1
00:00:15,356 --> 00:00:21,956
Speaker 1: Pushkin. If I close my eyes and picture a robot,

2
00:00:22,476 --> 00:00:28,316
Speaker 1: I see basically a metal guy arms, legs ahead, maybe

3
00:00:28,356 --> 00:00:31,836
Speaker 1: a couple of lights for eyes. And there's a reason

4
00:00:31,876 --> 00:00:34,556
Speaker 1: I see this. This is the way people have been

5
00:00:34,636 --> 00:00:37,796
Speaker 1: dreaming of robots for something like a hundred years now,

6
00:00:38,596 --> 00:00:41,756
Speaker 1: and today people are doing more than dreaming. They're spending

7
00:00:41,836 --> 00:00:46,356
Speaker 1: billions of dollars to build robots like this, to build humanoids,

8
00:00:46,396 --> 00:00:50,516
Speaker 1: to build robots that look like people. But it's really

9
00:00:50,556 --> 00:00:52,996
Speaker 1: hard to build a robot that looks like a guy.

10
00:00:53,556 --> 00:00:56,716
Speaker 1: The physics are complicated. You need a lot of moving parts,

11
00:00:57,396 --> 00:01:01,076
Speaker 1: which is why despite the amazing demos we keep seeing

12
00:01:01,076 --> 00:01:03,676
Speaker 1: on social media, and despite all the smart people and

13
00:01:03,716 --> 00:01:07,196
Speaker 1: all the money trying to make robots happen, almost nobody

14
00:01:07,276 --> 00:01:11,236
Speaker 1: has a humanoid robot at their house. Maybe another approach

15
00:01:11,276 --> 00:01:14,356
Speaker 1: would make more sense. Maybe instead of starting with the

16
00:01:14,436 --> 00:01:17,076
Speaker 1: human body and saying how can we make a robot

17
00:01:17,116 --> 00:01:20,396
Speaker 1: that looks like this, we should ask how can we

18
00:01:20,436 --> 00:01:24,476
Speaker 1: build an affordable robot to help people solve real problems

19
00:01:24,676 --> 00:01:40,036
Speaker 1: at home? Right now, I'm Jacob Goldstein, this is what's

20
00:01:40,036 --> 00:01:43,676
Speaker 1: your problem? And my guest today is Aaron Edzinger. Aaron

21
00:01:43,756 --> 00:01:46,796
Speaker 1: has been building robots for decades. A while back, he

22
00:01:46,876 --> 00:01:49,596
Speaker 1: sold a company to Google, then he became the head

23
00:01:49,596 --> 00:01:52,956
Speaker 1: of robotics at Google, and then he left to start

24
00:01:52,996 --> 00:01:56,396
Speaker 1: a company called Hello Robot, where he is now the CEO.

25
00:01:57,436 --> 00:02:00,236
Speaker 1: Aaron's problem is this, how do you build a robot

26
00:02:00,476 --> 00:02:04,076
Speaker 1: that is useful and affordable for at least some people

27
00:02:04,236 --> 00:02:07,916
Speaker 1: to buy and use at home. Hello Robot hasn't really

28
00:02:07,956 --> 00:02:11,276
Speaker 1: solved that problem yet, but they're working on it. They

29
00:02:11,276 --> 00:02:15,036
Speaker 1: have a robot called Stretch three. They've sold hundreds of them,

30
00:02:15,676 --> 00:02:19,436
Speaker 1: and the robots are proving truly useful to some people

31
00:02:19,596 --> 00:02:24,356
Speaker 1: in very difficult circumstances. Also, the robots don't look anything

32
00:02:24,556 --> 00:02:28,836
Speaker 1: like metal people. They're not humanoids. In our conversation, Aaron

33
00:02:28,836 --> 00:02:32,076
Speaker 1: and I talk about why home robots have barely progressed

34
00:02:32,116 --> 00:02:34,796
Speaker 1: since the roomba, what it actually takes to get a

35
00:02:34,876 --> 00:02:38,476
Speaker 1: robot to feed someone, and why the physical world is

36
00:02:38,556 --> 00:02:42,396
Speaker 1: so much harder for AI than language. One of the

37
00:02:42,516 --> 00:02:45,396
Speaker 1: interesting things to me about your approach about Stretch three,

38
00:02:45,516 --> 00:02:48,556
Speaker 1: the robot that you have in the world now is, frankly,

39
00:02:48,596 --> 00:02:51,636
Speaker 1: the things that it's not right. It's not autonomous, it's

40
00:02:51,676 --> 00:02:55,556
Speaker 1: not a humanoid, it's a robot. Arm that's operated by

41
00:02:55,596 --> 00:02:58,356
Speaker 1: remote control. And you know you've been in the field

42
00:02:58,396 --> 00:03:00,716
Speaker 1: for a long time. You were the director of robotics

43
00:03:00,756 --> 00:03:05,396
Speaker 1: at Google. So why did you leave Google to build this?

44
00:03:06,436 --> 00:03:10,556
Speaker 2: Oh it's it yeah, complicated answer, it's a good question.

45
00:03:11,916 --> 00:03:15,676
Speaker 2: In the end. You know, with envisioned robots being something

46
00:03:15,716 --> 00:03:20,316
Speaker 2: different than what they were becoming at that time, one

47
00:03:20,396 --> 00:03:25,276
Speaker 2: part of that is just about simplicity, right, So you know,

48
00:03:25,396 --> 00:03:28,516
Speaker 2: more and more the robots that we're building at Google

49
00:03:29,156 --> 00:03:32,716
Speaker 2: were becoming more complex, and you see that with the

50
00:03:32,796 --> 00:03:36,236
Speaker 2: humanoids today. There's this push towards kind of human level

51
00:03:36,476 --> 00:03:40,276
Speaker 2: dexterity capability. And our premise was, you know what, you

52
00:03:40,276 --> 00:03:43,836
Speaker 2: could do a lot today if you push on simplifying

53
00:03:44,436 --> 00:03:48,076
Speaker 2: the design, the approach. So the whole premise with Heller

54
00:03:48,156 --> 00:03:52,196
Speaker 2: robot is how do we practically deliver a value with

55
00:03:52,236 --> 00:03:55,316
Speaker 2: a robot to help people? And forget about should it

56
00:03:55,356 --> 00:03:58,116
Speaker 2: look like a human Let's really focus on just that

57
00:03:58,196 --> 00:04:01,596
Speaker 2: core question. And from that came to Toutch three, which

58
00:04:01,596 --> 00:04:05,596
Speaker 2: is a very minimalist robot. We like to think of

59
00:04:05,636 --> 00:04:08,116
Speaker 2: it almost like robotic cubism, right, So if you take

60
00:04:08,316 --> 00:04:11,436
Speaker 2: a human form and decompose it, put it back together,

61
00:04:11,876 --> 00:04:15,356
Speaker 2: in some other way. If you squint, it's recognizable as

62
00:04:15,356 --> 00:04:18,036
Speaker 2: a human. Maybe it can do eighty percent of what

63
00:04:18,116 --> 00:04:20,956
Speaker 2: a human or a humanoid could do. But it's much

64
00:04:20,956 --> 00:04:25,156
Speaker 2: more practical, deployable from a business perspective, but also just

65
00:04:25,196 --> 00:04:29,116
Speaker 2: from a usability and also autonomy. So the robot is autonomous.

66
00:04:29,436 --> 00:04:33,996
Speaker 2: We don't highlight a lot of that, but the autonomy

67
00:04:33,996 --> 00:04:37,076
Speaker 2: gets easier too when the robot gets simpler. So that's

68
00:04:37,116 --> 00:04:39,956
Speaker 2: the basic premise of starting the company.

69
00:04:40,236 --> 00:04:42,556
Speaker 1: Yeah, I mean it's interesting as you were saying that,

70
00:04:42,636 --> 00:04:45,116
Speaker 1: I was thinking of sort of the classic software thing

71
00:04:45,356 --> 00:04:48,996
Speaker 1: is the minimum viable product, right, Like, what's the simplest

72
00:04:48,996 --> 00:04:50,316
Speaker 1: thing we can get out into the world and then

73
00:04:50,436 --> 00:04:54,156
Speaker 1: learn what's going on with it, iterate, etc. It's like

74
00:04:54,196 --> 00:04:58,476
Speaker 1: you're building the minimum viable robot exactly exactly, So you

75
00:04:59,556 --> 00:05:03,356
Speaker 1: have built a robot and it is in the world.

76
00:05:03,676 --> 00:05:06,516
Speaker 1: The current version is called Stretch three. Tell me about

77
00:05:06,516 --> 00:05:08,356
Speaker 1: Stretch three, so.

78
00:05:08,356 --> 00:05:12,756
Speaker 2: S Touch three a third generation. It is fundamentally think

79
00:05:12,756 --> 00:05:16,596
Speaker 2: of it like a roomba with a big mast on it, right.

80
00:05:17,316 --> 00:05:20,676
Speaker 2: Comes from that perspective, right, and then you add to

81
00:05:20,756 --> 00:05:24,476
Speaker 2: that mast a telescoping arm, you know, telescoping the way

82
00:05:24,516 --> 00:05:27,356
Speaker 2: a car Antenta might telescope a turned sideways and you

83
00:05:27,396 --> 00:05:31,076
Speaker 2: put on the end of that telescoping arm a hand, so.

84
00:05:31,076 --> 00:05:33,316
Speaker 1: It's basically an arm. There's a room b with a

85
00:05:33,516 --> 00:05:36,756
Speaker 1: vertical pole going up what three feet or something about

86
00:05:36,756 --> 00:05:39,956
Speaker 1: four feet yep, four feet, and then basically a robot

87
00:05:40,036 --> 00:05:41,596
Speaker 1: arm what we think of as a robot arm on

88
00:05:41,596 --> 00:05:43,636
Speaker 1: top of that pole. Exactly a room but that can

89
00:05:43,716 --> 00:05:45,196
Speaker 1: drive a robot arm around.

