Transkrip penuh untuk Ep 97 - Artificial Intelligence 2.0: No Robots Were Harmed in the Making of This Episode

great stock market crash happened when he was 19 34 00:01:53.969 --> 00:01:57.819 and D-Day just after he turned 34. He was 59 35 00:01:57.819 --> 00:02:00.620 when Neil Armstrong walked on the moon and 79 when 36 00:02:00.659 --> 00:02:04.569 the Belgian music group Technotronic released their seminal techno anthem, 37 00:02:04.969 --> 00:02:09.069 Pump Up the Jam. And of course, there are the 38 00:02:09.169 --> 00:02:13.680 technological changes he himself witnessed during his life. Cars replacing 39 00:02:13.740 --> 00:02:17.060 horses and buggies, the beginning of commercial radio and later 40 00:02:17.139 --> 00:02:20.270 the rise of television. The weapons of mass destruction were 41 00:02:20.330 --> 00:02:23.030 new when he was a child and became unimaginable in 42 00:02:23.050 --> 00:02:26.590 the scope of their power when he was in middle age. Phonographs, 43 00:02:26.870 --> 00:02:31.210 record players, cassette tapes, CDs, he saw them all. The 44 00:02:31.250 --> 00:02:35.590 pill was invented, single-engine propeller-driven planes became jets, and on 45 00:02:35.750 --> 00:02:39.530 and on and on. It leads one to wonder, what 46 00:02:39.569 --> 00:02:43.129 will my grandkids say about the changes I have witnessed 47 00:02:43.460 --> 00:02:47.180 and the unforeseen changes I will yet witness? Is the 48 00:02:47.259 --> 00:02:49.819 moment at which we now find ourselves as significant as 49 00:02:49.879 --> 00:02:53.800 it feels? The spread of artificial intelligence into seemingly every 50 00:02:53.939 --> 00:02:56.650 aspect of our lives strikes many of us as not 51 00:02:56.710 --> 00:03:01.349 merely new, but a new kind of new, raising questions 52 00:03:01.409 --> 00:03:05.069 and challenges our ancestors never had to consider, along with 53 00:03:05.110 --> 00:03:09.349 more familiar ones. How do we use technology well without 54 00:03:09.469 --> 00:03:13.759 undermining our own humanity? How will AI and robotics change 55 00:03:13.879 --> 00:03:18.169 us as individuals and as societies? Do we want those changes? 56 00:03:18.710 --> 00:03:20.949 Will they make us better off, all things considered? Or 57 00:03:20.990 --> 00:03:25.469 will our high-tech lives be impoverished ones? Do we face 58 00:03:25.530 --> 00:03:29.289 the prospect of gaining the whole world, but losing our souls? 59 00:03:31.060 --> 00:03:34.259 I'm Matt Jordan. This is the Three Questions, Three Drinks podcast. 60 00:03:34.819 --> 00:03:55.400 Welcome to the conversation. Hello, everybody, and welcome to the 61 00:03:55.419 --> 00:03:57.639 Three Questions, Three Drinks podcast, the show where we make 62 00:03:57.680 --> 00:04:01.550 a few drinks and ask thought-provoking questions about an interesting topic. 63 00:04:02.250 --> 00:04:04.509 If we do it right, and let's face it, we 64 00:04:04.550 --> 00:04:07.389 always do. We create a conversation that you feel like 65 00:04:07.409 --> 00:04:09.740 you're a part of, and it gets better with every sip. 66 00:04:09.909 --> 00:04:11.639 My name is Matt Jordan, and I'll be the host 67 00:04:11.659 --> 00:04:15.180 for today's episode. Joining me, as usual, is a woman 68 00:04:15.500 --> 00:04:19.300 whose very name causes AI chatbots to tremble in their 69 00:04:19.379 --> 00:04:23.639 digital boots, the human face of the anti-tech revolution, Dr. 70 00:04:23.660 --> 00:04:24.459 Amanda Adams. 71 00:04:24.879 --> 00:04:25.490 You said it. 72 00:04:26.589 --> 00:04:30.310 And the man who puts the chat in ChatGPT, the one, 73 00:04:30.750 --> 00:04:33.230 the only, the organic, Chris Michaelet. 74 00:04:33.250 --> 00:04:33.310 Hi. 75 00:04:34.329 --> 00:04:36.350 Welcome, Dr. Adams. Welcome, Mr. Michael. 76 00:04:36.370 --> 00:04:37.870 When did this become a game show? 77 00:04:40.089 --> 00:04:41.449 Are you ready to play? 78 00:04:41.529 --> 00:04:43.699 Come on down, everyone. Let's go. 79 00:04:44.339 --> 00:04:45.860 Very well done. I'm enjoying it so far. 80 00:04:45.879 --> 00:04:46.089 Thank you. 81 00:04:46.100 --> 00:04:49.000 Thank you. Hey, before we pour the first drink, actually, 82 00:04:49.019 --> 00:04:50.759 before we drink the first drink, I do want to 83 00:04:50.779 --> 00:04:53.620 invite anybody listening, make sure that you are subscribed to 84 00:04:53.660 --> 00:04:55.360 this podcast. You don't want to miss a minute of 85 00:04:55.399 --> 00:04:58.199 this and hitting that subscribe button. Maybe go ahead and 86 00:04:58.259 --> 00:05:00.139 press the like button. I think we can guarantee you're 87 00:05:00.170 --> 00:05:02.550 going to enjoy it. It doesn't cost you anything. Just 88 00:05:02.629 --> 00:05:05.250 do it. Um, so we've got a brand new drink 89 00:05:05.470 --> 00:05:09.509 in front of us today to kick things off with. Um, let's, 90 00:05:09.970 --> 00:05:12.110 can we take a taste before I, before I tell 91 00:05:12.129 --> 00:05:14.360 you anything about it? It's a nice, it's. 92 00:05:14.139 --> 00:05:15.259 Like a watermelon color. 93 00:05:15.279 --> 00:05:16.899 Yes. It's a, let me start by that too. There's 94 00:05:16.980 --> 00:05:23.819 no watermelon in it, but type of color. Oh, I 95 00:05:23.860 --> 00:05:26.279 need to taste some. Maybe. 96 00:05:26.899 --> 00:05:28.879 No, not Campari. I was going to say Amaro Nonino. 97 00:05:29.160 --> 00:05:31.120 A little more sweetness, I think, would be good. 98 00:05:31.379 --> 00:05:32.019 It's not bad. 99 00:05:32.060 --> 00:05:32.759 It's interesting. It's good. 100 00:05:32.980 --> 00:05:36.160 It's got lemon juice in it. Mm-hmm. Yeah. It feels 101 00:05:36.199 --> 00:05:39.000 like Amaro. Am I wrong on that? 102 00:05:39.310 --> 00:05:40.019 You're wrong on that. 103 00:05:40.959 --> 00:05:42.529 Maybe bourbon? It's got to be bourbon. 104 00:05:42.569 --> 00:05:43.230 There is bourbon. 105 00:05:43.930 --> 00:05:44.889 There is Campari. 106 00:05:45.110 --> 00:05:45.810 Oh, there is Campari. 107 00:05:45.829 --> 00:05:47.949 I was right. There is lemon

juice. There's sweet vermouth. 108 00:05:48.730 --> 00:05:51.730 And there is some Luxardo Maraschino decor. Oh, yeah. Okay. 109 00:05:51.810 --> 00:05:56.319 And a dash of bitters. Ooh. This is... brand-new drink 110 00:05:56.879 --> 00:06:01.300 that is called the clever prompt and in honor of 111 00:06:01.379 --> 00:06:04.399 today's episode I took a picture of a whole bunch 112 00:06:04.439 --> 00:06:06.839 of the liquor that I had in my house and 113 00:06:06.879 --> 00:06:10.410 I said to AI I said to chat GPT I 114 00:06:10.430 --> 00:06:13.410 want to impress some friends with a delicious drink here's 115 00:06:13.430 --> 00:06:16.769 what I have on hand it is 80 degrees and 116 00:06:16.790 --> 00:06:19.879 I need something I can throw together fairly quickly And, uh, 117 00:06:20.009 --> 00:06:21.310 it gave me three different suggestions. 118 00:06:21.490 --> 00:06:21.769 Okay. 119 00:06:22.129 --> 00:06:24.230 And, uh, and then, um, yeah, this is, this is 120 00:06:24.290 --> 00:06:26.990 one that actually gave me three and I, uh, I 121 00:06:27.029 --> 00:06:28.819 refused all of those cause they all had watermelon in it, 122 00:06:28.970 --> 00:06:30.290 which is something I have, but then I ran out 123 00:06:30.310 --> 00:06:33.189 of time to make it like a watermelon puree. Um, 124 00:06:33.449 --> 00:06:35.370 so I asked for something that had no watermelon that 125 00:06:35.389 --> 00:06:37.470 I could do quickly. That was the fourth one. And 126 00:06:37.490 --> 00:06:37.930 that is this. 127 00:06:38.149 --> 00:06:40.769 All right. And it's an invention. Like it's new. They're 128 00:06:40.810 --> 00:06:43.639 not suggesting the program is lying to me. 129 00:06:43.649 --> 00:06:47.069 You know, heaven forbid that chat GPT would lie. Um, 130 00:06:47.790 --> 00:06:49.120 but cool. Well, it's not bad. 131 00:06:49.240 --> 00:06:50.019 Well, it's a, it's a. 132 00:06:49.980 --> 00:06:51.240 Pretty good drink. That's interesting. 133 00:06:51.279 --> 00:06:53.660 I thought, I thought immediately Amanda would like throw this 134 00:06:53.800 --> 00:06:55.420 on the ground. As soon as she found out it 135 00:06:55.459 --> 00:06:57.899 was AI generated, she'd just like throw it against the window. 136 00:06:58.500 --> 00:06:59.310 Not in this house. 137 00:07:01.129 --> 00:07:02.889 We are recording in Amanda's house. I do think you 138 00:07:02.910 --> 00:07:05.050 could use a little more, um, and actually mention this. 139 00:07:05.089 --> 00:07:06.550 If you, if the drink is not as sweet as 140 00:07:06.569 --> 00:07:09.389 you'd like a little more of the, uh, Luxardo liqueur 141 00:07:09.410 --> 00:07:10.790 would be appropriate. So maybe next time. 142 00:07:10.810 --> 00:07:12.949 I don't mind it at all. It tastes a little 143 00:07:13.029 --> 00:07:17.779 paper plain-ish to me, which is bourbon, Amaro Nonino, Luxardo, and, um, 144 00:07:18.970 --> 00:07:19.750 I can't remember the other one. 145 00:07:20.149 --> 00:07:20.769

Some kind of bitters? 146 00:07:20.990 --> 00:07:21.810 Maybe lemon juice? 147 00:07:21.870 --> 00:07:28.279 So you mean it recycled something that already existed without attribution? 148 00:07:28.399 --> 00:07:32.420 Well, hey, look, this is a natural segue to the 149 00:07:32.439 --> 00:07:34.939 question I wanted to ask because my question to kick 150 00:07:35.000 --> 00:07:38.959 us off as we think about technologies and AI in 151 00:07:39.000 --> 00:07:44.310 particular is what does a healthy use of artificial intelligence 152 00:07:44.370 --> 00:07:44.790 look like? 153 00:07:44.850 --> 00:07:46.189 How do we integrate technology? 154 00:07:46.819 --> 00:07:48.800 Or can we? Maybe the answer is there's no healthy 155 00:07:48.839 --> 00:07:52.540 way to do it. But what does a good integration 156 00:07:52.660 --> 00:07:54.490 of artificial intelligence into our lives look like? 157 00:07:54.670 --> 00:07:57.009 Can I just say first, before we dive into that, 158 00:07:57.029 --> 00:07:59.329 that this, I think, is kind of born from we 159 00:07:59.370 --> 00:08:03.829 had a spat over text, the three of us. I 160 00:08:03.870 --> 00:08:06.189 think it was over text where I had thrown out 161 00:08:06.230 --> 00:08:10.230 some question ideas for a prior episode. And Amanda was like, 162 00:08:10.310 --> 00:08:12.970 are those AI generated? And I was like, yes, of course. Actually, 163 00:08:13.029 --> 00:08:14.629 I was the one who said that. And you guys 164 00:08:14.649 --> 00:08:18.829 were both appalled and very angry at me for using 165 00:08:18.870 --> 00:08:21.449 AI to come up with some ideas for questions. I've 166 00:08:21.470 --> 00:08:22.939 never been so angry in my life. And I was like, 167 00:08:23.000 --> 00:08:26.319 this is a fully acceptable use of AI. And then 168 00:08:26.339 --> 00:08:29.079 we're like, whoa, whoa, whoa. Let's save this. Let's not 169 00:08:29.160 --> 00:08:31.220 do this now. Let's save this for a recording, which 170 00:08:31.259 --> 00:08:34.039 we're doing right now. And we also, this is like 171 00:08:34.250 --> 00:08:36.870 AI 2.0, because in March of 25, so a little 172 00:08:36.889 --> 00:08:40.159 over a year ago, Which was only like three episodes ago. 173 00:08:40.320 --> 00:08:41.340 Yeah. 174 00:08:41.860 --> 00:08:43.340 But, oh yeah, almost a year and a half ago now, 175 00:08:43.399 --> 00:08:46.139 we did an AI episode. And it does feel like 176 00:08:46.639 --> 00:08:51.240 things have shifted quite a bit, for me anyway, regarding AI, 177 00:08:52.039 --> 00:08:57.250 maybe my AI optimism, or maybe a little less than 178 00:08:57.269 --> 00:08:58.629 I used to. But, yeah, okay. 179 00:08:58.669 --> 00:08:59.710 We can dive in. 180 00:08:59.730 --> 00:09:01.110 So what's a healthy use of it? 181 00:09:02.149 --> 00:09:02.889 That's the question. 182 00:09:04.730 --> 00:09:08.909 The argument we were starting to have was, I was 183 00:09:08.950 --> 00:09:12.190 saying I think it's completely acceptable to use AI as 184 00:09:12.269 --> 00:09:16.649 a creative brainstorming source. I don't think it's acceptable to 185 00:09:16.809 --> 00:09:19.929 use it as– to just use the content it generates. 186 00:09:21.330 --> 00:09:22.850 I think that you can use it, though, to come 187 00:09:22.909 --> 00:09:25.429 up with ideas and to refine things. I think about 188 00:09:25.450 --> 00:09:28.840 it like it's a super, super– if you prompt it right, 189 00:09:29.629 --> 00:09:32.659 you can act like a very super bright friend of 190 00:09:32.700 --> 00:09:37.580 yours that is going to bounce ideas off. so amanda's 191 00:09:37.620 --> 00:09:40.919 like fuming so why why does this make you like 192 00:09:40.960 --> 00:09:42.220 a smoke coming off. 193 00:09:42.019 --> 00:09:45.080 I tried to generate smoke out of my nostrils but. 194 00:09:45.019 --> 00:09:47.399 Help me understand why you couldn't like start having a 195 00:09:47.440 --> 00:09:50.639 conversation to generate new ideas just like you'd have with 196 00:09:50.659 --> 00:09:52.039 a friend what's wrong with. 197 00:09:51.950 --> 00:09:57.889 That well um just the first like there's two sort 198 00:09:57.929 --> 00:10:00.500 of avenues to answer that one is just the idea 199 00:10:00.659 --> 00:10:05.730 of anthropomorphizing AI, like turning it into a human, talking 200 00:10:05.750 --> 00:10:08.710 about it like a friend. It is a text prediction 201 00:10:08.769 --> 00:10:12.250 or text probability program. That's what it is. So it 202 00:10:12.330 --> 00:10:14.659 generates like word by word by word what is the 203 00:10:14.700 --> 00:10:15.940 most likely thing to come next. 204 00:10:15.960 --> 00:10:16.620 Right. 205 00:10:16.639 --> 00:10:18.679 So it is not a person, right? It is not 206 00:10:18.720 --> 00:10:21.539 like a friend. It doesn't have feelings or desires or 207 00:10:21.559 --> 00:10:25.120 intentions or things like that. And so just there's that. 208 00:10:25.179 --> 00:10:26.429 But I think, I mean, like. 209 00:10:31.769 --> 00:10:34.620 Brainstorming like when we when I teach brainstorming or something 210 00:10:34.659 --> 00:10:37.679 right it's usually it does sometimes involve talking to somebody 211 00:10:37.759 --> 00:10:40.460 else or just like jotting down ideas like where it's 212 00:10:40.500 --> 00:10:43.480 a low commitment right all those sorts of things like 213 00:10:43.519 --> 00:10:45.840 I just would I just wonder why you would why 214 00:10:45.879 --> 00:10:48.919 a person would need it um about especially about something 215 00:10:48.960 --> 00:10:51.899 that's not like not something you need to research for 216 00:10:52.120 --> 00:10:52.899 right like just. 217 00:10:52.860 --> 00:10:56.100 A sort of concrete example let's say you're going to 218 00:10:56.179 --> 00:10:59.480 do a best man speech or a maid of honor 219 00:10:59.519 --> 00:11:02.820 speech at a wedding. And you're not good with this stuff. 220 00:11:02.899 -->

