v Speaker 2>or the fibers of your clothing are built from a

640
00:34:11.760 --> 00:34:15.320
<v Speaker 2>highly responsive programmable polymers. We are looking at a future

641
00:34:15.400 --> 00:34:18.920
<v Speaker 2>where tangible objects are just as adaptable as an operating system.

642
00:34:19.440 --> 00:34:22.239
<v Speaker 2>You might not physically replace materials when they wear out

643
00:34:22.679 --> 00:34:24.280
<v Speaker 2>or when you need them to be more rigid or

644
00:34:24.360 --> 00:34:27.480
<v Speaker 2>more porous. You might simply expose them to a specific

645
00:34:27.559 --> 00:34:30.519
<v Speaker 2>chemical trigger or frequency of light that uploads a new

646
00:34:30.519 --> 00:34:34.400
<v Speaker 2>sequence of structural instructions. But if physical matter is essentially

647
00:34:34.519 --> 00:34:37.719
<v Speaker 2>running on code, and that code can be updated remotely

648
00:34:37.760 --> 00:34:42.880
<v Speaker 2>through environmental triggers, it introduces a terrifying new vulnerability in

649
00:34:42.920 --> 00:34:45.840
<v Speaker 2>the digital world. Where there is code, there are exploits.

650
00:34:46.400 --> 00:34:49.079
<v Speaker 2>If the structural integrity of your car's breaks, where the

651
00:34:49.079 --> 00:34:51.920
<v Speaker 2>thermal insulation of your home relies on a highly specific

652
00:34:52.000 --> 00:34:55.920
<v Speaker 2>cascade of molecular logic gates, could a malicious actor introduce

653
00:34:55.960 --> 00:35:00.239
<v Speaker 2>a synthetic chemical virus. Could someone intentionally release a custom

654
00:35:00.239 --> 00:35:04.119
<v Speaker 2>design molecular sequence that acts as a false input, perfectly

655
00:35:04.159 --> 00:35:07.880
<v Speaker 2>fitting the binding sites of your programmable materials and instructing

656
00:35:07.960 --> 00:35:12.000
<v Speaker 2>them to subtly depolymerize or change their physical state. We've

657
00:35:12.000 --> 00:35:14.960
<v Speaker 2>spent thirty years learning how to defend our digital software

658
00:35:15.000 --> 00:35:18.320
<v Speaker 2>from hackers. As the fabric of physical reality itself waits

659
00:35:18.320 --> 00:35:20.599
<v Speaker 2>for its next software update. We may soon have to

660
00:35:20.599 --> 00:35:22.679
<v Speaker 2>figure out how to put a firewall around the very

661
00:35:22.719 --> 00:35:23.920
<v Speaker 2>atoms we live inside.