90
00:05:44,916 --> 00:05:47,476
Speaker 2: Your house, drive around, and it can reach up to

91
00:05:47,516 --> 00:05:51,036
Speaker 2: the countertop underneath the couch. That type of thing. Very

92
00:05:51,156 --> 00:05:54,916
Speaker 2: very simple design, and what we saw early on is

93
00:05:54,996 --> 00:05:57,876
Speaker 2: just by remote controlling it, you can do quite a

94
00:05:57,916 --> 00:05:59,956
Speaker 2: few things around a house that you care about. And

95
00:05:59,996 --> 00:06:03,436
Speaker 2: so our first launch video showed it picking up toys

96
00:06:03,556 --> 00:06:07,516
Speaker 2: tele operated, but picking up toys, wiping down countertops, putting

97
00:06:07,556 --> 00:06:10,796
Speaker 2: laundry in a laundry machine with a very simple device.

98
00:06:11,356 --> 00:06:13,356
Speaker 2: You know, I just come from Google, where we're building

99
00:06:13,356 --> 00:06:16,316
Speaker 2: a very complex device and to do the same type

100
00:06:16,316 --> 00:06:18,676
Speaker 2: of thing. And I just thought, you know, if these

101
00:06:18,756 --> 00:06:20,636
Speaker 2: robots are ever really going to get out to the

102
00:06:20,636 --> 00:06:23,596
Speaker 2: world as real products, it's going to be and look

103
00:06:23,716 --> 00:06:24,796
Speaker 2: and feel more like this.

104
00:06:25,436 --> 00:06:26,396
Speaker 1: How much does it cost?

105
00:06:27,396 --> 00:06:31,316
Speaker 2: Right now? It's twenty about twenty five thousand dollars. And

106
00:06:31,356 --> 00:06:34,916
Speaker 2: that's that's because we run the company off of sales.

107
00:06:34,916 --> 00:06:37,836
Speaker 2: We're not it's trying to you know, it's low volume.

108
00:06:37,996 --> 00:06:40,236
Speaker 1: So does that mean your your marginal cost is about

109
00:06:40,276 --> 00:06:42,316
Speaker 1: twenty five thousand dollars, you're selling it for about what

110
00:06:42,396 --> 00:06:42,916
Speaker 1: it costs.

111
00:06:43,076 --> 00:06:45,436
Speaker 2: No, so the opposite. So what I'm saying is it's

112
00:06:45,556 --> 00:06:49,316
Speaker 2: actually can be very low cost, but because we're funding

113
00:06:49,516 --> 00:06:52,076
Speaker 2: R and D and engineering and production.

114
00:06:51,796 --> 00:06:52,956
Speaker 1: You have a margin on it.

115
00:06:53,636 --> 00:06:57,156
Speaker 2: We have a real, operational, sustainable business selling robots today.

116
00:06:57,276 --> 00:06:58,516
Speaker 1: So you're a real company.

117
00:06:58,636 --> 00:07:02,036
Speaker 2: We are a real company selling robots to people. Yes,

118
00:07:02,236 --> 00:07:04,916
Speaker 2: And that was part of the strategy from the get gos.

119
00:07:05,796 --> 00:07:08,196
Speaker 2: You know, I've been to robotics for twenty five years

120
00:07:08,236 --> 00:07:13,596
Speaker 2: now and seen a lot of companies over promise what's

121
00:07:13,676 --> 00:07:19,036
Speaker 2: possible with the technology, and then you know, their capital,

122
00:07:19,116 --> 00:07:22,796
Speaker 2: their investors getting patient, and they end up closing up shop.

123
00:07:22,916 --> 00:07:24,636
Speaker 2: And we see this time and time again.

124
00:07:24,796 --> 00:07:27,196
Speaker 1: We're selling to Google and a happier version.

125
00:07:27,276 --> 00:07:29,876
Speaker 2: Yeah, well you could do that too, which yeah, but

126
00:07:30,156 --> 00:07:32,916
Speaker 2: you know, really we wanted to have a viable business,

127
00:07:33,196 --> 00:07:36,716
Speaker 2: built a product that people liked and loved, and then

128
00:07:37,156 --> 00:07:40,756
Speaker 2: from there then you have permission to take the time

129
00:07:40,796 --> 00:07:42,596
Speaker 2: that it needs to really to do it correctly.

130
00:07:43,796 --> 00:07:44,716
Speaker 1: How many have you sold?

131
00:07:45,516 --> 00:07:48,916
Speaker 2: Yes, we have about three hundred robots in the world

132
00:07:49,476 --> 00:07:53,796
Speaker 2: across twenty three countries, and about forty five to different institutions,

133
00:07:54,036 --> 00:07:58,436
Speaker 2: from top tier research labs to small startups, to people

134
00:07:58,476 --> 00:08:01,876
Speaker 2: looking at vertical farming, to people looking at healthcare, you know,

135
00:08:01,876 --> 00:08:03,436
Speaker 2: a whole range of things.

136
00:08:04,156 --> 00:08:09,476
Speaker 1: I saw a video about a man named Henry Vince

137
00:08:09,476 --> 00:08:13,316
Speaker 1: who uses one of your robots, and I found it

138
00:08:13,356 --> 00:08:18,116
Speaker 1: quite compelling. Yeah, tell me about Henry Evans, right.

139
00:08:18,596 --> 00:08:22,516
Speaker 2: Henry and his care partner and wife, Jane. You know,

140
00:08:22,556 --> 00:08:26,756
Speaker 2: they're in California, not too far from us. They've collaborated

141
00:08:26,836 --> 00:08:30,396
Speaker 2: with my co founder, Charlie camp for at least I

142
00:08:30,476 --> 00:08:33,636
Speaker 2: think fifteen years now, so it's been a long relationship.

143
00:08:33,676 --> 00:08:38,676
Speaker 2: Henry is quadriplegic from a stroke. Jane is the primary caregiver.

144
00:08:39,836 --> 00:08:42,596
Speaker 2: But Henry is, you know, an early adopter. He's really

145
00:08:42,676 --> 00:08:46,076
Speaker 2: excited to find ways that technology can give him agency,

146
00:08:47,116 --> 00:08:49,956
Speaker 2: not just that, but make life easier on chain and

147
00:08:51,596 --> 00:08:54,396
Speaker 2: so you know, he's been looking at robotics as a

148
00:08:54,436 --> 00:08:57,436
Speaker 2: solution partial solution for that for quite some time. We've

149
00:08:57,476 --> 00:09:01,396
Speaker 2: collaborated with him, and really when we found it, Hello robot,

150
00:09:01,516 --> 00:09:05,356
Speaker 2: you know, he was one of the key well i'd

151
00:09:05,356 --> 00:09:08,516
Speaker 2: say inspirations and users end users that we had in mind.

152
00:09:09,316 --> 00:09:11,836
Speaker 2: So through some some work that was funded by the NIH,

153
00:09:11,876 --> 00:09:13,996
Speaker 2: we've spent it last three years working with him and

154
00:09:14,036 --> 00:09:18,996
Speaker 2: some other people in similar situations looking at how Stretch

155
00:09:19,476 --> 00:09:22,396
Speaker 2: can can benefit him. And so, you know, Henry isn't

156
00:09:22,396 --> 00:09:26,436
Speaker 2: able to communicate, to talk except through I gays, so

157
00:09:26,476 --> 00:09:29,476
Speaker 2: he uses I gaze on a letter board to spill

158
00:09:29,476 --> 00:09:32,716
Speaker 2: out words and then he can also use I gays

159
00:09:32,796 --> 00:09:36,396
Speaker 2: to control a computer mouse, and through that he's able

160
00:09:36,396 --> 00:09:39,876
Speaker 2: to control Stretch to do simple tasks, things that might

161
00:09:39,956 --> 00:09:43,516
Speaker 2: seem you know, trivial, are you not so exciting to

162
00:09:43,596 --> 00:09:46,156
Speaker 2: other people, but for him, it's it's it's it's really

163
00:09:46,196 --> 00:09:49,876
Speaker 2: life changing. And so a really really nice story with

164
00:09:49,876 --> 00:09:53,996
Speaker 2: with Henry is, you know, we came in thinking like, well,

165
00:09:54,316 --> 00:09:58,516
Speaker 2: you know, Stretch will do very practical tasks very quickly.

166
00:09:58,596 --> 00:10:01,076
Speaker 2: What we learned is there's a whole range of things

167
00:10:01,076 --> 00:10:04,996
Speaker 2: that are much more social, emotional about connection that we

168
00:10:05,076 --> 00:10:08,316
Speaker 2: just didn't anticipate with the device like this. And so

169
00:10:08,396 --> 00:10:11,796
Speaker 2: one example is his granddaughter at the time was probably

170
00:10:11,836 --> 00:10:16,196
Speaker 2: about three or four. She was a little bit scared

171
00:10:16,236 --> 00:10:19,956
Speaker 2: of him because you know, he's immobile, he's in a

172
00:10:20,076 --> 00:10:23,796
Speaker 2: powerchair in bed, can't talk, and so they didn't really

173
00:10:23,796 --> 00:10:27,196
Speaker 2: have a relationship. And what we found is through Stretch,

174
00:10:27,276 --> 00:10:29,836
Speaker 2: because he could control Stretch, he could start to play

175
00:10:29,876 --> 00:10:32,236
Speaker 2: with her, interact with her, and in the physical world.