00:11:05.980 It doesn't come naturally to you. Is it okay to 221 00:11:06.039 --> 00:11:08.490 go to Chachi Petit or Claude or something like that 222 00:11:08.519 --> 00:11:10.029 and go, I've got to give this speech. Let me 223 00:11:10.070 --> 00:11:12.090 tell you about this person. Some of our childhood memories. 224 00:11:12.509 --> 00:11:14.570 I'm struggling to come up with a theme or an idea. 225 00:11:14.590 --> 00:11:16.750 I want to have a witty joke. I want to like, 226 00:11:17.210 --> 00:11:18.909 you know, do it in two or three minutes. Can 227 00:11:18.950 --> 00:11:22.240 you help me come up with something? I think it's 228 00:11:22.299 --> 00:11:25.240 okay to use that to get, to generate an idea 229 00:11:25.360 --> 00:11:27.820 and maybe come up with something that's like, oh, that's good. 230 00:11:28.279 --> 00:11:32.639 But then you use your own judgment to refine, edit, 231 00:11:32.740 --> 00:11:36.789 and put your own mark on it. So what's wrong 232 00:11:36.809 --> 00:11:39.110 with that? The context matters so much, right? 233 00:11:39.450 --> 00:11:41.769 And like, like one of the words you, I forget 234 00:11:41.789 --> 00:11:43.750 if it was in that specific question or earlier, you 235 00:11:43.769 --> 00:11:47.110 talked about using AI as a source for ideas and, 236 00:11:48.090 --> 00:11:50.289 And I think that's where I get worried, especially in 237 00:11:50.309 --> 00:11:52.330 the case of the best man speech at a wedding 238 00:11:52.830 --> 00:11:55.960 where like this is a very personal, like ought to 239 00:11:56.000 --> 00:12:01.799 be an intimate, like organic expression of what this relationship is. 240 00:12:02.659 --> 00:12:04.860 And I do think like I think I have a 241 00:12:04.899 --> 00:12:09.159 softer take on this stuff than Amanda does. But I 242 00:12:09.250 --> 00:12:13.419 think that. the idea of helping to like letting AI 243 00:12:13.480 --> 00:12:16.379 help us to refine things that are genuinely original with 244 00:12:16.460 --> 00:12:20.379 us feels to me very different from giving it kind 245 00:12:20.419 --> 00:12:22.399 of the first stab, if that makes sense. 246 00:12:22.460 --> 00:12:25.240 I mean, if, if you, if you told your, if, 247 00:12:25.759 --> 00:12:28.379 if I were a bride and my, my maid of 248 00:12:28.419 --> 00:12:33.080 honor said, I use chat GPT to, to like write 249 00:12:33.340 --> 00:12:37.279 your speech, I would be offended because, I mean, or 250 00:12:37.320 --> 00:12:38.440 another example. Here's another example. 251 00:12:38.480 --> 00:12:39.419 First of all, I've never admitted. 252 00:12:40.720 --> 00:12:43.320 Right, but the thing is, I mean, especially something like 253 00:12:43.340 --> 00:12:45.200 that where you know, or like the stakes are actually 254 00:12:45.220 --> 00:12:47.480 pretty high. And I don't mean like the pressure to 255 00:12:47.519 --> 00:12:50.139 deliver a perfect speech, but the stakes are high in 256 00:12:50.159 --> 00:12:52.840 terms of human connection and human relationship. Like your best friend, 257 00:12:52.860 --> 00:12:54.460 if you're the best man, like you're- I. 258 00:12:54.379 --> 00:12:57.769 Have two questions or maybe one. Okay, the first question is, 259 00:12:58.370 --> 00:13:02.330 what's the difference between that and say before AI existed, 260 00:13:02.389 --> 00:13:05.389 but Google still existed, Googling and saying, I need a 261 00:13:05.490 --> 00:13:09.250 funny- icebreaker for a toast I'm going to give at my, 262 00:13:09.269 --> 00:13:12.470 at my brother's wedding or something and having it come 263 00:13:12.549 --> 00:13:15.110 like the internet kind of spur you for ideas. What's 264 00:13:15.129 --> 00:13:19.389 the difference between that? And then I believe that if 265 00:13:19.450 --> 00:13:23.529 you do it right, the quality can be better than 266 00:13:23.590 --> 00:13:25.929 if you just sat there and struggled on your own. 267 00:13:26.090 --> 00:13:27.690 And I think there are some studies I have seen 268 00:13:27.730 --> 00:13:31.230 where they did a creative writing study where people were 269 00:13:31.309 --> 00:13:34.549 using chat GPT or some AI to help them like 270 00:13:34.669 --> 00:13:40.179 prompt some creativity And that versus, um, then not using in, 271 00:13:40.379 --> 00:13:43.039 in the like judgment of the quality of the short 272 00:13:43.120 --> 00:13:48.039 essays was better with the AI assist on like brainstorming 273 00:13:48.070 --> 00:13:52.269 and idea generation. I think that you could have a better, 274 00:13:52.509 --> 00:13:57.230 more funny, more, more touching speech. If you're using the 275 00:13:57.289 --> 00:13:59.450 resources that you have, I'm again, I'm. 276 00:13:59.250 --> 00:14:02.679 Not saying it's not genuine. Like it doesn't come like 277 00:14:03.019 --> 00:14:05.440 you it's you're. And here's the thing. So the difference between, 278 00:14:05.460 --> 00:14:07.830 I don't, here's the thing. When I hear people give 279 00:14:07.870 --> 00:14:10.379 speeches and I hear speeches like a best man speech 280 00:14:10.480 --> 00:14:14.340 or like a graduation speech or something like that, even 281 00:14:14.379 --> 00:14:17.950 before AI, I can see that the person Googled good 282 00:14:18.009 --> 00:14:25.090 graduation speech, whatever, right? And it's crap because it's not organic. 283 00:14:25.309 --> 00:14:29.049 It's not specific to the particular audience. I know of 284 00:14:29.250 --> 00:14:31.769 a college president who, when there was a shooting on campus, 285 00:14:32.200 --> 00:14:38.299 used ChatGPT to write an email to the campus that said, like, 286 00:14:38.460 --> 00:14:42.259 you know, I'm thinking of the people, our thoughts and prayers, whatever. 287 00:14:42.639 --> 00:14:46.250 Students were really, really mad in this situation. place I 288 00:14:46.289 --> 00:14:50.250 heard about rightly so do I want the president like 289 00:14:50.690 --> 00:14:53.360

outsourcing a serious thing like that. 290 00:14:53.149 --> 00:14:55.519 Well it might depend on the president might it not. 291 00:14:55.360 --> 00:14:57.580 No no it doesn't I mean the president of the 292 00:14:57.620 --> 00:15:00.000 university but it doesn't actually because like what you want 293 00:15:00.039 --> 00:15:02.600 in that moment is like a person who cares and 294 00:15:02.639 --> 00:15:04.259 a person who and same thing with the best man 295 00:15:04.279 --> 00:15:07.019 speech and what you're doing when you um I don't 296 00:15:07.039 --> 00:15:09.379 know like I just the more and so I don't 297 00:15:09.500 --> 00:15:10.990 know if there is that much difference but. 298 00:15:10.950 --> 00:15:14.049 It's it's where I struggle I cut you off sorry 299 00:15:14.389 --> 00:15:18.570 no go ahead Um, is like, are you supposed to 300 00:15:18.610 --> 00:15:22.610 just always do this 100% on your own? Or could 301 00:15:22.649 --> 00:15:24.350 you go and be like, I'm going to go to 302 00:15:24.389 --> 00:15:27.450 the library and look at joke books because I can't 303 00:15:27.509 --> 00:15:31.460 come up with a witty one-liner joke. And I like, 304 00:15:31.549 --> 00:15:33.200 I'm going to go and try to find a funny 305 00:15:33.340 --> 00:15:36.080 zinger that I can incorporate into. Where is the line 306 00:15:36.100 --> 00:15:40.500 between that and going, Hey, uh, Claude or, or perplexity 307 00:15:40.580 --> 00:15:43.620 or whatever it is. Um, I am struggling with. Give 308 00:15:43.659 --> 00:15:47.399 me like six ideas for a opening line that I 309 00:15:47.440 --> 00:15:50.799 might incorporate. And it might come back with 50 ideas, 310 00:15:50.799 --> 00:15:52.960 49 of which are dumb. And one you go, I 311 00:15:53.000 --> 00:15:56.159 like that. And it reminds me of me and John 312 00:15:56.250 --> 00:15:58.559 or whatever in this story. And you're like, that's it. 313 00:15:58.590 --> 00:16:00.750 And I'm off to the races. Like that to me 314 00:16:00.789 --> 00:16:03.809 is how I think it's totally fine to use. 315 00:16:05.230 --> 00:16:07.610 I mean, I hate all the speeches that are set 316 00:16:07.629 --> 00:16:09.169 up the way that you were imagining. So that's a 317 00:16:09.230 --> 00:16:10.309 problem in terms. 318 00:16:10.090 --> 00:16:12.230 Of public speakers. 319 00:16:12.330 --> 00:16:15.549 Cheesy speeches that are cliches suck. 320 00:16:15.590 --> 00:16:17.950 Can we just say that as a people, can we 321 00:16:17.990 --> 00:16:19.690 decide that we can move on from the you have 322 00:16:19.710 --> 00:16:23.220 to start your speech with a funny joke? Can we 323 00:16:23.240 --> 00:16:26.620 just say that's not true? But the actual point at hand, 324 00:16:26.840 --> 00:16:29.259 I think, Chris, it feels to me like the way 325 00:16:29.299 --> 00:16:31.750 you're describing this area keeps evolving a little bit. And 326 00:16:31.950 --> 00:16:34.009 for me, and again, I don't think Amanda and I 327 00:16:34.110 --> 00:16:35.750 are on the same page with this, but the way 328 00:16:35.769 --> 00:16:39.970 you just described it, I'm increasingly comfortable with it. If 329 00:16:40.809 --> 00:16:43.169 somebody genuinely is struggling to come up with something worth 330 00:16:43.190 --> 00:16:47.649 saying and gives some real information to one of these 331 00:16:47.799 --> 00:16:52.019 AI things and says, here's the relationship, here are the 332 00:16:52.059 --> 00:16:54.340 things that I think of. Can you help me tie 333 00:16:54.379 --> 00:16:58.059 this together in a coherent way? I guess I have 334 00:16:58.100 --> 00:17:00.500 a harder and harder time having a problem with that 335 00:17:01.019 --> 00:17:05.900 given the degree of human initiative that's going into it. 336 00:17:07.539 --> 00:17:11.119 I think it's wrong to outsource your thinking and the output, 337 00:17:11.200 --> 00:17:17.450 for sure. But I think as a creative source, you're 338 00:17:17.470 --> 00:17:19.589 fooling yourself to think that you can't use it to 339 00:17:19.630 --> 00:17:21.190 make the end product better. 340 00:17:22.150 --> 00:17:23.980 Well, there you go. That's the thing. 341 00:17:24.000 --> 00:17:26.039 You just used the word product. And I just want, 342 00:17:26.059 --> 00:17:27.099 that's what I was going to say. 343 00:17:27.180 --> 00:17:29.500 Product being whatever you're doing, the speech, the whatever. 344 00:17:29.519 --> 00:17:30.680 But I was going to say you used the word 345 00:17:30.799 --> 00:17:32.619 better because. Well, right. 346 00:17:32.700 --> 00:17:33.099 That too. 347 00:17:33.279 --> 00:17:37.289 What counts as good, better, or best really depends on 348 00:17:37.309 --> 00:17:39.250 the context. It depends on what kind of thing it 349 00:17:39.349 --> 00:17:40.750 is and what the purpose of the thing is. 350 00:17:41.109 --> 00:17:43.869 And also like in that case with us, a best 351 00:17:43.890 --> 00:17:48.599 man speech is the, I mean, is the goal to 352 00:17:48.799 --> 00:17:51.940 be like, to produce this, the like, this like sort 353 00:17:51.960 --> 00:17:55.750 of professional sounding, um, best witty. 354 00:17:55.789 --> 00:17:57.589 We've all been to weddings where the maid of honor 355 00:17:57.609 --> 00:18:00.029 drones on and it's like, when we were in sophomores 356 00:18:00.109 --> 00:18:03.130 in college and we remember getting choked up. 357 00:18:03.369 --> 00:18:03.569 Yep. 358 00:18:03.750 --> 00:18:03.990 Yeah. 359 00:18:04.109 --> 00:18:04.910 But they're terrible. 360 00:18:04.930 --> 00:18:06.569 I don't want to hear that. I want to hear 361 00:18:06.630 --> 00:18:10.470 somebody deliver something that is hilarious. 362 00:18:10.549 --> 00:18:13.980 I love them to start with Western Webster's dictionary defines 363 00:18:14.019 --> 00:18:15.160 friendship as, I mean. 364 00:18:15.180 --> 00:18:17.779 That's what you come up with unless you're using AI 365 00:18:17.799 --> 00:18:19.619 to