176
00:10:33,076 --> 00:10:36,076
Speaker 2: She would decorate stretch, put stickers on it. They would

177
00:10:36,116 --> 00:10:39,116
Speaker 2: have a tablet on it, so you know, his face

178
00:10:39,156 --> 00:10:44,076
Speaker 2: would appear, and soon they started to play. So they'd

179
00:10:44,076 --> 00:10:49,516
Speaker 2: played basketball, pick up, a little nerve, basketball, play different games,

180
00:10:49,876 --> 00:10:52,276
Speaker 2: and you know, she started to really kind of see

181
00:10:52,356 --> 00:10:55,236
Speaker 2: him in a different light and former relationship with him,

182
00:10:55,836 --> 00:10:58,436
Speaker 2: and so now she calls him Papa Wheeli because he's

183
00:10:58,436 --> 00:11:01,716
Speaker 2: in his chair and and she, you know, can be

184
00:11:01,756 --> 00:11:05,036
Speaker 2: excited to see him. So it really bridge that gap

185
00:11:05,076 --> 00:11:08,036
Speaker 2: between the two of them, and and the type of

186
00:11:08,036 --> 00:11:12,796
Speaker 2: thing that with robot that there's so much potential and

187
00:11:12,956 --> 00:11:17,276
Speaker 2: I think we really don't quite know how far we

188
00:11:17,316 --> 00:11:19,916
Speaker 2: can go on the sort of the social side of it.

189
00:11:19,996 --> 00:11:23,556
Speaker 2: You know, really this can be this empathetic technology that

190
00:11:23,716 --> 00:11:25,956
Speaker 2: brings so much to people, and that's that's kind of

191
00:11:25,956 --> 00:11:27,196
Speaker 2: what we're excited to explore.

192
00:11:29,156 --> 00:11:31,956
Speaker 1: As you were talking about the robot playing with a

193
00:11:32,276 --> 00:11:35,116
Speaker 1: you know, if you're a young child, the question of

194
00:11:35,156 --> 00:11:38,916
Speaker 1: safety crossed my mind, Like what do you have to

195
00:11:38,916 --> 00:11:41,076
Speaker 1: worry about in that setting? How do you deal with it?

196
00:11:41,516 --> 00:11:41,596
Speaker 2: Like?

197
00:11:41,636 --> 00:11:44,476
Speaker 1: How much is the robot weigh? Yeah, it's a great question.

198
00:11:44,516 --> 00:11:47,116
Speaker 2: So this robot weighs about fifty pounds, a little over

199
00:11:47,196 --> 00:11:51,196
Speaker 2: fifty fifty pounds most that weighs down in the base, right,

200
00:11:52,796 --> 00:11:55,956
Speaker 2: And it's really it's an important question because there's so

201
00:11:56,036 --> 00:12:00,716
Speaker 2: much happening in robotics right now where we're seeing humanoids

202
00:12:00,996 --> 00:12:05,236
Speaker 2: in the home doing dishes right, and these are scary machines.

203
00:12:05,356 --> 00:12:07,836
Speaker 2: And really the fundamental thing you have to think about

204
00:12:07,996 --> 00:12:12,636
Speaker 2: is just that physics of it. Right, when you have

205
00:12:12,756 --> 00:12:16,436
Speaker 2: a humanoid that's balancing, particularly a human scale one, the

206
00:12:16,436 --> 00:12:19,476
Speaker 2: amount of weight that's up high, the potential energy that's

207
00:12:19,516 --> 00:12:22,996
Speaker 2: there is significant, and when things go.

208
00:12:23,036 --> 00:12:26,916
Speaker 1: Wrong, if it falls over, you could smush things. That's fundamentally,

209
00:12:27,196 --> 00:12:28,436
Speaker 1: that's fundamentally it.

210
00:12:28,716 --> 00:12:31,156
Speaker 2: And so just the pure physics physics of that are

211
00:12:31,556 --> 00:12:35,076
Speaker 2: fundamentally you know, we can't cheat it, right, you can't,

212
00:12:35,396 --> 00:12:38,796
Speaker 2: and things things will fail. So really, with from the

213
00:12:38,836 --> 00:12:41,516
Speaker 2: get go for for hello, robot and stretch. We thought

214
00:12:41,556 --> 00:12:46,716
Speaker 2: about safety from just the physical properties of it. How

215
00:12:46,756 --> 00:12:49,596
Speaker 2: do we make it so it's very hard for this

216
00:12:49,676 --> 00:12:52,716
Speaker 2: device to cause harm? So first off, it should be

217
00:12:52,796 --> 00:12:57,076
Speaker 2: light weight, that potential energy, the mass should be down

218
00:12:57,116 --> 00:12:59,316
Speaker 2: low so that if if it falls over, it's not

219
00:12:59,396 --> 00:13:03,796
Speaker 2: gonna you know, smash and break something. And then you

220
00:13:03,836 --> 00:13:06,796
Speaker 2: think about the robot joints themselves.

221
00:13:06,436 --> 00:13:08,316
Speaker 1: And is there is there a pincer We haven't actually

222
00:13:08,356 --> 00:13:10,636
Speaker 1: talked about the sort of yes, gripper is it? Is

223
00:13:10,676 --> 00:13:12,516
Speaker 1: it like a two piece pincer.

224
00:13:12,596 --> 00:13:14,956
Speaker 2: It's a it's basically like a grabber that you buy

225
00:13:14,996 --> 00:13:17,636
Speaker 2: an Amazon, you know, like to pick up trash. Yeah,

226
00:13:17,676 --> 00:13:20,956
Speaker 2: it's one of those that we've attached to motor to essentially.

227
00:13:21,676 --> 00:13:24,996
Speaker 2: But if you think about, you know, the robot kind

228
00:13:24,996 --> 00:13:27,836
Speaker 2: of motors. If if you have to hold your arm

229
00:13:27,836 --> 00:13:31,396
Speaker 2: out against gravity all day long, you know you're gonna

230
00:13:31,436 --> 00:13:33,196
Speaker 2: need a big motor there, right, because that's a lot

231
00:13:33,236 --> 00:13:37,356
Speaker 2: of energy to support that. If you don't, if you

232
00:13:37,396 --> 00:13:39,996
Speaker 2: can design the robot in a minimal way where you're

233
00:13:39,996 --> 00:13:43,396
Speaker 2: not spending all of your time fighting energy or first off,

234
00:13:43,396 --> 00:13:46,476
Speaker 2: you're more energy efficient. Second you have a smaller motor

235
00:13:46,716 --> 00:13:50,276
Speaker 2: which will be less dangerous. And so the arm moves

236
00:13:50,316 --> 00:13:51,636
Speaker 2: in a way, it is designed in a way that

237
00:13:51,716 --> 00:13:53,676
Speaker 2: it's n have to fight gravity, and that just makes

238
00:13:53,676 --> 00:13:58,236
Speaker 2: it inherently safer. Right, And so when you're working around

239
00:13:58,836 --> 00:14:03,516
Speaker 2: older adults, people with disabilities, kids, pets, these things really

240
00:14:03,556 --> 00:14:06,436
Speaker 2: really matter. And so that's that's just the starting point

241
00:14:06,676 --> 00:14:10,596
Speaker 2: from the design. The intrinsic safety has to be there.

242
00:14:10,636 --> 00:14:12,316
Speaker 2: The physics of it has has to be right.

243
00:14:12,956 --> 00:14:17,996
Speaker 1: So people with quadriplegia, people with extremely limited mobility are

244
00:14:18,036 --> 00:14:23,156
Speaker 1: an obvious early use case. What else, like, what else

245
00:14:23,196 --> 00:14:25,036
Speaker 1: are people using this for? What else do you have

246
00:14:25,116 --> 00:14:26,516
Speaker 1: in mind for early use case?

247
00:14:26,636 --> 00:14:30,356
Speaker 2: Yeah, one of the most interesting use cases is in

248
00:14:30,476 --> 00:14:34,636
Speaker 2: the home around caregiving. It's a very challenging one. It

249
00:14:34,756 --> 00:14:36,556
Speaker 2: is further out, and so you know, there's a lot

250
00:14:36,556 --> 00:14:39,956
Speaker 2: of use cases nearer term which I think will be

251
00:14:40,036 --> 00:14:45,156
Speaker 2: in service, you know, back of retail logistics. Uh, And

252
00:14:45,156 --> 00:14:47,076
Speaker 2: we're seeing you know, data centers, a lot of stuff

253
00:14:47,116 --> 00:14:50,836
Speaker 2: like that that we are seeing interest in and people exploring.

254
00:14:50,756 --> 00:14:53,796
Speaker 1: Like picking and pecking. I mean there are specialized robots

255
00:14:53,836 --> 00:14:54,716
Speaker 1: for those things.

256
00:14:54,796 --> 00:14:58,956
Speaker 2: Right. Well, yeah, there's some places where there's not yet, right, So,

257
00:14:59,076 --> 00:15:02,636
Speaker 2: for example, micro fulfillment little fulfillment, you know, like even

258
00:15:02,716 --> 00:15:04,996
Speaker 2: like a you know, stocking a seven to eleven. There's

259
00:15:05,036 --> 00:15:08,076
Speaker 2: lots of things where a robot like this could play

260
00:15:08,076 --> 00:15:08,356
Speaker 2: a role.

261
00:15:09,396 --> 00:15:12,956
Speaker 1: Options And just to bit clear, my understanding is that

262
00:15:13,036 --> 00:15:17,396
Speaker 1: your robot is mainly not autonomous, that it's mainly remote

263
00:15:17,396 --> 00:15:19,596
Speaker 1: controlled tteley operated. Is that right? And what does that

264
00:15:19,676 --> 00:15:20,836
Speaker 1: mean for that use case?

265
00:15:21,076 --> 00:15:25,196
Speaker 2: Right? It can be autonomous. There are autonomous it's completely

266
00:15:25,196 --> 00:15:28,676
Speaker 2: open source as a whole ecosystem of open source software

267
00:15:28,676 --> 00:15:33,636
Speaker 2: people have built for it, including autonomy, and the generation

268
00:15:33,716 --> 00:15:39,996
Speaker 2: of robot we're working on now will really be autonomy powered, right,

269
00:15:40,036 --> 00:15:42,636
Speaker 2: so in terms of the sensing of the software. But

270
00:15:42,876 --> 00:15:46,196
Speaker 2: you know, back to that sort of really minimal approach. Yeah,

271
00:15:46,316 --> 00:15:48,276
Speaker 2: we're trying to say what can we do today that

272
00:15:48,316 --> 00:15:53,996
Speaker 2: has value? And for someone like Henry Evans, he controls

273
00:15:53,996 --> 00:15:56,996
Speaker 2: the robot, it has autonomy mixed in there, so he's

274
00:15:56,996 --> 00:16:01,236
Speaker 2: not controlling every motion necessarily, and so what we see

275
00:16:01,316 --> 00:16:06,676
Speaker 2: is as a steady progression from direct teleoperation to mixed

276
00:16:06,676 --> 00:16:10,876
Speaker 2: autonomy to full autonomy, but doing that really deployed, right,

277
00:16:10,916 --> 00:16:13,396
Speaker 2: So this is not something in a research lab. It's

278
00:16:13,476 --> 00:16:17,996
Speaker 2: actually in the product. And you know, we also see

279
00:16:18,156 --> 00:16:22,036
Speaker 2: a path from people with really severe mobility impairments like Henry.

280
00:16:23,116 --> 00:16:27,636
Speaker 2: They're interesting because they have a high motivation to use

281
00:16:27,676 --> 00:16:30,156
Speaker 2: the device, and it's been the time to control it, right,

282
00:16:30,156 --> 00:16:34,156
Speaker 2: They've got time and motivation. And then you can layer

283
00:16:34,196 --> 00:16:37,156
Speaker 2: in the economy over time, and as you do that,

284
00:16:37,316 --> 00:16:40,076
Speaker 2: now you can broaden the set of people who can

285
00:16:40,316 --> 00:16:44,956
Speaker 2: who benefit from it. I think from people like Henry.

286
00:16:45,996 --> 00:16:49,396
Speaker 2: The next step in terms of in home care is

287
00:16:49,796 --> 00:16:54,596
Speaker 2: a remote operation by a family member a care provider.

288
00:16:55,436 --> 00:16:57,716
Speaker 2: You know, for example, I'm in New Orleans right now,

289
00:16:57,996 --> 00:17:00,716
Speaker 2: my mother is back in California, and you know, I'd

290
00:17:00,716 --> 00:17:03,396
Speaker 2: love to hop on and check in, and you know,

291
00:17:03,836 --> 00:17:06,596
Speaker 2: of course you can zoom, but sometimes you need to

292
00:17:06,596 --> 00:17:08,716
Speaker 2: go into the other room, open the cabinet door, and

293
00:17:09,276 --> 00:17:12,316
Speaker 2: you know, find the medication or something like that. And

294
00:17:12,596 --> 00:17:15,396
Speaker 2: so again there's a human in the loop, but they're

295
00:17:15,396 --> 00:17:17,876
Speaker 2: not going to be controlling every motion. They're going to

296
00:17:17,916 --> 00:17:21,516
Speaker 2: be clicking on a map and saying go to the bedroom, open,

297
00:17:21,796 --> 00:17:24,836
Speaker 2: open the drawer, that type of thing. Yeah, And then

298
00:17:24,916 --> 00:17:27,836
Speaker 2: from there, of course, you can bring in more autonomy,

299
00:17:27,876 --> 00:17:32,236
Speaker 2: and suddenly you have in home autonomous care and eventually

300
00:17:32,276 --> 00:17:34,676
Speaker 2: in home robots that are doing the laundry and dishes

301
00:17:34,716 --> 00:17:38,116
Speaker 2: and everything that you know, people are excited about. But

302
00:17:38,156 --> 00:17:41,356
Speaker 2: for us, again, we're taking this very pragmatic approach about

303
00:17:41,356 --> 00:17:43,156
Speaker 2: how do we stair step, how do we kind of

304
00:17:43,196 --> 00:17:45,196
Speaker 2: climb our way up to that future?

305
00:17:47,716 --> 00:17:50,676
Speaker 1: Stairstep is an interesting choice of words for a robot

306
00:17:50,716 --> 00:17:56,716
Speaker 1: on wheels. But you know, one of the things people

307
00:17:56,756 --> 00:18:00,196
Speaker 1: always talk about in the context of robots in the

308
00:18:00,236 --> 00:18:04,036
Speaker 1: last few years is getting the data to train the

309
00:18:04,076 --> 00:18:07,996
Speaker 1: model for autonomy, right, And the obvious analog is large

310
00:18:08,036 --> 00:18:12,196
Speaker 1: language models got so good because the Internet is full

311
00:18:12,236 --> 00:18:15,436
Speaker 1: of text, right, and there is no Internet of the

312
00:18:15,476 --> 00:18:17,996
Speaker 1: physical world. There is no giant data set that you

313
00:18:17,996 --> 00:18:21,956
Speaker 1: can just train a model on so that robots can

314
00:18:22,036 --> 00:18:24,756
Speaker 1: understand how physics works in the real world. Right, how

315
00:18:24,756 --> 00:18:25,676
Speaker 1: are you dealing with that?

316
00:18:28,316 --> 00:18:32,076
Speaker 2: You know, I would say, in terms of AI physical

317
00:18:32,116 --> 00:18:36,996
Speaker 2: AI as it's called around robotics, we're taking, you know,

318
00:18:37,036 --> 00:18:40,756
Speaker 2: an approach that's really about taking things that are mature

319
00:18:40,916 --> 00:18:45,596
Speaker 2: and integrating them into our autonomy, versus what's called end

320
00:18:45,636 --> 00:18:47,956
Speaker 2: to end learning, where you know, you might collect a

321
00:18:48,076 --> 00:18:51,516
Speaker 2: huge data set, right, you might train the robot tele

322
00:18:51,516 --> 00:18:54,396
Speaker 2: operated to open the dishwasher a thousand times, collect that

323
00:18:54,516 --> 00:18:58,716
Speaker 2: data and train a deep learned model to do that task,

324
00:19:00,116 --> 00:19:02,316
Speaker 2: and you see very exciting results from that, and I

325
00:19:02,316 --> 00:19:06,476
Speaker 2: do think that will continue to progress, but in terms

326
00:19:06,516 --> 00:19:09,796
Speaker 2: of being ready to be deployed, it'll be take some time.

327
00:19:10,756 --> 00:19:15,596
Speaker 2: And the amount of data that's required beyond just a dishwasher, Well,

328
00:19:15,636 --> 00:19:18,036
Speaker 2: now it's a different dishwasher. Now it's a dishwashering different

329
00:19:18,076 --> 00:19:20,756
Speaker 2: lighting conditions. Oh there's something in front of it. That

330
00:19:21,436 --> 00:19:23,876
Speaker 2: complexity gets so large so quickly.

331
00:19:23,956 --> 00:19:26,916
Speaker 1: And it doesn't generalize, like there are not models where

332
00:19:26,956 --> 00:19:29,116
Speaker 1: you can teach it one dishwasher and it can make

333
00:19:29,196 --> 00:19:30,836
Speaker 1: inferences about other dishwashers.

334
00:19:31,276 --> 00:19:34,396
Speaker 2: There's inklings of that happening, right, but we're not there yet,

335
00:19:34,396 --> 00:19:38,476
Speaker 2: and resourcers are still really exploring what's possible. You know,

336
00:19:38,556 --> 00:19:41,396
Speaker 2: some are saying it's going to learn from YouTube, some

337
00:19:41,476 --> 00:19:43,956
Speaker 2: are saying it's going to be all on simulation. Some

338
00:19:44,036 --> 00:19:49,316
Speaker 2: are our hiring teams of operators to collect that data. Right.

339
00:19:49,956 --> 00:19:52,956
Speaker 2: Our view is, you know that that world is progressing

340
00:19:52,996 --> 00:19:58,476
Speaker 2: quickly and it'll sort itself out soon. You know, we

341
00:19:58,516 --> 00:20:01,516
Speaker 2: will be able to train the robot to open that dishwasher.

342
00:20:02,236 --> 00:20:04,436
Speaker 2: I think what's more interesting is how do you get

343
00:20:04,436 --> 00:20:07,356
Speaker 2: it to do you know the five or six important

344
00:20:07,396 --> 00:20:09,996
Speaker 2: tasks you care about and actually do it in a

345
00:20:10,076 --> 00:20:12,516
Speaker 2: robot that people feel is trusted in the home, it's

346
00:20:12,556 --> 00:20:15,036
Speaker 2: safe in the home, is affordable, you know, all the

347
00:20:15,076 --> 00:20:16,636
Speaker 2: things that actually to have a business.

348
00:20:17,476 --> 00:20:21,956
Speaker 1: Yeah, so you're basically it's again the like the physical

349
00:20:21,956 --> 00:20:24,396
Speaker 1: intelligence is the bright shiny object that you are not

350
00:20:24,556 --> 00:20:27,916
Speaker 1: chasing so that you can build a robot that works exactly.

351
00:20:27,996 --> 00:20:31,636
Speaker 2: And and you know, think about uh, you know, you know,

352
00:20:31,996 --> 00:20:34,756
Speaker 2: is Apple building a big AI team? Well, no, it's

353
00:20:34,796 --> 00:20:37,836
Speaker 2: partnering with the people who are doing the AI. And

354
00:20:37,996 --> 00:20:40,476
Speaker 2: I think you know, for us, we're more on that

355
00:20:40,516 --> 00:20:41,196
Speaker 2: side of the fence.

356
00:20:42,116 --> 00:20:44,916
Speaker 1: Are some of those AI gains that you are sort

357
00:20:44,916 --> 00:20:48,756
Speaker 1: of piggybacking on coming from autonomous vehicles? I mean when

358
00:20:48,756 --> 00:20:51,156
Speaker 1: you talk about saying go to the other room and

359
00:20:51,156 --> 00:20:53,796
Speaker 1: have it go there, I think of that, That's where

360
00:20:53,796 --> 00:20:55,956
Speaker 1: my mind goes, right that those are the robots out

361
00:20:55,996 --> 00:20:58,716
Speaker 1: in the world right now driving around San Francisco.