help you say this is stupid. 366 00:18:20.279 --> 00:18:22.039 But I'm just saying like, I mean, I don't know. 367 00:18:22.059 --> 00:18:24.799 I think like a person should stop. Speak from the heart, 368 00:18:24.859 --> 00:18:28.460 to use a cliche. So that's way better than something 369 00:18:28.529 --> 00:18:31.950 that's been perfected and tailored and refined in all these 370 00:18:31.970 --> 00:18:33.650 sorts of ways. And that could be at any wedding. 371 00:18:33.730 --> 00:18:35.869 It's so... I don't know. I don't like it. I 372 00:18:35.910 --> 00:18:37.630 also think... Let me answer your question about the difference 373 00:18:37.670 --> 00:18:40.630 between sources going to the library or looking on the 374 00:18:40.650 --> 00:18:44.200 internet at Reddit posts and all these different sources versus AI. 375 00:18:44.779 --> 00:18:47.099 There's something different that happens in the brain in that case. 376 00:18:47.359 --> 00:18:49.400 Which is when you look at lots of different sources, 377 00:18:49.559 --> 00:18:54.109 you are... cognitively reading these things, you're sifting through, you're 378 00:18:54.130 --> 00:18:55.950 making decisions, you're being, I like this, I like this, 379 00:18:55.990 --> 00:18:57.930 I don't like this, whatever. What happens when you use 380 00:18:57.990 --> 00:18:59.950 AI is that you've skipped that step. So you're not 381 00:19:00.109 --> 00:19:03.740 actually doing that cognitive work. So it doesn't express you 382 00:19:03.779 --> 00:19:04.380 in the same way. 383 00:19:04.519 --> 00:19:06.160 I think it's how you prompt it though. I think 384 00:19:06.200 --> 00:19:08.000 it's how you're setting it up. I think if you're 385 00:19:08.039 --> 00:19:11.380 really good at using AI, you're not just going, give 386 00:19:11.440 --> 00:19:15.519 me the answer. You're doing what I'm describing, which is like, um, 387 00:19:16.200 --> 00:19:18.779 trying to maybe, sorry, I don't want to cut you off. No, 388 00:19:18.819 --> 00:19:19.099 go ahead. 389 00:19:19.440 --> 00:19:21.720 I just wonder if maybe part of the disconnect here 390 00:19:22.500 --> 00:19:26.259 is that at least for me, one of the worries 391 00:19:26.880 --> 00:19:29.940 is that what you're describing, the thing that you're, you're characterizing, Chris, 392 00:19:29.960 --> 00:19:32.349 that I said, I think I'm okay with this way 393 00:19:32.390 --> 00:19:36.710 of doing it is also seems likely to be this tiny, tiny, 394 00:19:36.789 --> 00:19:39.430 tiny percentage of the population. Like, so like maybe it 395 00:19:39.549 --> 00:19:44.099 can be done well, but at a large scale, it's 396 00:19:44.119 --> 00:19:47.019 just not gonna be done well. Everything's gonna be... I 397 00:19:47.259 --> 00:19:48.170 do think there is. 398 00:19:48.069 --> 00:19:51.319 A real uh issue with the more people use the 399 00:19:51.460 --> 00:19:56.440 ai the more um like homogenized i was gonna say 400 00:19:56.480 --> 00:20:00.140 dilution

like everything is gonna start sounding the same well. 401 00:20:00.069 --> 00:20:01.549 It already is like if you read anything on the 402 00:20:01.589 --> 00:20:04.250 internet right it's not this it's this right yeah. 403 00:20:04.619 --> 00:20:07.460 But I think that's, I mean, here's a very old 404 00:20:07.519 --> 00:20:09.640 idea that goes back at least as far as the 405 00:20:09.680 --> 00:20:12.500 great Greek philosophers, that if you want to know what 406 00:20:12.779 --> 00:20:16.059 is good for something, you need to know what it's for, right? So, like, 407 00:20:16.079 --> 00:20:18.200 what makes a good knife a good knife is that, like, 408 00:20:18.259 --> 00:20:22.160 it can cut well, because that's what a knife is for. And, like, 409 00:20:22.240 --> 00:20:25.990 different activities are for different kinds of things. And so, like, 410 00:20:26.029 --> 00:20:30.289 a best man speech is for a celebration of an 411 00:20:30.410 --> 00:20:32.910 intimate kind of friendship. Like, it's not. 412 00:20:34.819 --> 00:20:39.900 For like entertainment. It's not supposed to be an amazing 413 00:20:40.259 --> 00:20:43.339 set done by a standup comic. It's supposed to be 414 00:20:43.440 --> 00:20:46.019 from the heart. And so the more you bring in 415 00:20:46.079 --> 00:20:49.960 an artificial source or quotes from stupid joke books you 416 00:20:49.980 --> 00:20:52.119 found in the library, the less you're doing what the 417 00:20:52.160 --> 00:20:53.059 thing is supposed to be. 418 00:20:53.680 --> 00:20:55.339 And I also want to say there is a thing 419 00:20:55.380 --> 00:20:56.980 I just wanted, this is a sort of a smaller 420 00:20:57.019 --> 00:21:00.000 point about writing or, or if you're speaking at whatever, 421 00:21:00.059 --> 00:21:01.700 but I assume you've sort of written it before. Um, 422 00:21:02.180 --> 00:21:04.119 even at the refining stage, if you use AI to 423 00:21:04.160 --> 00:21:08.140 quote refine it, um, that changes meaning. If you change 424 00:21:08.180 --> 00:21:10.740 the words, the sequence of words in a sentence, even 425 00:21:10.789 --> 00:21:13.210 if it's generally the same, it does change the meaning. 426 00:21:13.250 --> 00:21:16.170 It changes your voice. It changes the tone. It does 427 00:21:16.190 --> 00:21:18.630 all these things like words matter, the order they're in matter. 428 00:21:18.869 --> 00:21:21.789 So what I'm saying is I'm not like, what I'm 429 00:21:21.809 --> 00:21:24.650 saying is that it will probably make it sound more professional, 430 00:21:24.769 --> 00:21:27.069 but that's not what I don't even want that as 431 00:21:27.130 --> 00:21:27.839 a writing teacher. 432 00:21:28.190 --> 00:21:30.559 You are a PhD in English and you're very, a 433 00:21:30.619 --> 00:21:33.289 very great, you're a great writer. If somebody came to 434 00:21:33.309 --> 00:21:35.210 you and they said, I wrote this maid of honor 435 00:21:35.250 --> 00:21:37.210 speech and I need some help. Can you give me 436 00:21:37.250 --> 00:21:39.710 some feedback? Would you say, absolutely not. It needs to 437 00:21:39.730 --> 00:21:41.890 be in your voice and I can't help you professionalize 438 00:21:41.930 --> 00:21:43.170 it or sound better? 439 00:21:43.289 --> 00:21:44.990 No, I wouldn't say that, but I wouldn't rewrite it 440 00:21:45.049 --> 00:21:47.430 for them. I would give them feedback and I'd say, yeah, 441 00:21:48.190 --> 00:21:51.509 you're droning on a little bit. Stop crying. 442 00:21:51.529 --> 00:21:53.789 So let me ask you about yourself together. So the 443 00:21:53.829 --> 00:21:56.710 broader question, right, is about a healthy use of AI. 444 00:21:56.859 --> 00:21:59.220 I mean, I'd be curious to know. That's why I 445 00:22:00.440 --> 00:22:02.839 asked ChatGPT to invent a cocktail for me. Is because 446 00:22:02.859 --> 00:22:07.710 I wanted to, like, it seems to me, I'm totally 447 00:22:07.730 --> 00:22:09.190 okay with that. I'm not going to lose any sleep 448 00:22:09.430 --> 00:22:10.950 over that choice. I don't think I need to go 449 00:22:10.970 --> 00:22:14.369 to confession or anything like that. I like the cocktail, personally. 450 00:22:14.390 --> 00:22:14.990 I'm curious. 451 00:22:15.049 --> 00:22:19.230 I like it a lot myself. I'm curious, like, where 452 00:22:19.250 --> 00:22:21.279 does that land for you? The cocktail? Yeah. 453 00:22:22.019 --> 00:22:24.720 So, yeah, it's lower on my list of concerns. I mean, 454 00:22:24.740 --> 00:22:28.720 I think for me, the best thing about AI, and again, 455 00:22:28.880 --> 00:22:32.039 and this isn't something, well, I'll just say this first. 456 00:22:33.500 --> 00:22:36.859 is like its ability to sift through a ton of data, right? And, 457 00:22:37.140 --> 00:22:39.950 and like, I don't know, find the thing you ask 458 00:22:39.970 --> 00:22:41.569 it to find or whatever. Like that's not, so I 459 00:22:41.589 --> 00:22:45.130 think there are, there are potentials there. But it should 460 00:22:45.170 --> 00:22:49.599 never substitute for experience or certainly for like humans doing 461 00:22:49.819 --> 00:22:53.640 critical thinking or, values and moral judgments, et cetera. But 462 00:22:53.799 --> 00:22:55.339 in the cocktail case, like it's not the end of 463 00:22:55.359 --> 00:22:57.599 the world. What if you only did this? What if 464 00:22:57.640 --> 00:23:00.140 you only use AI to create cocktails and you lost 465 00:23:00.319 --> 00:23:03.650 out on the human, like, like reading about what we 466 00:23:03.690 --> 00:23:05.569 do on this podcast, where we talk about the history 467 00:23:05.650 --> 00:23:07.849 or lack of, um, about the cocktails. 468 00:23:07.869 --> 00:23:09.470 It was a history of history. It was short. 469 00:23:09.490 --> 00:23:10.410 Right. It was today. 470 00:23:10.430 --> 00:23:11.190 It was like an hour ago. 471 00:23:11.250 --> 00:23:11.470 Right. 472 00:23:12.049 --> 00:23:14.440 Um, or like, or like the, the rich,

you know, 473 00:23:14.460 --> 00:23:17.079 the sort of rich history of whatever, or like the 474 00:23:17.140 --> 00:23:19.480 pleasure of being like yourself. 475 00:23:19.839 --> 00:23:20.079 Okay. 476 00:23:20.099 --> 00:23:22.720 I like this in this cocktail book. I read, you know, 477 00:23:22.940 --> 00:23:24.940 here's how you make a Manhattan, but I think I'm 478 00:23:24.960 --> 00:23:26.880 going to do this twist. Like I'm going to change 479 00:23:26.900 --> 00:23:28.970 it this way, you know, and you can do that with, 480 00:23:29.150 --> 00:23:31.190 with the NAI production too. But I just mean like, 481 00:23:31.670 --> 00:23:34.829 there's something, I don't know. Like I, I hear people 482 00:23:34.869 --> 00:23:36.529 say like, Oh, I use AI to create a meal 483 00:23:36.569 --> 00:23:38.230 plan for me for my week. Or I use AI 484 00:23:38.289 --> 00:23:40.589 to do these like basic life skills. And to me, 485 00:23:40.630 --> 00:23:45.269 that's like infantilizing in the sense that you're like, Why 486 00:23:45.309 --> 00:23:49.009 can't you do it yourself? Again, especially if it's... Here's 487 00:23:49.029 --> 00:23:52.349 what I would say. Hold on. Especially if it's something pleasurable. 488 00:23:53.150 --> 00:23:56.930 Why would you forfeit that, like making a cocktail? You 489 00:23:56.950 --> 00:23:57.380 know what I mean? 490 00:23:58.039 --> 00:23:58.619 Earlier this year. 491 00:23:58.660 --> 00:24:00.359 Or writing something creative like a podcast. 492 00:24:01.420 --> 00:24:03.599 Earlier this year, I started to get in better shape 493 00:24:03.640 --> 00:24:06.380 and I used AI. I was having this long conversation 494 00:24:06.400 --> 00:24:08.900 with a Cheshire PT, telling it what I was eating, 495 00:24:08.980 --> 00:24:12.480 what my workouts were, and I found it incredibly helpful 496 00:24:12.609 --> 00:24:16.359 to track... like my protein intake. It would give me 497 00:24:16.420 --> 00:24:18.809 suggestions on workouts. I would tell it, this is, these 498 00:24:18.849 --> 00:24:20.569 are the weights and the equipment that I have in 499 00:24:20.589 --> 00:24:24.829 my company gym, um, helped me design a meal based 500 00:24:24.849 --> 00:24:27.829 on what I have in the refrigerator. That's, you know, healthy, whatever. 501 00:24:28.349 --> 00:24:32.049 It was exceptionally helpful. And I was like, man, this 502 00:24:32.109 --> 00:24:34.329 is like people, everybody should be doing this. This is 503 00:24:34.349 --> 00:24:37.920 like making me better as far, like I'm healthier or whatever. Yeah. 504 00:24:38.059 --> 00:24:40.559 You're putting like a personal trainer and nutritionist out of work, 505 00:24:40.640 --> 00:24:40.960 but yeah. 506 00:24:41.690 --> 00:24:41.950 Yeah. 507 00:24:42.150 --> 00:24:43.230 And that's what's going to happen. 508 00:24:43.269 --> 00:24:43.869 Sure. Yeah. 509 00:24:43.990 --> 00:24:46.339 But I think I see it. I don't know. I don't. 510 00:24:46.720 --> 00:24:49.960