362
00:20:58,996 --> 00:21:02,516
Speaker 2: Yeah, I mean it's slightly different in terms of the

363
00:21:02,556 --> 00:21:06,356
Speaker 2: requirements the sensors, but it's certainly certainly related. I mean,

364
00:21:06,396 --> 00:21:10,596
Speaker 2: we've all benefited from he's wrapping gains and sensing and

365
00:21:10,636 --> 00:21:14,436
Speaker 2: compute and and you know even ll MS is they're

366
00:21:14,476 --> 00:21:18,836
Speaker 2: being used to help their robots plan and think and replan. Right.

367
00:21:19,036 --> 00:21:22,876
Speaker 2: So you know, when I was started in robotics twenty

368
00:21:22,996 --> 00:21:26,516
Speaker 2: twenty years ago, I couldn't imagine being able to say

369
00:21:26,516 --> 00:21:28,836
Speaker 2: to the robot, hey, go to the other room and

370
00:21:28,916 --> 00:21:32,276
Speaker 2: find the blue mug that's next to the microwave. Right,

371
00:21:32,356 --> 00:21:34,796
Speaker 2: Like the context of that was just like, how are

372
00:21:34,836 --> 00:21:36,596
Speaker 2: we going to solve that? Are we going to write

373
00:21:36,636 --> 00:21:38,716
Speaker 2: every rule into the robot and that's.

374
00:21:38,556 --> 00:21:41,636
Speaker 1: Like commodified like those are That's that's the easy part

375
00:21:41,716 --> 00:21:43,036
Speaker 1: now at some.

376
00:21:43,116 --> 00:21:46,956
Speaker 2: Level, yes, I mean certainly there's more to be done.

377
00:21:47,036 --> 00:21:49,396
Speaker 2: I'd say, you know, the real challenges today are going

378
00:21:49,396 --> 00:21:54,556
Speaker 2: to be around desk to manipulation and being able to

379
00:21:54,596 --> 00:21:58,796
Speaker 2: do things that are really in dynamic, you know, kind

380
00:21:58,836 --> 00:22:00,076
Speaker 2: of home type environments.

381
00:22:00,316 --> 00:22:02,876
Speaker 1: So when you when you talk about you know, dexter

382
00:22:03,036 --> 00:22:05,116
Speaker 1: sort of fine motor skill type things, what are you

383
00:22:05,196 --> 00:22:07,076
Speaker 1: thinking of? What do you want the robot to be

384
00:22:07,076 --> 00:22:09,596
Speaker 1: able to do in the relatively near term that it

385
00:22:09,636 --> 00:22:10,276
Speaker 1: can't do now?

386
00:22:11,596 --> 00:22:13,996
Speaker 2: I think just broaden the range of tasks. Right. So

387
00:22:14,236 --> 00:22:18,876
Speaker 2: one thing that's really exciting to this user group is feeding, right,

388
00:22:18,916 --> 00:22:21,356
Speaker 2: I think feeding. Feeding is one of the things where

389
00:22:22,116 --> 00:22:24,996
Speaker 2: right now, so your partner care partner has to come

390
00:22:25,036 --> 00:22:27,676
Speaker 2: and feed you, and that there's a number of reasons

391
00:22:27,676 --> 00:22:29,916
Speaker 2: why that's just they would prefer to do it themselves.

392
00:22:29,996 --> 00:22:30,436
Speaker 2: Of course.

393
00:22:30,556 --> 00:22:35,316
Speaker 1: The basic it's a basic function of adult life is

394
00:22:35,356 --> 00:22:36,036
Speaker 1: you feed yourself.

395
00:22:36,076 --> 00:22:39,596
Speaker 2: You feed yourself, and depending on someone changes that relationship

396
00:22:39,676 --> 00:22:43,196
Speaker 2: with that person, right, And and so feeding is one

397
00:22:43,196 --> 00:22:46,276
Speaker 2: that we've looked at a fair amount. It is challenging.

398
00:22:46,276 --> 00:22:49,716
Speaker 2: There are safety considerations, so we're careful about what utensils

399
00:22:49,756 --> 00:22:53,836
Speaker 2: and what type of food. But I think if it

400
00:22:53,876 --> 00:22:57,876
Speaker 2: can feed, if you can feed yourself, if you can

401
00:22:58,036 --> 00:23:00,356
Speaker 2: get yourself a glass of water, you can do basic

402
00:23:00,556 --> 00:23:03,796
Speaker 2: you know, hygiene maintenance things for yourself. I think that's

403
00:23:03,876 --> 00:23:07,036
Speaker 2: kind of the base line of regaining a sense of agency, right,

404
00:23:08,356 --> 00:23:08,956
Speaker 2: And what do you have.

405
00:23:08,956 --> 00:23:11,276
Speaker 1: To figure out to make feeding really work that you

406
00:23:11,316 --> 00:23:12,636
Speaker 1: haven't figured out yet?

407
00:23:12,836 --> 00:23:16,276
Speaker 2: Yeah, it's a mixture of we have. Food is kind

408
00:23:16,316 --> 00:23:20,236
Speaker 2: of messy, mushy, right, it's not like a rigid object.

409
00:23:20,356 --> 00:23:24,076
Speaker 1: Well, and it's it's super heterogeneous, right, Food is many

410
00:23:24,116 --> 00:23:25,236
Speaker 1: different kinds of.

411
00:23:25,196 --> 00:23:29,676
Speaker 2: Ways exactly, and so but there is you know, one

412
00:23:29,716 --> 00:23:31,836
Speaker 2: thing we've seen is you know, the caregiver is willing

413
00:23:31,876 --> 00:23:34,396
Speaker 2: to prepare the food, maybe put it in a special tray,

414
00:23:34,956 --> 00:23:37,356
Speaker 2: certain types of food that we agree it can work with.

415
00:23:37,436 --> 00:23:40,636
Speaker 2: So but of course you'd love that to get broader.

416
00:23:40,716 --> 00:23:43,036
Speaker 2: But you know, food in itself is challenging. Then there's

417
00:23:43,076 --> 00:23:46,196
Speaker 2: the safety challenges. But also, you know, a really interesting

418
00:23:46,236 --> 00:23:48,756
Speaker 2: part of this, which kind of relates to you know,

419
00:23:48,796 --> 00:23:53,436
Speaker 2: where AI is in general, is understanding the person, right,

420
00:23:54,756 --> 00:23:57,436
Speaker 2: are they ready for another bite? You know, are they

421
00:23:58,316 --> 00:24:01,036
Speaker 2: you know, what side would they like to be fed from.

422
00:24:01,156 --> 00:24:04,756
Speaker 2: That's all these little things that are about that human

423
00:24:05,076 --> 00:24:10,516
Speaker 2: interaction that make the product useful, exciting, acceptable in the home.

424
00:24:11,076 --> 00:24:13,996
Speaker 2: And so much of what's happening in ai AI right

425
00:24:14,036 --> 00:24:18,196
Speaker 2: now doesn't really get at the human side of the story.

426
00:24:18,276 --> 00:24:21,076
Speaker 2: And I think that's one thing where we would love

427
00:24:21,116 --> 00:24:25,596
Speaker 2: to see a advancements is really having better understanding of

428
00:24:26,276 --> 00:24:29,876
Speaker 2: these sort of more subtle cues and interactions that the

429
00:24:29,956 --> 00:24:32,036
Speaker 2: robot will need in order to be accepted.

430
00:24:35,996 --> 00:24:48,676
Speaker 1: We'll be back in just a minute. So I'm curious

431
00:24:48,956 --> 00:24:53,236
Speaker 1: because you have been working on robots for more than

432
00:24:53,276 --> 00:24:58,356
Speaker 1: twenty years, right, I mean, so much has happened in software, yeah, right,

433
00:24:58,396 --> 00:25:00,836
Speaker 1: and even in other kinds of hardware, Like you were

434
00:25:01,276 --> 00:25:04,156
Speaker 1: working on robots before there was an iPhone, right, and

435
00:25:04,196 --> 00:25:07,396
Speaker 1: so so much has changed, and it feels like, at

436
00:25:07,436 --> 00:25:09,796
Speaker 1: least as a as a layperson, certainly the home like

437
00:25:10,476 --> 00:25:13,436
Speaker 1: robots are not around the way iPhones are around. And

438
00:25:13,476 --> 00:25:17,156
Speaker 1: even now large language models are around. And so has

439
00:25:17,236 --> 00:25:18,916
Speaker 1: less happened with robots than you thought.

440
00:25:20,676 --> 00:25:24,076
Speaker 2: I would say less in terms of deployed robots, right,

441
00:25:24,076 --> 00:25:27,476
Speaker 2: Like the rumba is still one of the most successful robots,

442
00:25:27,516 --> 00:25:30,596
Speaker 2: but there aren't many other examples like it, right, and

443
00:25:30,636 --> 00:25:31,836
Speaker 2: so why is that?

444
00:25:32,436 --> 00:25:32,676
Speaker 1: Yes?

445
00:25:32,756 --> 00:25:32,876
Speaker 2: What?

446
00:25:33,076 --> 00:25:34,556
Speaker 1: Like that's a great way to ask it, and that

447
00:25:34,716 --> 00:25:37,316
Speaker 1: I mean, you worked with Rodney Brooks, right, one of

448
00:25:37,356 --> 00:25:40,956
Speaker 1: the rumbook guys at MIT, and that's like the only

449
00:25:41,116 --> 00:25:44,076
Speaker 1: robot anyone can name, and there are lots of failed

450
00:25:44,276 --> 00:25:46,076
Speaker 1: companies that were rumba.

451
00:25:45,916 --> 00:25:49,436
Speaker 2: Like right, yeah, yeah, and so yeah, So I did

452
00:25:49,476 --> 00:25:52,196
Speaker 2: my PhD with rod Brooks at MIT and at the

453
00:25:52,236 --> 00:25:56,516
Speaker 2: time he was spinning up a robot, inventing the rumba.