Well, that's, that's not the same thing. 511 00:24:50.119 --> 00:24:53.740 Part of where, why I'm not super uncomfortable with say 512 00:24:53.880 --> 00:24:57.339 having the, having a robot create a meal plan is 513 00:24:57.480 --> 00:25:00.519 like most of us can't afford personal trainers and nutrition. 514 00:25:01.220 --> 00:25:02.980 I can't afford, my wife and I can't afford a 515 00:25:03.089 --> 00:25:05.490 personal assistant to help us with the little tasks that 516 00:25:05.509 --> 00:25:08.160 get dropped in a typical week. But like it's, we 517 00:25:08.180 --> 00:25:11.289 have a lot going on and like, it's a chore, um, that, 518 00:25:11.450 --> 00:25:14.109 that my wife has typically taken on to figure out 519 00:25:14.170 --> 00:25:16.950 what are this meals or this week's meals. Um, and 520 00:25:17.289 --> 00:25:20.960 I think to, uh, at least occasionally, or maybe even 521 00:25:21.000 --> 00:25:24.240 consistently ask for help from a computer, it's, it's very, 522 00:25:24.259 --> 00:25:27.619 it's hard for me to articulate a reason why that 523 00:25:27.660 --> 00:25:29.960 would be objectionable. And by the way, I'm not saying 524 00:25:29.980 --> 00:25:33.140 that I'm sure that I'd I'm kind of playing devil's advocate. 525 00:25:33.200 --> 00:25:35.059 I'm not sure that I like that conclusion. I just 526 00:25:35.079 --> 00:25:35.769 can't see what's wrong with it. 527 00:25:35.779 --> 00:25:38.140 Amanda, if you were ever diagnosed, and God forbid, with 528 00:25:38.200 --> 00:25:40.910 a rare disease, would you use AI to help learn 529 00:25:40.950 --> 00:25:44.589 about it and Maybe research it. Because people are doing 530 00:25:44.630 --> 00:25:46.809 this very successfully. Yeah. 531 00:25:46.910 --> 00:25:48.819 I mean, here's the thing, though. In that case, why 532 00:25:48.859 --> 00:25:52.980 wouldn't I read papers published by doctors on it? I 533 00:25:53.000 --> 00:25:53.319 don't understand. 534 00:25:53.339 --> 00:25:57.119 Because the AI can do it way better than the doctors. 535 00:25:57.119 --> 00:25:57.240 100%. 536 00:25:57.240 --> 00:25:57.700 Absolutely. 537 00:25:57.720 --> 00:25:59.849 We know that it's making a ton of mistakes, first 538 00:25:59.869 --> 00:26:01.109 of all. But even if it wasn't. 539 00:26:01.130 --> 00:26:04.549 The breakthroughs in medicine with AI are astonishing. And they 540 00:26:04.609 --> 00:26:05.529 will be astonishing. 541 00:26:05.549 --> 00:26:08.839 But I'm talking about information. I'm talking about how I 542 00:26:08.880 --> 00:26:10.160 get my information about that thing. 543 00:26:10.180 --> 00:26:13.029 As if doctors don't make mistakes. I mean, I think 544 00:26:13.109 --> 00:26:15.589 it's true. 545 00:26:15.670 --> 00:26:18.230 I know, of course. Here's the thing. I want to 546 00:26:18.289 --> 00:26:21.440 know who the source is on any question of value 547 00:26:21.480 --> 00:26:23.839 or importance. And AI does not tell me what that 548 00:26:23.880

--> 00:26:26.680 source is in the sense of like it presents itself 549 00:26:26.839 --> 00:26:31.210 as an oracle, as an objective source. piece of work information. 550 00:26:31.230 --> 00:26:33.630 It can, I know, but I'm saying if people just 551 00:26:33.730 --> 00:26:36.650 read AI, like I just would never, it has always 552 00:26:36.710 --> 00:26:38.930 mattered to me to go through Google and be like, Oh, 553 00:26:38.950 --> 00:26:40.390 I'm not going to read the Reddit posts. I'm going 554 00:26:40.410 --> 00:26:42.569 to read the, the Mayo Clinic or whatever. I'm going 555 00:26:42.589 --> 00:26:46.109 to read the, you know, whatever it is. Um, when I'm, uh, 556 00:26:46.170 --> 00:26:48.470 doom searching, there's a lot here and. 557 00:26:48.339 --> 00:26:50.700 We're closing and I needed to move on to a 558 00:26:50.700 --> 00:26:53.680 new topic, but like, is there, I don't know how 559 00:26:53.720 --> 00:26:56.299 I would answer this, but is there a clear answer 560 00:26:56.339 --> 00:26:59.200 to the question we started with? What does healthy use 561 00:26:59.220 --> 00:27:01.039 of AI look like? I mean, it sounds like your 562 00:27:01.059 --> 00:27:03.490 answer is there's no such thing. Is that too strong? 563 00:27:03.650 --> 00:27:06.150 I mean, I think like I don't want to speak 564 00:27:06.230 --> 00:27:11.150 for sort of scientific. experimentation and discoveries and things like that. 565 00:27:11.170 --> 00:27:12.670 You know what I mean? But as like a person 566 00:27:12.710 --> 00:27:14.289 living in the world, who's not in that field and 567 00:27:14.309 --> 00:27:16.769 like just going through my life, I don't, I'm don't 568 00:27:17.029 --> 00:27:19.309 plan on using it. I don't use it now unless 569 00:27:19.490 --> 00:27:21.490 I'm forced to, or I don't know that I'm using it, 570 00:27:21.730 --> 00:27:22.390 which is possible. 571 00:27:22.690 --> 00:27:22.910 Right. 572 00:27:23.230 --> 00:27:24.970 But I mean, like I'm, I don't like that it 573 00:27:25.009 --> 00:27:26.769 comes up first in a Google search. Like I, like 574 00:27:26.809 --> 00:27:28.970 I said, I like to know who those sources are, but, um, 575 00:27:29.029 --> 00:27:31.549 and pick my own source. But, um, so I don't, 576 00:27:32.269 --> 00:27:34.329 I just, I don't know what, I don't know what 577 00:27:34.390 --> 00:27:38.910 problem it's fixing again. Um, health, medical discoveries aside, those 578 00:27:38.950 --> 00:27:40.930 sorts of things. I'm not saying those don't exist, but 579 00:27:40.970 --> 00:27:44.009 for me in my day-to-day life, it's not fixing any problems. 580 00:27:44.970 --> 00:27:48.569 I know let's wrap up, but I do think there's many, 581 00:27:48.630 --> 00:27:51.809 many use cases that it will help you. I've referenced 582 00:27:51.829 --> 00:27:54.029 this before. I think on this podcast where like my 583 00:27:54.230 --> 00:27:56.839 garage door broke, I literally took a picture of the 584 00:27:56.880 --> 00:27:58.980 garage door model type. And I was like, this is 585 00:27:59.099 --> 00:28:01.609 what's happening. This is a noise it's making. it immediately 586 00:28:01.630 --> 00:28:04.309 was like, here's the three possible fixes. Go through them 587 00:28:04.349 --> 00:28:06.549 in this order. And it fixed it immediately and saved 588 00:28:06.569 --> 00:28:09.289 me probably $ 600 calling a garage door person. 589 00:28:09.470 --> 00:28:12.880 Or probably saved you time. And I'm saying, yes, it did. 590 00:28:12.910 --> 00:28:15.730 Sifting through other information on the internet or videos you 591 00:28:15.759 --> 00:28:16.099 could have found. 592 00:28:16.119 --> 00:28:19.519 Or something to me. I'm thinking that's the heart of 593 00:28:19.579 --> 00:28:21.079 my answer to this, at least for now. I reserve 594 00:28:21.099 --> 00:28:23.259 the right to change my mind as the robots increasingly 595 00:28:23.299 --> 00:28:25.559 take over. And as we all keep thinking about this. 596 00:28:25.599 --> 00:28:31.099 But I think that In context where efficiency really is 597 00:28:31.119 --> 00:28:33.500 a good thing, and it's not always, and I think 598 00:28:33.519 --> 00:28:36.140 this is, Chris, a place where I maybe am reluctant 599 00:28:36.160 --> 00:28:39.980 to agree with some of your suggestions. Efficiency isn't always 600 00:28:40.680 --> 00:28:43.130 what we make it out to be. Understood. But in 601 00:28:43.210 --> 00:28:46.509 context where efficiency is genuinely valuable, then I think I'm 602 00:28:46.549 --> 00:28:47.269 very comfortable using it. 603 00:28:53.000 --> 00:28:56.480 That's fine in many contexts like the garage door fix 604 00:28:56.619 --> 00:29:00.849 versus judgment. I don't want to outsource my judgment on 605 00:29:00.950 --> 00:29:04.589 a speech or whatever. But I'm still going to use 606 00:29:04.630 --> 00:29:07.829 it to brainstorm and come up with ideas. I think 607 00:29:07.849 --> 00:29:10.839 it matters what the thing is. and what they, to 608 00:29:12.339 --> 00:29:14.059 talk like a philosopher, what the goods in terms of 609 00:29:14.079 --> 00:29:16.160 the practice are, like what are the things that make 610 00:29:16.299 --> 00:29:20.339 this good? And does AI contribute to those goods or 611 00:29:20.579 --> 00:29:21.500 does it diminish them? 612 00:29:21.880 --> 00:29:25.640 And if I could just say, yeah, absolutely. The question 613 00:29:25.660 --> 00:29:27.339 of efficiency, I think is really good because I do 614 00:29:27.400 --> 00:29:29.480 think there's an issue where we look at a problem 615 00:29:29.509 --> 00:29:32.130 where like, we're all too busy. And instead of saying, 616 00:29:32.210 --> 00:29:34.529 how could we change the structure of our society? So 617 00:29:34.549 --> 00:29:37.490 we're not so busy. let's do this thing that we 618 00:29:37.569 --> 00:29:42.990 outsource to. To me, is it a problem that we 619 00:29:43.039 --> 00:29:46.079 don't have time to cook food,

one of the most 620 00:29:46.160 --> 00:29:50.740 basic things to human life, right? Yes, it is a problem, right? 621 00:29:51.380 --> 00:29:53.059 So instead of saying, how could we make it so 622 00:29:53.099 --> 00:29:55.619 that more people have that time and the money to 623 00:29:55.680 --> 00:30:01.869 buy groceries and live a healthier life, we say, we 624 00:30:01.890 --> 00:30:04.130 sort of say, well, let's have this technological fix. And 625 00:30:04.170 --> 00:30:06.569 to me, like that, I'm not interested in that. I'm 626 00:30:06.650 --> 00:30:08.390 interested in as much as I can. And again, I 627 00:30:08.529 --> 00:30:11.170 don't have control over society, but in my own life, um, 628 00:30:11.230 --> 00:30:12.690 what control I do have, I want to try to 629 00:30:12.730 --> 00:30:15.339 go the other way and actually try to like make my, 630 00:30:15.539 --> 00:30:16.980 change my life so that I'm living in a way 631 00:30:17.019 --> 00:30:17.420 that's healthier. 632 00:30:17.440 --> 00:30:21.500 That's well said. I, this isn't, I am curious, maybe 633 00:30:21.579 --> 00:30:23.200 we should do a part three of this at some point, 634 00:30:23.220 --> 00:30:25.000 because I am, I am curious, like, I wonder if, 635 00:30:25.569 --> 00:30:29.789 if hypothetically AI could do so much labor that we 636 00:30:29.829 --> 00:30:32.690 do now, like factory jobs and menial tasks and things 637 00:30:32.710 --> 00:30:36.210 like that, that it allowed every person in America to 638 00:30:36.250 --> 00:30:41.150 have a universal basic income of a certain amount, would 639 00:30:41.190 --> 00:30:44.029 you be acceptable? Would you be okay with that? I'm serious. 640 00:30:44.049 --> 00:30:46.069 You're throwing this in now at minute. 641 00:30:46.089 --> 00:30:47.049 Sorry. 642 00:30:47.230 --> 00:30:48.529 I have so much to say about that. 643 00:30:48.549 --> 00:30:51.490 But I do think that's where we're headed. The arc 644 00:30:51.549 --> 00:30:55.289 of humanity's progress with technology is, is going to get there. 645 00:30:55.680 --> 00:30:57.029 But the problem is you're saying that, nope. 646 00:30:57.049 --> 00:30:57.630 I don't want that. 647 00:30:57.829 --> 00:30:59.069 No, I'm not. No, no, no, I'm not saying that. 648 00:30:59.089 --> 00:31:01.349 I want everyone to be frustrated by doing shit every day. 649 00:31:01.369 --> 00:31:02.710 No, no, no. She definitely didn't say that. 650 00:31:02.730 --> 00:31:03.509 I didn't say that. 651 00:31:03.769 --> 00:31:04.890 I didn't say that. 652 00:31:04.970 --> 00:31:07.390 I think that you guys need anywhere from 30 to 653 00:31:07.390 --> 00:31:08.190 60 seconds. 654 00:31:08.430 --> 00:31:08.750 We do. 655 00:31:08.809 --> 00:31:10.730 To simmer down a little bit. So why don't we 656 00:31:10.769 --> 00:31:15.569 go ahead and hear a word from whatever sponsors the algorithms, 657 00:31:15.710 --> 00:31:19.130 the robots give us today. So there'll be more 3Q3D 658 00:31:19.170 --> 00:31:25.619 coming up right after this. And we are back with 659 00:31:25.859 --> 00:31:27.839 question and drink number two now that folks, I think, 660 00:31:27.859 --> 00:31:30.809 have calmed down a little bit here at 3Q3D headquarters. 661 00:31:31.569 --> 00:31:33.829 Are you guys feeling all right? Are we good? 662 00:31:33.970 --> 00:31:36.140 I'm going to need to talk to ChatGPT about it 663 00:31:36.160 --> 00:31:36.799 and see how I feel. 664 00:31:38.150 --> 00:31:38.630 I feel fine. 665 00:31:39.670 --> 00:31:42.450 Dr. Adams, I believe the question and the drink are yours. 666 00:31:43.079 --> 00:31:45.900 So for this episode, I brought back an old favorite, 667 00:31:46.460 --> 00:31:47.480 which is the sidecar. 668 00:31:47.559 --> 00:31:49.539 I knew it was the sidecar. I could see it. 669 00:31:49.599 --> 00:31:49.880 I know. 670 00:31:50.119 --> 00:31:50.920 It's a beautiful drink. 671 00:31:50.940 --> 00:31:52.359 I think this is the third time I've had it 672 00:31:52.460 --> 00:31:54.640 on the show, but I love it so much. It's 673 00:31:54.660 --> 00:31:59.359 the perfect summer drink. So it's cognac and Cointreau and 674 00:31:59.420 --> 00:32:04.390 lemon juice with a sugar rim, and it's the best. 675 00:32:04.589 --> 00:32:07.630 It's like a wonderful, it's like a bright... yeah drink 676 00:32:07.670 --> 00:32:08.849 it drink it they're drinking. 677 00:32:08.589 --> 00:32:12.049 It all right let's go it's got a sugar that's delightful. 678 00:32:12.650 --> 00:32:16.420 Yeah it's it's the best so the. 679 00:32:15.980 --> 00:32:20.400 Perfect drink for a summer day it's just so refreshing. 680 00:32:19.920 --> 00:32:22.660 Yeah it's hard to find yes it's hard to find 681 00:32:22.720 --> 00:32:26.289 a brown liquor a dark liquor that's good for summer 682 00:32:26.670 --> 00:32:28.470 and if you are a fan as i am of 683 00:32:28.509 --> 00:32:32.650 darker liquor um yeah This is a good choice. 684 00:32:32.710 --> 00:32:33.269 That's really good. 685 00:32:33.490 --> 00:32:35.410 Yeah, and it's an old, just, you know, I've said 686 00:32:35.430 --> 00:32:37.289 this on the podcast before, but it's like an early 687 00:32:37.710 --> 00:32:40.759 20th century, invented at Harry's New York Bar in Paris 688 00:32:41.980 --> 00:32:44.980 by the bartender there and named for a patron, a 689 00:32:45.000 --> 00:32:48.400 guy who would come in riding in the sidecar of 690 00:32:48.460 --> 00:32:50.180 a chauffeur-driven motorcycle. 691 00:32:50.200 --> 00:32:50.299 Yeah. 692 00:32:51.269 --> 00:32:51.509 Awesome. 693 00:32:51.789 --> 00:32:52.009 I know. 694 00:32:52.289 --> 00:32:53.650 I want to know who that guy is. I want 695 00:32:53.670 --> 00:32:54.190 to talk to that guy. 696 00:32:54.210 --> 00:32:55.789 You would drink this drink. That's funny. Yeah. 697 00:32:55.890 --> 00:32:56.250 I don't know. 698 00:32:56.349 --> 00:32:57.670 The only downside with this drink is I feel like 699 00:32:57.690 --> 00:32:59.349

I'm going to get an instant canker sore on my 700 00:32:59.410 --> 00:33:01.230 tongue with there's so much sugar in it. 701 00:33:02.089 --> 00:33:04.089 Well, there's no, it's just the sugar rim though. 702 00:33:04.150 --> 00:33:04.490 That's it. 703 00:33:04.690 --> 00:33:06.809 There's no sugar in the, there's no like simple syrup 704 00:33:06.869 --> 00:33:09.230 or anything. So the actual drink is not, but it's, 705 00:33:09.430 --> 00:33:12.039 it's Cointreau and cognac. So it's not like not sweet, 706 00:33:12.079 --> 00:33:12.519 but it's not. 707 00:33:12.539 --> 00:33:12.930 Yeah. 708 00:33:13.210 --> 00:33:13.559 Okay. 709 00:33:13.599 --> 00:33:14.099 Very good. All right. 710 00:33:14.119 --> 00:33:15.259 You're off to a great start. 711 00:33:15.319 --> 00:33:18.700 Okay. Well, It's all downhill from here. Now, here's my question. 712 00:33:18.740 --> 00:33:23.690 And actually, your 11th hour Hail Mary toss at the 713 00:33:23.829 --> 00:33:28.650 end of the last segment about sort of general how 714 00:33:28.730 --> 00:33:30.750 are we as a society going to how's AI going 715 00:33:30.769 --> 00:33:32.609 to change how we live as a society does tee 716 00:33:32.630 --> 00:33:35.150 this up. And so my question is this. It's actually about. 717 00:33:36.450 --> 00:33:40.480 our society, but specifically American democracy and how, and here's 718 00:33:40.519 --> 00:33:43.779 my question, does a population that's kind of reliant on AI, 719 00:33:44.059 --> 00:33:47.630 and I would add in, you know, AI as a, well, 720 00:33:47.750 --> 00:33:50.910 let me finish the question first. Does a population that's 721 00:33:50.950 --> 00:33:54.509 reliant on AI pose a threat to the future of democracy? Like, 722 00:33:54.549 --> 00:33:59.039 can we have a democracy built of people who, who 723 00:33:59.140 --> 00:34:03.380 are maybe over-reliant on AI or, or, or just reliant. Yes. 724 00:34:03.480 --> 00:34:05.019 That part of it. But then I would also say, 725 00:34:05.059 --> 00:34:09.050 how does that work? Um, you know, like, uh, what 726 00:34:09.070 --> 00:34:12.710 happens when the robots put everybody out of work and, um, 727 00:34:12.789 --> 00:34:16.190 that has not been great for democracies in the past when, um, 728 00:34:16.429 --> 00:34:17.909 big changes like that have happened. 729 00:34:17.949 --> 00:34:19.429 So not great for people in general. 730 00:34:19.489 --> 00:34:20.010 For sure. 731 00:34:20.409 --> 00:34:20.590 Yeah. 732 00:34:20.710 --> 00:34:24.269 I mean, but, and then downstream from that democracies, this. 733 00:34:24.150 --> 00:34:27.139 Is, here's a real question before we get into that, that, Because, 734 00:34:27.179 --> 00:34:28.320 I mean, I think you know a lot more about 735 00:34:28.340 --> 00:34:29.840 the history of these things than I do. Like, is 736 00:34:29.880 --> 00:34:33.059 the issue that there is a, when there's mass unemployment 737 00:34:33.079 --> 00:34:35.579 and whatnot, that there's

a unique threat to democracy or 738 00:34:35.619 --> 00:34:38.639 that it's just a massive social cataclysm and it's a 739 00:34:38.659 --> 00:34:40.139 threat to whatever structures are in place? 740 00:34:40.219 --> 00:34:42.800 I think violence is often, whether it's to a democracy 741 00:34:42.860 --> 00:34:44.860 or to other, I mean, the French Revolution, right, where 742 00:34:44.880 --> 00:34:47.420 you have poor people, right, who are rising up and 743 00:34:47.460 --> 00:34:50.260 a middle class who's. whatever we can get into the 744 00:34:50.300 --> 00:34:52.730 history of the french revolution another time but. 745 00:34:52.650 --> 00:34:56.269 That's probably not on today's docket this fell asleep. 746 00:34:56.210 --> 00:34:58.329 But i mean um but no i mean like i 747 00:34:58.369 --> 00:35:00.710 think of historically i think too of like the 19th 748 00:35:00.730 --> 00:35:03.469 century the industrial revolution in england and there was like 749 00:35:03.639 --> 00:35:05.739 getting close to a french revolution type thing and they 750 00:35:05.780 --> 00:35:08.039 made some massive changes and things like that to try 751 00:35:08.059 --> 00:35:10.940 to head that off successfully so um you know i 752 00:35:11.119 --> 00:35:13.599 there i think like i think it's for me it's 753 00:35:13.639 --> 00:35:15.139 something i think about a lot i think about democracy 754 00:35:15.159 --> 00:35:18.949 a lot in general and and particularly like Yeah. So 755 00:35:19.010 --> 00:35:20.570 go ahead. I won't say more. I have more to say, 756 00:35:20.590 --> 00:35:20.949 but I'll wait. 757 00:35:23.590 --> 00:35:26.429 I mean, I'll go ahead on this. I can't because 758 00:35:26.449 --> 00:35:29.320 I think this is one of the things that will 759 00:35:29.380 --> 00:35:33.739 sound like a joke. And I'm just speaking plainly in 760 00:35:33.780 --> 00:35:37.239 my adult life. most of which has been spent in 761 00:35:37.619 --> 00:35:43.139 public higher education, I have soured on democracy. I'm not 762 00:35:43.159 --> 00:35:45.679 saying I had a better option. I'm pretty sympathetic with 763 00:35:45.699 --> 00:35:48.000 the old idea that democracy is the absolute worst form 764 00:35:48.019 --> 00:35:49.610 of government, except that all the others are even worse. 765 00:35:51.150 --> 00:35:58.269 But teaching college and meeting so many young people who, frankly, 766 00:35:58.289 --> 00:35:59.989 and I've had some wonderful students, and if any of 767 00:36:00.010 --> 00:36:01.869 them are listening, I'm certainly not talking about anybody who's 768 00:36:01.909 --> 00:36:05.210 listening to this podcast, but Except you, Joe. Oh, Joe. 769 00:36:05.269 --> 00:36:08.429 Don't get me started on Joe. But, I mean, so 770 00:36:08.510 --> 00:36:12.619 many of these college students were such poor thinkers, so 771 00:36:12.920 --> 00:36:18.739 easily swayed by whatever was trendy or cool or whatever. 772 00:36:19.380 --> 00:36:22.110 And I'm like... like why do these people get to 773 00:36:22.210 --> 00:36:24.949 vote is something that I've wondered more than once in 774 00:36:24.989 --> 00:36:27.989 the past. And I think that to answer the question 775 00:36:28.010 --> 00:36:32.809 that's at hand, the same dynamics that, um, led me 776 00:36:32.849 --> 00:36:35.309 to start thinking that sort of thing, 10, 15, 20 777 00:36:35.309 --> 00:36:40.219 years ago, um, are accelerated and greatly enhanced by, by 778 00:36:40.280 --> 00:36:44.159 artificial intelligence. Um, that the more democracy doesn't work when 779 00:36:44.199 --> 00:36:48.320 you have a population that can't think. And, um, The 780 00:36:48.440 --> 00:36:51.469 great danger from my point of view of AI in 781 00:36:51.510 --> 00:36:56.070 our culture is that it's creating a populace that cannot think. 782 00:36:56.710 --> 00:36:58.250 So I would say, yeah, I think there's a threat 783 00:36:58.269 --> 00:36:58.949 to democracy there. 784 00:37:01.269 --> 00:37:06.300 I feel that is a case of us romanticizing how 785 00:37:07.079 --> 00:37:09.320 our ancestors must have all been brilliant when most of 786 00:37:09.360 --> 00:37:12.980 them were illiterate and very susceptible to the same propaganda 787 00:37:13.159 --> 00:37:15.599 we are today. I'm maybe saying this just to be 788 00:37:15.619 --> 00:37:19.570 a little bit argumentative Because I do think AI is 789 00:37:19.630 --> 00:37:25.809 like propaganda on steroids times 100 times 1,000. Or maybe 790 00:37:25.889 --> 00:37:29.409 like the algorithm that AI is fueling via social media. 791 00:37:29.570 --> 00:37:32.860 I think social media is worse than when we talk AI. 792 00:37:33.059 --> 00:37:36.820 I'm doing that in quotes. I think social media is 793 00:37:37.179 --> 00:37:41.320 way worse. But I think we're deluding ourselves to think 794 00:37:41.360 --> 00:37:50.139 that our ancestors were all brilliant philosophers who thought... They 795 00:37:50.159 --> 00:37:55.650 were equally susceptible to dumb stuff. Our ancestors were all brilliant. 796 00:37:55.750 --> 00:37:58.210 As soon as you start making universal claims, of course 797 00:37:58.230 --> 00:37:59.190 that's going to be easy to refute. 798 00:37:59.210 --> 00:38:01.940 But you're making universal claims that everybody young is easily 799 00:38:01.980 --> 00:38:02.480 susceptible to this. 800 00:38:02.500 --> 00:38:06.500 I didn't say everybody. I'm talking about general trends, and 801 00:38:06.539 --> 00:38:09.420 I think we can say things confidently about... a populace 802 00:38:09.800 --> 00:38:12.219 at a general level, because you can look, we can 803 00:38:12.280 --> 00:38:14.309 actually go back and look at like the kinds of 804 00:38:14.429 --> 00:38:19.550 books that sold well in 1950 or 1900 or 1800 805 00:38:19.550 --> 00:38:21.889 versus the kinds of books that sell well today. And that, 806 00:38:22.030 --> 00:38:24.250 that does say something about where we are, like where 807

00:38:24.269 --> 00:38:26.800 the median American is in terms of an ability to 808 00:38:26.940 --> 00:38:31.619 process complex ideas, speak in, And process complex sentences and 809 00:38:31.659 --> 00:38:35.679 so on. That's a pretty good test for where we 810 00:38:35.760 --> 00:38:37.639 are as a people. Though there's going to be exceptions 811 00:38:37.679 --> 00:38:41.139 in both directions. Blah, blah, blah, blah, blah. 812 00:38:41.300 --> 00:38:49.889 My thought is that I feel there is a... backlash 813 00:38:49.969 --> 00:38:54.449 to technology that is developing among Gen Alpha, right? 814 00:38:54.469 --> 00:38:57.000 They're so young. They're like children. So Gen Z is 815 00:38:57.019 --> 00:38:58.619 the most anti-AI generation. 816 00:38:58.679 --> 00:38:59.119 Okay. 817 00:38:59.280 --> 00:39:03.099 So maybe the optimist to me, because I'm an eternal 818 00:39:03.179 --> 00:39:07.300 optimist about most things, is that there is a bit 819 00:39:07.320 --> 00:39:12.550 of a skepticism that develops among the, say, average person, 820 00:39:12.610 --> 00:39:16.920 average American that goes, Is what I'm reading bullshit? How 821 00:39:16.980 --> 00:39:18.800 can I verify? Because you're going to have to think 822 00:39:18.880 --> 00:39:20.610 that's going to have to be the default is that 823 00:39:20.679 --> 00:39:23.409 is this a deep fake? Is this video real? How 824 00:39:23.429 --> 00:39:26.710 do I know it's real? And I think that's a 825 00:39:26.769 --> 00:39:29.130 good thing. A little bit of a skepticism to say, 826 00:39:29.150 --> 00:39:31.139 I don't know about that. this. I got to verify 827 00:39:31.179 --> 00:39:33.599 my source here before I judge. 828 00:39:33.619 --> 00:39:35.579 Yeah. It's weird that we've gotten to the point so 829 00:39:35.619 --> 00:39:39.199 quickly where like, even as an, as an old person, uh, I'm, 830 00:39:39.460 --> 00:39:41.550 I'm seeing stuff pop up online and I'm like, I 831 00:39:41.590 --> 00:39:43.190 have no idea if this is real or not. Like, 832 00:39:43.230 --> 00:39:46.670 I can't believe that this, this political figure did or 833 00:39:46.730 --> 00:39:48.150 did not say this. Cause for all I know, this 834 00:39:48.170 --> 00:39:50.809 was created by their enemies, uh, And it's totally fake. 835 00:39:51.909 --> 00:39:54.650 Yeah. But, you know, I guess I sit there and go, man, 836 00:39:55.550 --> 00:40:00.179 I the average American is really vulnerable, really naive. 837 00:40:00.380 --> 00:40:00.579 Right. 838 00:40:00.800 --> 00:40:04.940 And like very persuadable. I mean, I would. 839 00:40:06.039 --> 00:40:08.030 Yeah, for sure. And I mean, I think that's we've 840 00:40:08.050 --> 00:40:10.510 talked in the past about I've said that I think 841 00:40:10.789 --> 00:40:14.570 like media is inseparable. The question of like our media 842 00:40:14.610 --> 00:40:17.480 and I mean that broadly, including social media is. is 843 00:40:17.539 --> 00:40:20.869 inseparable from questions of democracy, unfortunately. That's where, to me, 844 00:40:20.929 --> 00:40:24.710 the problem lies. And I agree with Matt that I 845 00:40:24.769 --> 00:40:25.210 think that. 846 00:40:26.750 --> 00:40:28.110 That young people should not have the vote. 847 00:40:28.309 --> 00:40:35.849 No, absolutely not. I don't agree. I am a pro-democracy. Democracy, 848 00:40:35.889 --> 00:40:40.880 to me, is a first principle. And I do think AI, 849 00:40:42.030 --> 00:40:44.400 insofar as it does sort of encourage people, and I'm 850 00:40:44.420 --> 00:40:46.739 not saying you have to interact with it this way, 851 00:40:46.760 --> 00:40:49.099 but encourage people not to think critically. And it does, 852 00:40:49.199 --> 00:40:53.559 I think fundamentally, um, that's a problem for democracy. I 853 00:40:53.579 --> 00:40:55.139 will say, I don't think like. 854 00:40:55.099 --> 00:40:57.800 You're saying that, sorry, you're saying AI teaches you not 855 00:40:57.840 --> 00:41:00.449 to think critically this. Okay. I wasn't sure if I 856 00:41:00.489 --> 00:41:00.820 heard you right. 857 00:41:00.909 --> 00:41:03.349 Or at least the, the way that people use AI 858 00:41:03.449 --> 00:41:06.269 is that, yeah. Um, and I, so, and then, and 859 00:41:06.289 --> 00:41:07.989 then I just want to say an aside though, that 860 00:41:08.050 --> 00:41:10.849 I don't think a democracy has, like, I don't think 861 00:41:10.889 --> 00:41:14.090 a person needs to be a complex, sophisticated, critical thinker 862 00:41:14.150 --> 00:41:15.880 to deserve the vote. So if we go back in 863 00:41:15.889 --> 00:41:18.019 our history a little bit, they had literacy tests to 864 00:41:18.039 --> 00:41:20.079 keep black people from voting, et cetera. And like, that's 865 00:41:20.659 --> 00:41:24.559 bullshit because people should have representation in a democracy. 866 00:41:24.679 --> 00:41:27.059 And there, I want to be clear there. I mean, there, 867 00:41:27.340 --> 00:41:29.420 though I was not being like trying to be funny, 868 00:41:29.480 --> 00:41:32.219 I also do want to acknowledge a lot more nuance 869 00:41:32.260 --> 00:41:33.019 to this stuff than. 870 00:41:32.960 --> 00:41:37.110 I, but here's, here's where I see another problem, which 871 00:41:37.190 --> 00:41:42.659 is like, is, is, um, the control of information. AI is, 872 00:41:43.139 --> 00:41:45.510 is own, I mean, AI, we, we interact with AI 873 00:41:45.969 --> 00:41:47.789 in terms of like most of us, right? Not, I 874 00:41:47.829 --> 00:41:49.949 wouldn't say like specialized programs, but most of us interact 875 00:41:49.969 --> 00:41:54.030 with AI through just a couple or few corporations, right? 876 00:41:54.050 --> 00:41:56.550 And those corporations are run by specific people and those 877 00:41:56.570 --> 00:42:01.699 specific people have particular intentions. It may be, um, money, 878 00:42:01.840 --> 00:42:03.139 keep you on there as long as you can, so 879 00:42:03.159