454
00:25:56,556 --> 00:25:58,756
Speaker 2: And around the time I finished, he had just sold

455
00:25:58,756 --> 00:26:01,476
Speaker 2: his million rumba and he's he inspired me to go

456
00:26:01,516 --> 00:26:02,796
Speaker 2: off to my own company.

457
00:26:03,156 --> 00:26:05,476
Speaker 1: I mean, it must have seemed like they're here home

458
00:26:05,556 --> 00:26:08,636
Speaker 1: robots are here, like you would think at that moment,

459
00:26:08,636 --> 00:26:09,956
Speaker 1: it's like, oh, hey, here we go.

460
00:26:10,236 --> 00:26:13,956
Speaker 2: Yeah yeah, well but things seemed very hard, right, so

461
00:26:13,996 --> 00:26:17,596
Speaker 2: it's like, okay, a vacuum cleaner that bounces around off walls,

462
00:26:18,636 --> 00:26:21,316
Speaker 2: like we can do that, right, Yeah. And I think

463
00:26:21,436 --> 00:26:25,356
Speaker 2: the insight from Rumba that I still carry is, you know,

464
00:26:25,396 --> 00:26:29,476
Speaker 2: they thought really carefully about the price, the complexity of

465
00:26:29,476 --> 00:26:32,756
Speaker 2: the device. They weren't trying to There was a competitor

466
00:26:32,876 --> 00:26:34,676
Speaker 2: call the Electrial Lucks at the time that was about

467
00:26:34,716 --> 00:26:37,836
Speaker 2: three thousand dollars had all sorts of cool features and

468
00:26:37,916 --> 00:26:41,756
Speaker 2: computer and sensing. But with the Rumbo, what they did

469
00:26:41,796 --> 00:26:44,756
Speaker 2: is they went to the mall and they asked people,

470
00:26:45,436 --> 00:26:47,596
Speaker 2: how much can you spend on something without having to

471
00:26:47,636 --> 00:26:49,836
Speaker 2: ask your significant other for permission?

472
00:26:49,916 --> 00:26:50,876
Speaker 1: Oh?

473
00:26:50,916 --> 00:26:53,076
Speaker 2: And the answer was about two hundred dollars.

474
00:26:53,316 --> 00:26:56,236
Speaker 1: I was just talking to a friend who basically does

475
00:26:56,276 --> 00:26:59,156
Speaker 1: a B to B business. That's business a business, and

476
00:26:59,196 --> 00:27:01,676
Speaker 1: apparently that's a sales strategy in business as well. Like

477
00:27:01,916 --> 00:27:05,356
Speaker 1: the person who you're selling to they have some spending

478
00:27:05,396 --> 00:27:07,916
Speaker 1: authority that they can spend on their own. Yes, of course,

479
00:27:07,956 --> 00:27:10,916
Speaker 1: it's about a thing times greater. Right, It's like they

480
00:27:10,916 --> 00:27:13,596
Speaker 1: can spend one hundred thousand dollars or whatever without asking

481
00:27:13,636 --> 00:27:15,996
Speaker 1: somebody else, So you really want to price it like that,

482
00:27:16,076 --> 00:27:17,276
Speaker 1: otherwise it's much harder.

483
00:27:17,356 --> 00:27:21,156
Speaker 2: So anyway, yeah, exactly, But you know, I think the

484
00:27:21,276 --> 00:27:24,196
Speaker 2: room to success in part was they made a minimal

485
00:27:24,236 --> 00:27:29,516
Speaker 2: product within that spending, you know, authority and kind of

486
00:27:29,756 --> 00:27:31,876
Speaker 2: made it all work. And I think the reason we

487
00:27:31,876 --> 00:27:34,196
Speaker 2: haven't seen a lot beyond that is it's hard. It's

488
00:27:34,236 --> 00:27:37,476
Speaker 2: been hard traditionally to find the things that match all

489
00:27:37,516 --> 00:27:41,836
Speaker 2: those constraints. That technology can be there, the use case

490
00:27:41,876 --> 00:27:44,796
Speaker 2: can be there, right clearly in home care, there's so

491
00:27:44,876 --> 00:27:48,036
Speaker 2: much need for help in the home, but can we

492
00:27:48,076 --> 00:27:50,156
Speaker 2: do it at a price? And is it technology there?

493
00:27:50,676 --> 00:27:52,716
Speaker 2: You know, five thousand dollars would be a lot of

494
00:27:52,756 --> 00:27:54,556
Speaker 2: money for most families.

495
00:27:54,396 --> 00:27:57,716
Speaker 1: Yes, although for a caregiver, right like, if you narrow

496
00:27:57,756 --> 00:27:59,676
Speaker 1: it not to like doing the dishes.

497
00:27:59,716 --> 00:28:01,116
Speaker 2: But yeah, I.

498
00:28:01,076 --> 00:28:07,076
Speaker 1: Mean, you know, many people have been bankrupted by just

499
00:28:07,116 --> 00:28:11,676
Speaker 1: needing to hire a simple caregiver, right Like, people could

500
00:28:11,716 --> 00:28:13,956
Speaker 1: definitely figure out how to spend five thousand dollars on

501
00:28:13,996 --> 00:28:18,556
Speaker 1: a robot that could significantly replace a caregiver. One would think,

502
00:28:18,716 --> 00:28:20,236
Speaker 1: I guess that's your bet, that's your.

503
00:28:20,596 --> 00:28:23,236
Speaker 2: That's partial partially our bet. Yeah, and you know, I

504
00:28:23,236 --> 00:28:25,956
Speaker 2: think it's not replacing, it's it's augmenting, you know, I

505
00:28:25,996 --> 00:28:28,676
Speaker 2: think think what makes sense, yes, because we're not going

506
00:28:28,716 --> 00:28:30,996
Speaker 2: to do everything, and nor should we, you know, but

507
00:28:31,116 --> 00:28:34,716
Speaker 2: there's just aren't enough people ours caregiving hours in the

508
00:28:35,076 --> 00:28:37,036
Speaker 2: in the world today to take care of everyone who

509
00:28:37,036 --> 00:28:38,956
Speaker 2: needs care. That's just the truth of it.

510
00:28:39,116 --> 00:28:43,156
Speaker 1: Yeah, zooming out from from Hello robot, from your company

511
00:28:43,196 --> 00:28:46,276
Speaker 1: to the field more generally, like what's the landscape look

512
00:28:46,356 --> 00:28:48,636
Speaker 1: to you? How does the landscape look to you right now?

513
00:28:48,876 --> 00:28:52,476
Speaker 2: Yeah, it's fascinating there. There's a lot of first off,

514
00:28:52,476 --> 00:28:56,116
Speaker 2: a lot of capital flowing into it. Into two areas really,

515
00:28:56,196 --> 00:28:58,756
Speaker 2: one is robot foundation models. You know, this is sort

516
00:28:58,756 --> 00:29:02,476
Speaker 2: of deep learning the alms of robots. And then the

517
00:29:02,916 --> 00:29:05,956
Speaker 2: other would be humanoids, right, and sometimes there's both, right,

518
00:29:06,036 --> 00:29:09,396
Speaker 2: but there's a lot of excitement around that, and would

519
00:29:09,396 --> 00:29:11,396
Speaker 2: say generally they both have a little ways to go

520
00:29:11,476 --> 00:29:14,716
Speaker 2: before they get deployed and really applied.

521
00:29:15,236 --> 00:29:18,996
Speaker 1: I mean the foundation model one, you know, an ll

522
00:29:19,116 --> 00:29:22,956
Speaker 1: M for robots sounds good to me, It seems interesting

523
00:29:22,996 --> 00:29:24,756
Speaker 1: to me. Tell me about what's happening there.

524
00:29:24,916 --> 00:29:29,276
Speaker 2: Yeah, Well, it's first off, it's remarkable what you see

525
00:29:29,876 --> 00:29:33,396
Speaker 2: the robots doing with these these models, right, so peeling

526
00:29:33,396 --> 00:29:38,316
Speaker 2: an egg or tying shoelaces or folding laundry. And they're

527
00:29:38,316 --> 00:29:41,316
Speaker 2: getting better, right, They're they're getting they're having long I think.

528
00:29:41,156 --> 00:29:42,956
Speaker 1: The peeling an egg was teleoperated.

529
00:29:43,076 --> 00:29:45,876
Speaker 2: Okay, fair enough, but they are doing things like that

530
00:29:45,876 --> 00:29:48,716
Speaker 2: that you think, wow, I just could not imagine this.

531
00:29:50,076 --> 00:29:52,436
Speaker 2: And and they're getting better, and that they're having longer

532
00:29:53,596 --> 00:29:55,916
Speaker 2: memory so they can do stuff over a longer period

533
00:29:55,956 --> 00:29:59,276
Speaker 2: of time. They're getting better at recovering from from issues.

534
00:30:00,756 --> 00:30:03,676
Speaker 2: But also they haven't yet learned to generalize. Right, So

535
00:30:03,716 --> 00:30:07,716
Speaker 2: if you think about what they're learning, that's such as

536
00:30:07,756 --> 00:30:10,876
Speaker 2: we were saying, such richness to diversity in the environments

537
00:30:10,876 --> 00:30:14,076
Speaker 2: they meant to operate, and they're learning on one kind

538
00:30:14,076 --> 00:30:16,836
Speaker 2: of little island and the whole sea of data that

539
00:30:16,836 --> 00:30:21,076
Speaker 2: they need access to to really be generally useful. And

540
00:30:21,436 --> 00:30:23,876
Speaker 2: so that's kind of a real challenge that that field

541
00:30:23,916 --> 00:30:26,316
Speaker 2: is trying to work through, is how do you get

542
00:30:26,356 --> 00:30:30,756
Speaker 2: to that generality and that robustness to that diversity. And

543
00:30:30,796 --> 00:30:33,476
Speaker 2: it's very much an unsolved problem. I think they're making progress,

544
00:30:34,116 --> 00:30:36,156
Speaker 2: but it also means to me that this is not

545
00:30:36,276 --> 00:30:40,196
Speaker 2: something that we're going to be deploying into a robot

546
00:30:40,196 --> 00:30:41,236
Speaker 2: product anytime soon.

547
00:30:42,116 --> 00:30:44,436
Speaker 1: You know, it's interesting because it seems like one of

548
00:30:44,476 --> 00:30:50,036
Speaker 1: the frankly big surprises about large language models is how

549
00:30:50,036 --> 00:30:54,836
Speaker 1: good they are at generalizing, right, how they often are

550
00:30:54,916 --> 00:30:57,036
Speaker 1: much better than specialist models.

551
00:30:57,116 --> 00:30:57,316
Speaker 2: Right.

552
00:30:57,356 --> 00:30:59,756
Speaker 1: Just take a frontier model and throw anything at it

553
00:30:59,836 --> 00:31:01,916
Speaker 1: and it will be really good. Not quite, but kind of.

554
00:31:02,596 --> 00:31:05,276
Speaker 1: And so it's interesting to hear that that's not at

555
00:31:05,276 --> 00:31:08,956
Speaker 1: all the case with robot foundation models. Why do you

556
00:31:09,196 --> 00:31:09,636
Speaker 1: pose that.

557
00:31:09,676 --> 00:31:12,756
Speaker 2: Is, Well, you know, I think one thing is just

558
00:31:12,956 --> 00:31:17,036
Speaker 2: the amount of data that's available, Right, So if you

559
00:31:17,076 --> 00:31:20,116
Speaker 2: think of a ll M, one simple way to think

560
00:31:20,116 --> 00:31:21,956
Speaker 2: of it, it's going to predict the next word that

561
00:31:21,996 --> 00:31:24,196
Speaker 2: you say, and it has the whole corpus of the

562
00:31:24,276 --> 00:31:28,156
Speaker 2: Internet to make that prediction. Right, for a robot to

563
00:31:28,396 --> 00:31:30,316
Speaker 2: predict the next thing that's going to happen when it's

564
00:31:30,476 --> 00:31:32,996
Speaker 2: peeling the egg, there's not a lot of data out

565
00:31:32,996 --> 00:31:38,316
Speaker 2: there for it to understand that. And some people are saying, well,

566
00:31:38,316 --> 00:31:42,796
Speaker 2: I can do that in simulation, but very hard to

567
00:31:42,836 --> 00:31:46,076
Speaker 2: imagine simulating a lot of that kind of complexity that's

568
00:31:46,076 --> 00:31:47,516
Speaker 2: actually in the world, just.

569
00:31:47,516 --> 00:31:49,356
Speaker 1: Because like the world is so diverse. What do you

570
00:31:49,436 --> 00:31:50,596
Speaker 1: even simulate or how.

571
00:31:50,556 --> 00:31:54,676
Speaker 2: Do you even simulate an egg being peeled? And so,

572
00:31:55,356 --> 00:31:59,156
Speaker 2: you know, I think it's a challenge because robots are

573
00:31:59,196 --> 00:32:00,196
Speaker 2: in the physical world.

574
00:32:00,876 --> 00:32:06,116
Speaker 1: Is there something as I understand it, the basic model,

575
00:32:06,276 --> 00:32:09,596
Speaker 1: the transformer model that powers all of the frontier elements

576
00:32:09,596 --> 00:32:14,996
Speaker 1: today was developed based on the nature of language itself, right, Like,

577
00:32:16,876 --> 00:32:18,756
Speaker 1: is there a thought that that's maybe just not the

578
00:32:18,796 --> 00:32:23,236
Speaker 1: best fundamental model for understanding the physical world for training robots.

579
00:32:23,396 --> 00:32:26,316
Speaker 1: Might it be the case that some very different approach

580
00:32:26,396 --> 00:32:27,636
Speaker 1: ends up working much better.

581
00:32:28,156 --> 00:32:30,116
Speaker 2: Yeah, I know less about that. I do know there

582
00:32:30,556 --> 00:32:33,916
Speaker 2: is a large interest in world models today that can

583
00:32:33,996 --> 00:32:37,116
Speaker 2: kind of infer the physics of the world. And I

584
00:32:37,156 --> 00:32:41,196
Speaker 2: think one thing that that is beneficial is there is

585
00:32:41,236 --> 00:32:45,276
Speaker 2: so much structure in the world, both from physics.

586
00:32:44,916 --> 00:32:48,076
Speaker 1: Like laws of physics, laws of physics, yep, but also.

587
00:32:47,876 --> 00:32:51,196
Speaker 2: In the way that we design our homes, our appliances

588
00:32:51,396 --> 00:32:53,796
Speaker 2: are you know, there is a lot of structure to

589
00:32:53,836 --> 00:32:57,956
Speaker 2: be gained and learned, and so you can imagine these

590
00:32:58,036 --> 00:33:02,396
Speaker 2: models become very good at predicting what your home is

591
00:33:02,396 --> 00:33:03,716
Speaker 2: going to look like because it's seen a lot of

592
00:33:03,716 --> 00:33:07,476
Speaker 2: homes and they don't vary that much fundamentally, so so

593
00:33:07,556 --> 00:33:10,436
Speaker 2: that there is there is an enough structure to learn

594
00:33:10,556 --> 00:33:14,556
Speaker 2: from what that model and how that transpires. I don't know,

595
00:33:14,636 --> 00:33:15,796
Speaker 2: but people are exploring that.

596
00:33:16,276 --> 00:33:19,516
Speaker 1: I mean, one of the one of the obvious challenges

597
00:33:19,596 --> 00:33:22,516
Speaker 1: is we're just making the robots come live in our world, right,

598
00:33:22,556 --> 00:33:26,196
Speaker 1: And I could imagine, for a case like someone with quadriplegia,

599
00:33:26,756 --> 00:33:29,276
Speaker 1: it might I could imagine a universe where it's like, well,

600
00:33:29,356 --> 00:33:32,036
Speaker 1: if you set up your room this way, not you

601
00:33:32,076 --> 00:33:33,756
Speaker 1: have to remodel your house, but you have to do

602
00:33:33,876 --> 00:33:36,676
Speaker 1: some set of things, then our robot can do a lot.

603
00:33:36,716 --> 00:33:40,036
Speaker 1: And I can imagine that being worthwhile in many settings.

604
00:33:40,476 --> 00:33:42,796
Speaker 1: Is that something you are doing or thinking about? Sort

605
00:33:42,836 --> 00:33:44,356
Speaker 1: of a compromised solution like that?

606
00:33:44,716 --> 00:33:48,276
Speaker 2: Absolutely? And you know, as an example, a Stretch three

607
00:33:48,356 --> 00:33:51,956
Speaker 2: as it is today, has a hard time opening some refrigerators, right,

608
00:33:51,956 --> 00:33:55,636
Speaker 2: It's just the magnetic force is too high. But you

609
00:33:55,676 --> 00:33:59,036
Speaker 2: can attach a little you know, grabbing thing that breaks

610
00:33:59,036 --> 00:34:02,076
Speaker 2: the seal and makes it easy to open it. Right,

611
00:34:02,116 --> 00:34:04,676
Speaker 2: And so it's a three dollars apart, you stick it

612
00:34:04,676 --> 00:34:07,316
Speaker 2: onto your fridge. Now Stretch can open their fridge for

613
00:34:07,396 --> 00:34:12,196
Speaker 2: you and for for these users, you know, those accommodations

614
00:34:12,236 --> 00:34:15,636
Speaker 2: make a lot of sense, and uh, you know, I

615
00:34:15,636 --> 00:34:18,556
Speaker 2: do think you know, maybe over time you'll have special

616
00:34:18,556 --> 00:34:20,916
Speaker 2: plates at the robot nose it cannot you know, grab

617
00:34:20,956 --> 00:34:23,876
Speaker 2: and not break, and you know, there will be accommodations

618
00:34:23,876 --> 00:34:29,196
Speaker 2: that people make for it. But you know, for for now,

619
00:34:29,276 --> 00:34:32,636
Speaker 2: we are using and counting on those types of accommodations.

620
00:34:33,836 --> 00:34:37,076
Speaker 1: So I want to do two kind of forward looking

621
00:34:37,396 --> 00:34:41,796
Speaker 1: things first, the sad one and the happy one. Basically okay,

622
00:34:41,836 --> 00:34:44,196
Speaker 1: first is how might it not work out? And then

623
00:34:44,236 --> 00:34:45,636
Speaker 1: the second is what does it look like if it

624
00:34:45,636 --> 00:34:49,276
Speaker 1: does work out for the low robots, So like what

625
00:34:49,396 --> 00:34:52,956
Speaker 1: might go wrong for for your company in a you know,

626
00:34:53,036 --> 00:34:54,596
Speaker 1: significant way, right.

627
00:34:57,156 --> 00:34:59,596
Speaker 2: Well, there's there's always the risk of just the business

628
00:34:59,596 --> 00:35:01,796
Speaker 2: doesn't work out. I think, I think from from the

629
00:35:01,836 --> 00:35:05,756
Speaker 2: technology and product side, it's really about acceptance, right, So

630
00:35:05,916 --> 00:35:09,516
Speaker 2: I think implicit to bringing this technology and your life

631
00:35:09,556 --> 00:35:12,156
Speaker 2: into your home, there has to be trust, right, and

632
00:35:12,236 --> 00:35:16,396
Speaker 2: people have to trust and want this technology around them,

633
00:35:16,756 --> 00:35:19,756
Speaker 2: and so I think, you know, there there's always a

634
00:35:19,836 --> 00:35:21,756
Speaker 2: risk that people like you know, what we're going to

635
00:35:21,996 --> 00:35:25,396
Speaker 2: we're going to find different ways to addrust the caregiving problem,

636
00:35:25,796 --> 00:35:28,796
Speaker 2: you know, which you know, I should. There's lots of

637
00:35:28,876 --> 00:35:31,076
Speaker 2: going to be lots of solutions to that, but robots

638
00:35:31,116 --> 00:35:33,516
Speaker 2: may may not turn out to be part of that story.