--> 00:42:05.820 they make the chatbots engaging, et cetera. It may be political. 880 00:42:06.079 --> 00:42:08.699 We know some of the more famous people have particular 881 00:42:08.739 --> 00:42:14.280 political views, and they work the answers to AI to meet, 882 00:42:14.840 --> 00:42:18.679 in some cases, to match their worldview. And I just 883 00:42:18.739 --> 00:42:21.539 worry that the more reliant we are on that because 884 00:42:21.599 --> 00:42:23.920 people see it as like an objective source of information 885 00:42:23.940 --> 00:42:26.199 or they see it as all knowing, they're not going 886 00:42:26.219 --> 00:42:29.230 into it skeptically in the same way. And like that 887 00:42:29.269 --> 00:42:32.889 you would, if you like, I don't know, I, it 888 00:42:32.929 --> 00:42:34.889 just seems to me, people are sort of the control 889 00:42:34.929 --> 00:42:37.650 there is, is, is a little bit terrifying in terms 890 00:42:37.710 --> 00:42:38.730 of the number of people. 891 00:42:38.750 --> 00:42:40.610 I think before you go further, Chris, I just, I 892 00:42:40.650 --> 00:42:43.070 think it's interesting that I want to agree with both 893 00:42:43.090 --> 00:42:45.880 of you. Like I agree with Chris's sense that people 894 00:42:45.920 --> 00:42:51.059 are increasingly skeptical, but I agree with your statement, but 895 00:42:51.139 --> 00:42:54.880 not of AI. If it comes to us through ChatGPT, 896 00:42:55.059 --> 00:42:56.360 it seems like people are very credulous. 897 00:42:56.989 --> 00:43:00.829 Not at all. The last poll I saw was 17% 898 00:43:00.829 --> 00:43:05.789 of Americans are optimistic about our AI future. The vast 899 00:43:05.829 --> 00:43:07.980 majority of people are like, this is scary as hell. 900 00:43:08.019 --> 00:43:10.340 Yeah, that has to do with labor and data centers 901 00:43:10.360 --> 00:43:10.900 and things like that. 902 00:43:11.139 --> 00:43:14.130 I don't know what the answer to this is, but 903 00:43:14.150 --> 00:43:16.489 it would be interesting to get the same kind of 904 00:43:16.530 --> 00:43:20.760 survey to say, if you put a question into chat GPT, 905 00:43:20.889 --> 00:43:22.789 how confident are you that the answer it gives you 906 00:43:22.889 --> 00:43:23.869 is objectively true? 907 00:43:24.150 --> 00:43:26.449 I'd say the vast majority would say, absolutely not. Most 908 00:43:26.489 --> 00:43:28.809 people are aware that chat GPT hallucinates. 909 00:43:28.849 --> 00:43:31.130 Then why do people ask them to write their best 910 00:43:31.170 --> 00:43:33.349 man speeches? Do you know what I mean? I'm just 911 00:43:33.369 --> 00:43:36.530 saying like that, that is a credulous, like you're saying, 912 00:43:36.710 --> 00:43:37.329 I trust this. 913 00:43:37.610 --> 00:43:39.260 I think if you went out on the street and 914 00:43:39.300 --> 00:43:41.440 you talked to somebody and said, do you like, who's 915 00:43:41.460 --> 00:43:43.960 like a passing user of AI in any way? I 916

00:43:44.000 --> 00:43:47.900 think it's, pretty well known nowadays that AI hallucinates and is, 917 00:43:48.239 --> 00:43:52.010 it's gotten better, but it's still not objectively, you cannot 918 00:43:52.429 --> 00:43:53.429 take it for a fact. 919 00:43:53.670 --> 00:43:56.170 Well, my problem is less, I agree. Like there's that 920 00:43:56.210 --> 00:43:58.989 whole side of like it getting facts wrong and that's important. 921 00:43:59.590 --> 00:44:02.309 I think that will probably continue to improve. My problem 922 00:44:02.369 --> 00:44:05.110 is not questions of fact. It's questions of judgment and 923 00:44:05.150 --> 00:44:07.429 value and things like that, where people trust it to 924 00:44:07.469 --> 00:44:10.239 sort of... Like, where they're asking, what should I... Should 925 00:44:10.280 --> 00:44:11.179 I divorce my wife? 926 00:44:11.699 --> 00:44:12.179 Should I. 927 00:44:13.639 --> 00:44:16.340 How do I, you know, deal with this employee? 928 00:44:16.719 --> 00:44:17.179 How do I. 929 00:44:17.420 --> 00:44:21.610 Like, these questions, those sorts of things, where, like... Who's 930 00:44:21.670 --> 00:44:23.730 behind that? And the problem is it's sort of hidden. 931 00:44:23.789 --> 00:44:26.329 Whereas if you buy a newspaper, I'm going way back here. 932 00:44:27.489 --> 00:44:29.030 And this is true also of the internet, what I'm saying, 933 00:44:29.050 --> 00:44:30.980 by the way, about AI. But like, because you can 934 00:44:31.000 --> 00:44:33.719 hide behind anonymity. But if you buy a newspaper or 935 00:44:33.760 --> 00:44:38.380 you watch NBC News in 1980, there's a corporation, journalism, 936 00:44:38.400 --> 00:44:41.880 a sort of standard of credibility there. First of all, 937 00:44:41.920 --> 00:44:44.800 in terms of not making up facts, but also you 938 00:44:46.139 --> 00:44:48.719 can assign a source. And the thing about AI is 939 00:44:48.739 --> 00:44:50.769 you can't assign a source in the way, and I 940 00:44:50.829 --> 00:44:54.230 think that is really dangerous. And again, it's just sort 941 00:44:54.269 --> 00:44:58.210 of hidden, and people don't think about it. So imagine, 942 00:44:58.369 --> 00:45:03.369 I'm just saying, it is an authoritarian, totalitarian dictator's dream AI. 943 00:45:04.389 --> 00:45:07.780 I think so is the media and anybody who controls media, 944 00:45:07.840 --> 00:45:10.699 which is like several, it's like a small handful of 945 00:45:10.719 --> 00:45:13.900 corporations that are owned by several people who are very powerful. 946 00:45:14.280 --> 00:45:16.840 And you can persuade the masses through the media. You 947 00:45:16.860 --> 00:45:20.480 could do it through social media. And I think AI 948 00:45:20.699 --> 00:45:23.980 is another example, but this is what's been happening from 949 00:45:24.019 --> 00:45:24.699 time immorium. 950 00:45:25.760 --> 00:45:28.500 No, not from time immorium, but I would say in 951 00:45:28.699 --> 00:45:31.719 our lives. But I

would also just say that the 952 00:45:31.840 --> 00:45:35.969 more we move our lives into that digital kind of... 953 00:45:36.590 --> 00:45:39.909 uh virtual AI space the more we say like I 954 00:45:39.949 --> 00:45:42.989 don't I mean like literally there are some people who 955 00:45:43.190 --> 00:45:45.530 can't lead their you know they can't like live their 956 00:45:45.570 --> 00:45:48.079 lives like don't have that are like what do I 957 00:45:48.119 --> 00:45:49.960 say to my date she's in the bathroom right like 958 00:45:49.980 --> 00:45:52.079 how do I talk to this person it is sad 959 00:45:52.119 --> 00:45:53.840 but I mean the more and that's going that goes 960 00:45:53.900 --> 00:45:55.840 on more and more and people are increasing that kind 961 00:45:55.860 --> 00:45:59.860 of use of AI and like that has political ramifications 962 00:45:59.880 --> 00:46:02.070 because the less you can look around your own life 963 00:46:02.110 --> 00:46:05.090 and be like, I'm hungry, I don't like this, I 964 00:46:05.110 --> 00:46:08.199 don't like this, whatever, and instead sort of get all your, like... 965 00:46:08.590 --> 00:46:10.989 believe what a chatbot is telling you about your own life. 966 00:46:11.570 --> 00:46:13.920 Like the more we move into those spaces, that to 967 00:46:13.960 --> 00:46:16.880 me is dangerous. And I'm feeling more and more like 968 00:46:16.900 --> 00:46:17.460 Ron Swanson. 969 00:46:17.480 --> 00:46:21.059 There's another layer of problem here that like, I mean, Amanda, 970 00:46:21.079 --> 00:46:23.019 what you keep circling back to is not being able 971 00:46:23.059 --> 00:46:26.280 to trust that you're getting information from good sources when 972 00:46:26.320 --> 00:46:28.090 it's filtered by chat GPT or whatever. 973 00:46:28.110 --> 00:46:29.929 Not just information judgment. 974 00:46:30.150 --> 00:46:33.510 Sure. Okay. Yeah. But then there's this broader issue that 975 00:46:33.530 --> 00:46:36.610 we'd be facing, whether we had AI or not. with like, 976 00:46:37.269 --> 00:46:41.090 like what's gone on with credentials in America in our lifetime. 977 00:46:41.130 --> 00:46:44.539 Like it's so many people have initials after their names. 978 00:46:44.659 --> 00:46:47.739 So like there are people who are incompetent. 979 00:46:47.800 --> 00:46:49.079 Like CHD. 980 00:46:49.239 --> 00:46:51.840 It's true. I mean, Amanda and I both have PhDs 981 00:46:51.880 --> 00:46:53.960 for those who don't know. And I know Chris is joking, 982 00:46:54.000 --> 00:46:56.420 but there's a lot of idiots out there who have 983 00:46:56.440 --> 00:46:57.949 a PhD in something or other. 984 00:46:58.010 --> 00:47:00.650 Well, hold on. In traditional educational systems, you go through 985 00:47:00.690 --> 00:47:02.610 a whole lot to get an PhD. 986 00:47:02.630 --> 00:47:05.289 The point is, there's a lot of educated morons out there. 987 00:47:05.389 --> 00:47:08.369 Education is one thing. It's not necessarily. 988 00:47:08.389 --> 00:47:10.199 I don't want to get into the whole thing about 989 00:47:10.360 --> 00:47:14.139 higher ed and whatnot. I just think it's undeniable that there... 990 00:47:15.019 --> 00:47:19.110 There are whole academic disciplines where under the sway of 991 00:47:19.170 --> 00:47:23.110 particular political views or moral views or whatever, like there's 992 00:47:23.130 --> 00:47:27.409 certain things that that whole field can't... Sorry, I said 993 00:47:27.449 --> 00:47:30.030 I didn't want to get to hire it, and we shouldn't. 994 00:47:30.530 --> 00:47:34.030 Suffice it to say, there are questions about whom you 995 00:47:34.070 --> 00:47:37.059 can trust that have nothing to do with AI and 996 00:47:37.079 --> 00:47:40.440 that are ubiquitous in our culture. I don't know what 997 00:47:40.460 --> 00:47:42.980 to say about that, but I guess maybe this is 998 00:47:43.000 --> 00:47:44.480 the only place where I'm pushing back, is it's not 999 00:47:44.539 --> 00:47:46.480 clear to me that if we could wave a magic 1000 00:47:46.519 --> 00:47:49.019 wand and all the AI disappeared tomorrow, I don't know 1001 00:47:49.039 --> 00:47:51.159 if the problem that you're pointing to is one that 1002 00:47:51.199 --> 00:47:52.719 would be in any way improved. 1003 00:47:52.860 --> 00:47:54.789 My sense is there's a lot of people who'd be 1004 00:47:54.849 --> 00:47:59.530 fine with AI if their particular political party was in charge. Right. 1005 00:47:59.590 --> 00:48:00.309 And that's a problem. 1006 00:48:00.489 --> 00:48:00.670 Yeah. 1007 00:48:00.690 --> 00:48:05.219 That's called authoritarianism. I mean, I just read 1984 and like, 1008 00:48:05.260 --> 00:48:09.880 that's what's happening with technology there, but like the, no, I, I, I, okay. 1009 00:48:11.269 --> 00:48:14.269 Sam Altman has proposed a sort of personal agent, right? 1010 00:48:14.369 --> 00:48:17.989 That your AI agent that is tapped into your email, 1011 00:48:18.090 --> 00:48:20.079 and maybe some of you, maybe you do this already 1012 00:48:20.099 --> 00:48:22.880 with some things, but your email, your internet search history, 1013 00:48:23.280 --> 00:48:27.679 you're listening to your conversations with your family so that 1014 00:48:27.699 --> 00:48:33.050 it can always be helping you, right? I find that terrifying. 1015 00:48:33.090 --> 00:48:34.429 That's really creepy. 1016 00:48:34.449 --> 00:48:37.969 That a corporation, let's be clear, it is a corporation, 1017 00:48:38.289 --> 00:48:41.320 would have access to your entire life like that and 1018 00:48:41.380 --> 00:48:44.900 could mess it. Well, right, but that's not the case. 1019 00:48:45.219 --> 00:48:47.659 I guess that's what I'm saying here is that what's 1020 00:48:47.719 --> 00:48:51.610 happening here is not government. It's private corporations. Sure, but 1021 00:48:51.619 --> 00:48:54.260 that's not here, I'm saying. But at least if it 1022 00:48:54.300 --> 00:48:56.440 were here, then