639
00:35:34,916 --> 00:35:37,116
Speaker 2: We're bullish that it will, but there it's a real

640
00:35:37,196 --> 00:35:40,276
Speaker 2: kind of complex dynamic between the people and the technology.

641
00:35:41,916 --> 00:35:44,876
Speaker 2: So that's that's sort of the the bare case. The

642
00:35:44,876 --> 00:35:47,796
Speaker 2: bowl case would be, you know, ten years out. Call

643
00:35:47,836 --> 00:35:50,476
Speaker 2: it a five thousand dollars device that's in your home

644
00:35:51,396 --> 00:35:55,756
Speaker 2: that can provide caregiving support in the times when there's

645
00:35:55,796 --> 00:35:59,396
Speaker 2: not care present. So basically it's just kind of facilitating

646
00:35:59,596 --> 00:36:02,396
Speaker 2: quality of life at home, agency at home as you

647
00:36:02,436 --> 00:36:05,996
Speaker 2: want to kind of age independently without having so when

648
00:36:05,996 --> 00:36:09,596
Speaker 2: you're when your grandkids come over, you're not they're not

649
00:36:09,636 --> 00:36:12,076
Speaker 2: doing chores for you. You're interacting, right because of the

650
00:36:12,156 --> 00:36:13,716
Speaker 2: robots helping fill that gap.

651
00:36:17,396 --> 00:36:19,676
Speaker 1: We'll be back in a minute with the lightning round.

652
00:36:33,996 --> 00:36:35,316
Speaker 1: Let's finish with the lightning round.

653
00:36:35,596 --> 00:36:37,276
Speaker 2: Okay, do you.

654
00:36:37,196 --> 00:36:37,956
Speaker 1: Still make art?

655
00:36:38,876 --> 00:36:42,756
Speaker 2: Oh? I not for a while, But that is how

656
00:36:42,956 --> 00:36:47,796
Speaker 2: I got into robotics, was doing robotic sculpture, and then

657
00:36:47,916 --> 00:36:52,276
Speaker 2: that led one thing led to another. But yeah, really

658
00:36:52,996 --> 00:36:55,676
Speaker 2: fascinated by well, love art but also love you know,

659
00:36:55,756 --> 00:36:57,596
Speaker 2: kind of robotics and how it relates to art.

660
00:36:58,196 --> 00:37:00,836
Speaker 1: What's the least practical robot you've ever built?

661
00:37:02,476 --> 00:37:06,516
Speaker 2: Wow? Uh? So let's see in the in the nineteen

662
00:37:06,596 --> 00:37:09,836
Speaker 2: nineties in San Francisco, we had a robotic art performance

663
00:37:09,876 --> 00:37:14,796
Speaker 2: troupe and it was a theatrical multimedia performance of robots

664
00:37:14,956 --> 00:37:19,476
Speaker 2: on stage acting performing. You know, this was a kind

665
00:37:19,476 --> 00:37:25,676
Speaker 2: of robotic gallery of kind of characters from the from

666
00:37:25,716 --> 00:37:29,796
Speaker 2: the local neighborhood. So these were a street musician and

667
00:37:31,156 --> 00:37:35,396
Speaker 2: you know, someone who was got of causing trouble and

668
00:37:35,436 --> 00:37:39,236
Speaker 2: so they were kind of robotic denizens of the Tenderloin

669
00:37:39,316 --> 00:37:42,676
Speaker 2: of nineteen nineties, San Francisco put onto stage.

670
00:37:43,956 --> 00:37:45,956
Speaker 1: What was the first robot you ever built?

671
00:37:46,636 --> 00:37:49,756
Speaker 2: It was actually the Robotic Street Preacher. So it was

672
00:37:49,796 --> 00:37:53,436
Speaker 2: a little robot that would go around and big spinning

673
00:37:53,756 --> 00:37:57,636
Speaker 2: head of a speaker and it would be on stage

674
00:37:57,636 --> 00:38:01,476
Speaker 2: and perform and we would sit there and tell operate it.

675
00:38:01,556 --> 00:38:04,036
Speaker 2: And it had, you know, no real autonomy. This is

676
00:38:04,676 --> 00:38:06,916
Speaker 2: you know, everything with the hands salted, handbuilt and their

677
00:38:06,956 --> 00:38:12,356
Speaker 2: computers were super tiny, are super a small capability, And

678
00:38:12,396 --> 00:38:14,836
Speaker 2: we put that on stage and performed it.

679
00:38:15,716 --> 00:38:17,956
Speaker 1: Did you ever try and build robots when you were

680
00:38:17,996 --> 00:38:18,396
Speaker 1: a kid?

681
00:38:19,836 --> 00:38:23,796
Speaker 2: No, you know, Strangely, I never had a fascination with

682
00:38:23,956 --> 00:38:27,076
Speaker 2: robusts as a kid. I came out of grew up

683
00:38:27,076 --> 00:38:31,196
Speaker 2: in Washington State on a farm, and uh, you know,

684
00:38:31,516 --> 00:38:33,756
Speaker 2: I think my interest in robots came from having to

685
00:38:33,756 --> 00:38:36,516
Speaker 2: do farm work and learning about automation.

686
00:38:36,476 --> 00:38:38,916
Speaker 1: Like Henry Ford. You know, Henry Ford was the same

687
00:38:39,036 --> 00:38:41,316
Speaker 1: story about a farm and thought that was too much

688
00:38:41,396 --> 00:38:43,756
Speaker 1: labor exactly, and I just thought, you know, there's got

689
00:38:43,796 --> 00:38:45,076
Speaker 1: to be a better way than this.

690
00:38:45,316 --> 00:38:48,836
Speaker 2: My my mother's family, which is Mexican, had a history

691
00:38:48,876 --> 00:38:52,156
Speaker 2: and kind of uh kind of working in the fields

692
00:38:52,196 --> 00:38:55,276
Speaker 2: and kind of migrant labor, and so kind of the

693
00:38:55,356 --> 00:38:59,276
Speaker 2: relationship to labor and work kind of inspired my interest

694
00:38:59,316 --> 00:39:01,356
Speaker 2: in automation and then roproducts.

695
00:39:02,476 --> 00:39:04,316
Speaker 1: Did you ever try and build a machine to do

696
00:39:04,396 --> 00:39:05,236
Speaker 1: your chores for you?

697
00:39:05,316 --> 00:39:09,116
Speaker 2: I absolutely did. It never worked, but it was fun,

698
00:39:09,556 --> 00:39:09,756
Speaker 2: you know.

699
00:39:10,276 --> 00:39:12,156
Speaker 1: I mean, I guess that's still what you're working on

700
00:39:12,196 --> 00:39:12,876
Speaker 1: at some level.

701
00:39:13,276 --> 00:39:13,796
Speaker 2: Exactly.

702
00:39:15,116 --> 00:39:17,036
Speaker 1: What's one thing you've learned working with Frank Gary?

703
00:39:17,476 --> 00:39:22,756
Speaker 2: Oh? Right, So we had a project looking at this

704
00:39:23,076 --> 00:39:28,756
Speaker 2: conceptual but a large robotic kind of botanical garden and

705
00:39:28,876 --> 00:39:31,916
Speaker 2: this sole thing in Singapore, so that we were looking

706
00:39:31,956 --> 00:39:36,516
Speaker 2: at developing that with him and Uh, you know, I'd say, actually,

707
00:39:36,556 --> 00:39:41,476
Speaker 2: the thing that was most interesting is just he really

708
00:39:41,476 --> 00:39:43,596
Speaker 2: put a lot of trust into the young people in

709
00:39:43,916 --> 00:39:48,476
Speaker 2: the studio and really was generous in that way. And

710
00:39:48,596 --> 00:39:49,996
Speaker 2: you know, it's been an inspiration for me.

711
00:39:50,796 --> 00:39:53,716
Speaker 1: That's a lovely surprise one. This is I guess some

712
00:39:54,116 --> 00:39:56,876
Speaker 1: bias that I have. I imagine a famous architect as

713
00:39:56,956 --> 00:39:57,836
Speaker 1: not being like that.

714
00:39:58,156 --> 00:40:01,396
Speaker 2: So to hear that so lovely, very very generous in

715
00:40:01,436 --> 00:40:04,316
Speaker 2: a way that can be inspiring to to the to

716
00:40:04,636 --> 00:40:07,036
Speaker 2: people in their career. So I would try to carry

717
00:40:07,076 --> 00:40:08,316
Speaker 2: that forward in my own work.

718
00:40:09,276 --> 00:40:10,636
Speaker 1: Uh, thank you for your time.

719
00:40:11,116 --> 00:40:13,196
Speaker 2: Yeah, it's been a pleasure. Thank you.

720
00:40:20,876 --> 00:40:25,156
Speaker 1: Aaron Edzinger is the co founder and CEO of Hello Robot.

721
00:40:25,916 --> 00:40:28,956
Speaker 1: Today's show was produced by Gabriel Hunter Chang and edited

722
00:40:28,956 --> 00:40:33,116
Speaker 1: by Lydia Jane Kott. Our engineer this week was Hansdale Sheet.

723
00:40:33,276 --> 00:40:35,836
Speaker 1: We're always looking for ideas for who to talk to

724
00:40:35,956 --> 00:40:38,836
Speaker 1: and what to cover on the show. You can email

725
00:40:38,956 --> 00:40:42,516
Speaker 1: us at problem at Pushkin dot fm. You can find

726
00:40:42,516 --> 00:40:44,876
Speaker 1: me on x at Jacob Goldstein. You can find me

727
00:40:44,916 --> 00:40:48,116
Speaker 1: on LinkedIn. I'm Jacob Goldstein. Thank you very much for

728
00:40:48,196 --> 00:40:50,356
Speaker 1: listening to the show, and we'll be back next week

729
00:40:50,396 --> 00:40:57,436
Speaker 1: with another episode