we'd vote on it. You know what 1023 00:48:56.480 --> 00:48:59.780 I mean? That's the difference. Here, we're sort of having 1024 00:48:59.800 --> 00:49:02.460 these things washed over us, and we don't really have 1025 00:49:02.500 --> 00:49:04.739 choices with private corporations. 1026 00:49:05.019 --> 00:49:06.550 I thought we were going to solve this issue in 1027 00:49:06.550 --> 00:49:07.070 15 minutes. 1028 00:49:08.139 --> 00:49:10.070 Clearly, we haven't. It does feel like we didn't completely 1029 00:49:10.070 --> 00:49:12.460 solve it, but also that we should probably consider moving 1030 00:49:12.460 --> 00:49:14.090 on to the next question on the list. Okay. I 1031 00:49:14.130 --> 00:49:16.159 have a good question. You know, it occurs to me, 1032 00:49:16.389 --> 00:49:18.920 before we go to breaking your question, you know, I 1033 00:49:18.940 --> 00:49:24.320 don't think we've ever done a... forms of government 3Q3D. 1034 00:49:24.489 --> 00:49:26.139 Let's do it. Let's talk about it. 1035 00:49:26.139 --> 00:49:27.380 What 3Q3D were boring to me? 1036 00:49:29.559 --> 00:49:30.860 We'll tell far jokes. 1037 00:49:30.880 --> 00:49:33.320 I don't know any. I'm not smart enough to do that, 1038 00:49:33.400 --> 00:49:33.739 for real. 1039 00:49:35.239 --> 00:49:38.239 I'm not either, but Amanda is. Amanda is, but I'm not. 1040 00:49:38.260 --> 00:49:39.139 She'll help us. 1041 00:49:39.219 --> 00:49:40.969 Not a question of smarts, I just think about it 1042 00:49:40.989 --> 00:49:41.329 a lot. 1043 00:49:42.650 --> 00:49:45.610 Well, let's hear what our corporate overlords want us to 1044 00:49:45.650 --> 00:49:48.250 purchase next. And we'll be back in a few minutes 1045 00:49:48.570 --> 00:49:59.940 with another drink, another question, and more life-changing discussions. Welcome back, everybody. 1046 00:50:00.300 --> 00:50:03.630 It's three questions, three drinks. And next up is Mr. 1047 00:50:03.670 --> 00:50:04.510 Christopher Michael. 1048 00:50:04.929 --> 00:50:08.190 Oh, let's go. I have a cocktail here, folks. 1049 00:50:09.659 --> 00:50:10.000 Okay. 1050 00:50:10.099 --> 00:50:13.940 This one's interesting. I, Amanda, don't be mad, but I 1051 00:50:14.239 --> 00:50:16.840 went to Google. I didn't go to AI. I said, 1052 00:50:16.860 --> 00:50:19.199 I need a drink that is like got some sort 1053 00:50:19.239 --> 00:50:22.739 of AI theme to it or whatever. And it came 1054 00:50:22.780 --> 00:50:25.440 back and said, there is a, I don't know if 1055 00:50:25.460 --> 00:50:28.699 this is a man or woman, but somebody who is 1056 00:50:28.739 --> 00:50:33.110 a professor at MIT teaches math and moonlights at the 1057 00:50:33.150 --> 00:50:37.070 four seasons bar in Boston and created a drink called 1058 00:50:37.090 --> 00:50:42.369 the algorithm. Um, This person, Francois Magritte, whatever. 1059 00:50:42.570 --> 00:50:42.869 I don't know. 1060 00:50:43.670 --> 00:50:47.590 So created this drink and it is, and it, I 1061 00:50:47.630 --> 00:50:49.510 liked the name out of the algorithm. 1062 00:50:49.829 --> 00:50:50.329 Yeah. Okay. 1063 00:50:50.969 --> 00:50:54.340 And I thought, well, they're a math professor, so it 1064 00:50:54.380 --> 00:50:58.599 seems AI like, right? And the drink is, it's one 1065 00:50:58.619 --> 00:51:05.139 and a half ounces gin, part Amaro, part Curacao, orange Curacao, 1066 00:51:05.699 --> 00:51:09.159 a little Galliano, some fresh lemon juice. Some dashes of 1067 00:51:09.179 --> 00:51:11.699 orange bitters. Oh, I was supposed to put tonic on it. 1068 00:51:13.280 --> 00:51:15.260 Oh, I don't think I have tonic. 1069 00:51:16.039 --> 00:51:16.719 Go get it. 1070 00:51:16.820 --> 00:51:19.559 Yeah, hang on one second. I'm not drinking this without 1071 00:51:19.599 --> 00:51:20.000 the tonic. 1072 00:51:20.059 --> 00:51:25.099 Yeah, me neither. So, of course, this is inspired by 1073 00:51:25.219 --> 00:51:27.340 Al Gore. The Al Gore rhythm. 1074 00:51:27.360 --> 00:51:29.219 You made that up. 1075 00:51:29.380 --> 00:51:33.619 That's not my joke. I was just filling the time 1076 00:51:33.659 --> 00:51:34.570 while you got your tonic. 1077 00:51:35.750 --> 00:51:39.219 Okay, so there we go. The algorithm. Try it. 1078 00:51:39.230 --> 00:51:42.389 All right. I'm splashing it a little. 1079 00:51:42.489 --> 00:51:44.329 It's an interesting concoction. 1080 00:51:44.650 --> 00:51:45.269 Yeah, let's see. 1081 00:51:45.329 --> 00:51:46.219 Let's try it. 1082 00:51:51.550 --> 00:51:54.389 It's good. It tastes kind of like Matt's. 1083 00:51:57.250 --> 00:52:02.030 No? Well, it has a similar profile. What? I mean. 1084 00:52:02.050 --> 00:52:03.969 I love this so much, what's happening right now. 1085 00:52:03.989 --> 00:52:04.239 See if. 1086 00:52:05.280 --> 00:52:06.139 Are you mad that I said that? 1087 00:52:06.960 --> 00:52:11.349 I've never been so angry. No, I think it's different. 1088 00:52:11.369 --> 00:52:14.800 I have to tell you something. I just lied completely. 1089 00:52:15.400 --> 00:52:19.480 There is no person... named Francois, whatever they make up. 1090 00:52:19.519 --> 00:52:20.179 Magritte? 1091 00:52:20.320 --> 00:52:21.179 I was like, the artist? 1092 00:52:21.219 --> 00:52:23.059 Did you do the same thing I did to you? 1093 00:52:23.139 --> 00:52:26.780 I did the exact same thing that you did. I 1094 00:52:26.840 --> 00:52:29.500 took a picture of my bar and I said, these 1095 00:52:29.539 --> 00:52:32.889 are the ingredients. Create a drink that would be like AI-like. 1096 00:52:33.690 --> 00:52:36.670 And I said, it tastes like that. 1097 00:52:36.750 --> 00:52:39.750 Exactly. The homogenization. 1098 00:52:39.889 --> 00:52:42.969 You're right. But here's the funny thing is, it came back, 1099 00:52:43.030 --> 00:52:47.139 the first drink it came back with It said, you 1100 00:52:47.179 --> 00:52:50.500 should create this drink called the neural network, blah, blah, blah. 1101 00:52:50.599 --> 00:52:54.159

And it said, use the blue curacao you have. And 1102 00:52:54.199 --> 00:52:57.420 I wrote, where is the blue curacao in the picture? 1103 00:52:57.460 --> 00:52:58.119 There is none. 1104 00:52:58.219 --> 00:52:58.500 I don't. 1105 00:52:58.579 --> 00:53:00.239 And it goes, oh, you're right. My mistake. 1106 00:53:00.519 --> 00:53:01.900 I'm like, you idiot. 1107 00:53:03.639 --> 00:53:08.690 Amanda's face right now is so disgusted. So I was like, 1108 00:53:08.909 --> 00:53:10.590 come up with a different drink, you idiot. 1109 00:53:10.610 --> 00:53:12.110 Did you say you idiot? 1110 00:53:12.150 --> 00:53:12.369 Yeah. 1111 00:53:12.429 --> 00:53:12.909 No, I didn't. 1112 00:53:13.190 --> 00:53:14.449 I said, try harder. 1113 00:53:14.469 --> 00:53:15.090 You're doing fine. 1114 00:53:15.849 --> 00:53:17.780 And, uh, you don't need to be nice to it. 1115 00:53:17.800 --> 00:53:20.619 You guys, it's a computer program. It's not conscious. 1116 00:53:20.639 --> 00:53:22.719 And it was like, okay, right, right. And then it 1117 00:53:22.739 --> 00:53:24.760 came back and it said, you should try this. Uh, 1118 00:53:24.820 --> 00:53:27.949 we'll call it the algorithm. And then I was concerned 1119 00:53:27.969 --> 00:53:30.650 that if I admitted that I had used AI to 1120 00:53:30.690 --> 00:53:32.650 create this drink, that you would immediately hate it. And 1121 00:53:32.670 --> 00:53:34.489 that's why I lied and made up the story about 1122 00:53:34.530 --> 00:53:35.489 the MIT professor. 1123 00:53:35.570 --> 00:53:35.969 Okay. 1124 00:53:36.230 --> 00:53:37.090 So yeah. 1125 00:53:37.110 --> 00:53:37.369 Yeah. 1126 00:53:38.300 --> 00:53:41.239 AI leads to moral choices. 1127 00:53:41.860 --> 00:53:42.059 No. 1128 00:53:42.480 --> 00:53:42.840 Anyway. 1129 00:53:42.920 --> 00:53:43.340 I don't know. 1130 00:53:43.360 --> 00:53:43.739 Hey. Hey. 1131 00:53:44.099 --> 00:53:45.539 I'll try anything. 1132 00:53:45.659 --> 00:53:46.239 I'm open. 1133 00:53:46.260 --> 00:53:49.019 It does seem like it's a very close cousin to 1134 00:53:49.059 --> 00:53:49.440 Matt's drink. 1135 00:53:49.460 --> 00:53:49.639 I know. 1136 00:53:49.679 --> 00:53:51.840 It's yellow, everyone. And I think what it's doing is. 1137 00:53:51.719 --> 00:53:55.420 It's just taking all the ingredients and mixing them all together. 1138 00:53:55.820 --> 00:54:00.219 Yeah, so it's basically like what a college student who 1139 00:54:00.239 --> 00:54:02.219 was having a party and freaking out about what to 1140 00:54:02.260 --> 00:54:03.000 make would do. 1141 00:54:03.300 --> 00:54:04.820 Yeah, which is fine by me. 1142 00:54:04.960 --> 00:54:06.539 It's like a Long Island iced tea. 1143 00:54:07.179 --> 00:54:07.840 Which is great. 1144 00:54:07.980 --> 00:54:10.219 You guys find them way more similar than I do. Like, 1145 00:54:10.260 --> 00:54:12.760 mine was bourbon-based. This is gin-based. I don't know. 1146 00:54:12.780 --> 00:54:14.780 I don't know what to tell you. 1147 00:54:14.800 --> 00:54:15.889 This is a bright yellow. 1148 00:54:15.909 --> 00:54:16.929 I'm taking a picture. 1149 00:54:16.949 --> 00:54:17.909 We do have an Instagram. 1150 00:54:18.130 --> 00:54:20.289 I have an AIDAR. 1151 00:54:20.829 --> 00:54:21.269 I just know. 1152 00:54:21.289 --> 00:54:22.010 A-A-A-DAR. 1153 00:54:22.090 --> 00:54:26.039 A-A-A-DAR. AIDAR, sorry. 1154 00:54:27.079 --> 00:54:28.059 Okay, I have a question. 1155 00:54:28.260 --> 00:54:29.900 Okay, ask a question. All right. 1156 00:54:31.679 --> 00:54:34.329 I mentioned earlier that I'm an optimist about most things, 1157 00:54:34.369 --> 00:54:39.369 but I am conflicted on AI. So let's talk about 1158 00:54:39.530 --> 00:54:45.510 our future and project 10 or 20 years forward. What 1159 00:54:45.590 --> 00:54:50.300 is our future like? My opinion is the AI toothpaste 1160 00:54:50.340 --> 00:54:52.610 is out of the tube. Like it's no going back. 1161 00:54:52.650 --> 00:54:56.269 Like we're going to have more and more AI driving things, 1162 00:54:56.849 --> 00:54:59.289 literally driving like our cars. I mean, 20 years from now, 1163 00:54:59.429 --> 00:55:02.949 I predict very few people are going to actually drive cars. 1164 00:55:02.969 --> 00:55:06.210 They'll be driven by the car, which is going to 1165 00:55:06.230 --> 00:55:09.429 be way safer and lead to far fewer fatalities and injuries, 1166 00:55:09.769 --> 00:55:12.530 which is a very good thing for society. But the 1167 00:55:12.570 --> 00:55:16.369 point is, what is our AI future like? Are you 1168 00:55:16.429 --> 00:55:20.780 bullish or are you pessimistic on what is transpiring today? 1169 00:55:21.090 --> 00:55:24.159 in the decades to come. Amanda, you look like an 1170 00:55:24.179 --> 00:55:25.610 optimist about this, so go ahead. 1171 00:55:26.659 --> 00:55:28.059 I reject the premise of the question. 1172 00:55:28.199 --> 00:55:28.400 Why? 1173 00:55:28.420 --> 00:55:31.179 Are you serious? I don't like the question about anything, 1174 00:55:31.260 --> 00:55:34.420 are you optimistic or pessimistic, because they both are passive positions. 1175 00:55:35.440 --> 00:55:37.460 What I mean is, if you're an optimist, you're like, well, 1176 00:55:37.619 --> 00:55:38.840 I think it's going to go great. And if you're 1177 00:55:38.860 --> 00:55:41.440 a pessimist, you're like, too bad, we're screwed, right? 1178 00:55:41.460 --> 00:55:44.039 It's a confirmation bias. You see what you want to see, right? 1179 00:55:44.539 --> 00:55:45.260 No, no. 1180 00:55:45.340 --> 00:55:48.570 What I mean is that that assumes we have no 1181 00:55:48.650 --> 00:55:50.929 say in what happens in the future. And I am 1182 00:55:50.989 --> 00:55:53.820 bullish about democracy. I'm bullish about our ability to. 1183 00:55:53.829 --> 00:55:56.760 It didn't sound like that in the last question. Well, no. 1184 00:55:58.139 --> 00:55:58.320 Right. 1185 00:55:58.400 --> 00:56:00.079 Like, there is a threat to that. But

I'm bullish 1186 00:56:00.139 --> 00:56:01.960 about our ability to control things. One of the things 1187 00:56:01.980 --> 00:56:04.780 that really bothers me about these, like, tech oligarchs that, like, 1188 00:56:04.820 --> 00:56:07.400 control everything is that they talk about things as if 1189 00:56:07.440 --> 00:56:09.699 democracy doesn't exist or the people have no say. So, like, 1190 00:56:09.880 --> 00:56:11.380 AI is going to do this. AI is going to 1191 00:56:11.420 --> 00:56:15.090 do this. We have control over what happens through if 1192 00:56:15.139 --> 00:56:18.530 we had a functioning democracy. But I guess so, like, 1193 00:56:18.630 --> 00:56:24.769 my thing is I don't– I want– my main– belief 1194 00:56:24.829 --> 00:56:27.409 and my main message to people is always like, don't 1195 00:56:27.550 --> 00:56:29.929 fall for the idea that quote, it's out of the toothpaste. 1196 00:56:29.989 --> 00:56:32.230 Not that AI is going back in the tooth, whatever. 1197 00:56:32.769 --> 00:56:36.039 Like I, what I mean is that we have a 1198 00:56:36.079 --> 00:56:40.139 real say over how that gets implemented, how the effects 1199 00:56:40.179 --> 00:56:42.679 it has on society, all these sorts of things. If 1200 00:56:42.760 --> 00:56:45.059 we want, if we take that power and control, I mean, 1201 00:56:45.360 --> 00:56:47.940 cause like it doesn't have to be something that just 1202 00:56:48.019 --> 00:56:50.530 happens to us is my point. Cause a lot of 1203 00:56:50.570 --> 00:56:52.309 this is stuff we never asked for and we can 1204 00:56:52.349 --> 00:56:52.579 say no. 1205 00:56:54.989 --> 00:56:57.630 Okay. I am holding my tongue. I'm going to bite 1206 00:56:57.650 --> 00:56:58.949 my tongue. Because, Matt, you go. 1207 00:56:59.010 --> 00:57:01.869 No, I think it's a really good point. And at 1208 00:57:01.889 --> 00:57:08.179 the same time, it doesn't seem fatalistic to look at 1209 00:57:08.460 --> 00:57:11.659 trends and how things are going and have a prediction. 1210 00:57:11.699 --> 00:57:12.059 And I think. 1211 00:57:14.010 --> 00:57:16.050 I think my attitude has changed in even just the 1212 00:57:16.090 --> 00:57:19.489 last month or two. I think I'm feeling more optimistic 1213 00:57:19.550 --> 00:57:21.889 than I was. And it's partly because I've picked up 1214 00:57:22.210 --> 00:57:25.829 on the anti-AI attitude of a lot of younger people. 1215 00:57:25.849 --> 00:57:28.610 We were talking about it earlier. Gen Z kids, sorry, 1216 00:57:28.929 --> 00:57:34.360 having like... They don't like it. And I think if 1217 00:57:34.400 --> 00:57:37.300 I had to put money on an outcome right now, 1218 00:57:37.820 --> 00:57:40.519 I think the most likely outcome is that we actually... 1219 00:57:41.250 --> 00:57:45.969 learn to use AI in a relatively human way, that 1220 00:57:46.110 --> 00:57:49.590 it gets integrated into our life the way most other 1221 00:57:49.639 --> 00:57:52.039 technologies have. Not that there's

no downside, not that there's 1222 00:57:52.079 --> 00:57:55.039 nothing bad that comes of it, but that on the whole, 1223 00:57:56.800 --> 00:58:00.869 we do an okay job of using this stuff in 1224 00:58:00.909 --> 00:58:04.530 a way that actually makes life better without costing us 1225 00:58:04.610 --> 00:58:08.309 too much. I'm not sure about that. I wouldn't put 1226 00:58:08.349 --> 00:58:09.929 a lot of money on it. But if I had 1227 00:58:09.949 --> 00:58:11.530 to put money on one outcome, that's the one that 1228 00:58:11.550 --> 00:58:12.389 I think is the most likely. 1229 00:58:13.969 --> 00:58:18.039 Yeah, for me, it boils down to, like, I think 1230 00:58:18.119 --> 00:58:21.260 about the human progress we have made with technology in 1231 00:58:21.280 --> 00:58:26.159 the last 200 years. In 1820, 94% of people lived 1232 00:58:26.239 --> 00:58:30.599 in grinding poverty in the world. 94%. It was awful. 1233 00:58:30.719 --> 00:58:35.070 Your life expectancy was crap. You worked until you die. Like, 1234 00:58:35.110 --> 00:58:39.159 there was no leisure. The average... That's not right. 1235 00:58:39.360 --> 00:58:40.239 That's not true. But go ahead. 1236 00:58:40.440 --> 00:58:43.579 Okay. I will give you the source on that. In 1820, 1237 00:58:43.579 --> 00:58:47.599 if you were born somewhere in the world, the vast 1238 00:58:47.699 --> 00:58:50.110 majority of people lived in grinding poverty. 1239 00:58:50.179 --> 00:58:52.679 Depends what you mean by poverty and also.... 1240 00:58:54.190 --> 00:58:56.110 Let me just keep going, okay? And I will say 1241 00:58:56.230 --> 00:58:59.750 that when you, again, life expectancy. 1242 00:59:00.070 --> 00:59:02.730 Like, We have less leisure today than we did in 1924. 1243 00:59:02.730 --> 00:59:06.590 100%, that's wrong. That is 100% wrong. The average person 1244 00:59:06.610 --> 00:59:09.090 in America works 37 hours a week. They used to 1245 00:59:09.110 --> 00:59:12.380 work 61 hours. a week in 1920. And I will 1246 00:59:12.420 --> 00:59:15.690 give you this fact. That's different than 1820. In 1820, 1247 00:59:15.690 --> 00:59:18.070 94% of people born in the world lived in grinding poverty. 1248 00:59:18.110 --> 00:59:19.289 By 1950, it was 50%. By 1990, it was 30%. 1249 00:59:19.289 --> 00:59:26.539 By 2025, it is under 30%. 10%, 9.4% of the 1250 00:59:26.599 --> 00:59:29.900 world lives in poverty. The vast majority of people do not. 1251 00:59:30.380 --> 00:59:33.869 And the truth is that yes, there are like rich 1252 00:59:33.949 --> 00:59:36.329 people out there that have more wealth than you can imagine. 1253 00:59:36.750 --> 00:59:40.289 But the average person in the world lives a life 1254 00:59:40.469 --> 00:59:44.190 that the kings of 200 years ago could not fathom, 1255 00:59:44.710 --> 00:59:48.119 could not fathom. in terms of what their, their ability 1256 00:59:48.219 --> 00:59:50.280 to do things is in terms. 1257 00:59:50.119 --> 00:59:52.980 Of material material. That's different. 1258 00:59:53.760 --> 00:59:53.960 Yeah. 1259 00:59:54.019 --> 00:59:56.179 I mean, we can talk about like what people actually 1260 00:59:56.320 --> 00:59:59.090 do with their time, but as far as like people 1261 00:59:59.150 --> 01:00:01.889 being able to live a life that they would want 1262 01:00:01.949 --> 01:00:05.170 to live and have abundant food to have leisure time 1263 01:00:05.250 --> 01:00:08.429 to do, to have a device in their pocket that 1264 01:00:08.469 --> 01:00:10.739 gives them every answer to every question at any point 1265 01:00:10.760 --> 01:00:15.679 in time to have The technological progress we have made 1266 01:00:15.719 --> 01:00:19.360 in 200 years is amazing for humanity. And I think 1267 01:00:19.860 --> 01:00:22.159 AI will continue to push that forward and that we 1268 01:00:22.199 --> 01:00:25.079 will look back and go, the majority of people are 1269 01:00:25.179 --> 01:00:28.659 still better off. The majority of people's quality of life. Now, again, 1270 01:00:28.699 --> 01:00:32.059 I'm not saying that people are going to be leading 1271 01:00:32.079 --> 01:00:34.639 more meaningful, happier lives, because I don't know that that's 1272 01:00:34.699 --> 01:00:37.780 any different than it was maybe where you're rolling your 1273 01:00:37.820 --> 01:00:40.039 eyes about is that maybe people had more meaning in 1274 01:00:40.059 --> 01:00:42.030 their lives 200 years ago even if it was short 1275 01:00:42.130 --> 01:00:45.969 and poverty stricken and nobody had indoor plumbing or any 1276 01:00:46.090 --> 01:00:50.000 anything like that like I lost my train of thought. 1277 01:00:50.800 --> 01:00:55.110 Well, no, but the question of whether technology is progress 1278 01:00:55.449 --> 01:00:58.030 is the question we're talking about today on this podcast. 1279 01:00:58.050 --> 01:01:01.250 And it is in question. It is not a given. 1280 01:01:01.530 --> 01:01:05.329 The progress we're making to lift people out of poverty 1281 01:01:05.369 --> 01:01:10.929 through technology is astonishingly good for humanity. Again, I am 1282 01:01:11.179 --> 01:01:13.300 not saying that people are finding more meaning in life. 1283 01:01:13.340 --> 01:01:15.119 And that's the bigger problem, I think, with AI is 1284 01:01:15.139 --> 01:01:19.030 that people are leading lonelier lives and, you know, if you're, 1285 01:01:19.170 --> 01:01:21.130 if you're interacting and your best friend is a chatbot, 1286 01:01:21.150 --> 01:01:24.090 that's a serious problem. But as far as like being 1287 01:01:24.369 --> 01:01:27.500 able to live a life where you couldn't do that 1288 01:01:27.500 --> 01:01:30.460 200 years ago, or even 50 years ago, the way 1289 01:01:30.480 --> 01:01:32.860 that you can now is, is astonishing. 1290 01:01:34.119 --> 01:01:39.039 Okay. I think I'm just saying like, okay, I have 1291 01:01:39.059 --> 01:01:40.400 lots of stuff to say about that. We can, we 1292 01:01:40.420 -->

01:01:41.469 can stick to AI. 1293 01:01:42.369 --> 01:01:45.880 Um, I think the march of progress will continue to 1294 01:01:45.940 --> 01:01:48.789 lift the vast majority of people out of grinding poverty. 1295 01:01:49.349 --> 01:01:52.130 The question is, do we become better as a species? 1296 01:01:52.250 --> 01:01:53.289 Do we become better people? 1297 01:01:53.889 --> 01:01:54.110 Right. 1298 01:01:54.190 --> 01:01:57.170 And I'm just saying, when you're talking about living in 1299 01:01:57.210 --> 01:02:00.760 grinding poverty, I think if you ask indigenous cultures in 1300 01:02:00.800 --> 01:02:04.829 the 17th century... I'm just saying like, are they counting 1301 01:02:04.869 --> 01:02:07.530 them as living in grinding property? Do you know what 1302 01:02:07.550 --> 01:02:09.869 I mean? Like that's, that's a very, like they, they 1303 01:02:09.889 --> 01:02:12.809 had outdoor plumbing, right. But like it was a intact 1304 01:02:12.829 --> 01:02:14.570 culture and like all these sorts of ways. So I don't, 1305 01:02:14.650 --> 01:02:15.590 I mean, they didn't have an iPhone. 1306 01:02:16.469 --> 01:02:17.119 I don't think they care. 1307 01:02:17.139 --> 01:02:19.980 You often hear of Americans who go do like service 1308 01:02:20.000 --> 01:02:23.039 projects and things in what we call third world countries 1309 01:02:23.079 --> 01:02:25.559 and so on, who, who lack so much of what 1310 01:02:25.579 --> 01:02:28.380 we have in terms of material goods. And I, by 1311 01:02:28.519 --> 01:02:30.880 no means do I want to minimize the suffering that 1312 01:02:30.900 --> 01:02:34.050 goes along with that. It is also the case that 1313 01:02:34.699 --> 01:02:38.019 you always hear people coming back from these experiences, Americans, saying, like, wow, 1314 01:02:38.039 --> 01:02:40.119 they have so much less than we do, but they're 1315 01:02:40.159 --> 01:02:42.179 so happy. There's no question. 1316 01:02:42.260 --> 01:02:45.519 What matters in life, no matter how prosperous we are, 1317 01:02:45.659 --> 01:02:47.949 is who you're sitting around the dinner table with or 1318 01:02:47.989 --> 01:02:50.329 who you're hanging out with, the connections that we make 1319 01:02:50.369 --> 01:02:53.630 with people. That's what matters the most in life. But 1320 01:02:54.590 --> 01:02:57.730 you don't want to have a life expectancy of 32 1321 01:02:57.730 --> 01:02:58.349 like it was in 1820. 1322 01:02:58.329 --> 01:02:58.469 Right. 1323 01:02:59.969 --> 01:03:03.030 To have like disease and famine and all these things 1324 01:03:03.449 --> 01:03:04.909 that we have largely eliminated. 1325 01:03:04.929 --> 01:03:07.550 But here's the thing. I don't like this idea that 1326 01:03:07.949 --> 01:03:10.150 it's sort of like all or nothing. Like we have 1327 01:03:10.190 --> 01:03:13.949 to accept all new technology or nothing. I think medical 1328 01:03:13.969 --> 01:03:17.510 breakthroughs are amazing. I love that part of science and technology. 1329 01:03:17.750 --> 01:03:19.179 And AI is going to cure cancer. 1330 01:03:19.219 --> 01:03:20.380 That doesn't mean, okay. 1331 01:03:21.039 --> 01:03:22.440 Which is not a bad thing. 1332 01:03:22.579 --> 01:03:24.719 Of course, of course. That doesn't mean it's a good 1333 01:03:24.739 --> 01:03:26.980 thing for people to carry around a phone. Like, I 1334 01:03:27.000 --> 01:03:28.179 don't think that's actually a good thing. 1335 01:03:28.280 --> 01:03:29.119 Yeah, I agree. 1336 01:03:29.199 --> 01:03:32.079 So what I mean is, like, we could do a 1337 01:03:32.099 --> 01:03:35.800 better job instead of letting, again, things happen to us, 1338 01:03:36.179 --> 01:03:39.099 make choices. And the thing that worries me, like, at 1339 01:03:39.139 --> 01:03:42.019 the end of the last segment, or two segments ago, 1340 01:03:42.280 --> 01:03:44.239 you talked about the idea of, and this is a 1341 01:03:44.579 --> 01:03:46.760 sort of broadly promised thing by a lot of leaders, 1342 01:03:46.820 --> 01:03:50.590 the CEOs of AI companies, that we'll all have more 1343 01:03:50.619 --> 01:03:52.710 leisure time because AI will do our work for us. 1344 01:03:53.610 --> 01:03:56.670 I want to see the policy plan on that because 1345 01:03:56.989 --> 01:03:59.090 when in our country have people been put out of 1346 01:03:59.130 --> 01:04:02.309 work and they've been just fine? Yeah. I want to 1347 01:04:02.369 --> 01:04:02.949 see how that's going to go. 1348 01:04:02.949 --> 01:04:03.550 No, that's a really good point. 1349 01:04:03.630 --> 01:04:07.050 And, like, that, to me, like, they're talking about it 1350 01:04:07.150 --> 01:04:10.590 as if, again, we don't have a, like, democracy doesn't 1351 01:04:10.610 --> 01:04:12.679 exist and we don't have policies. Like, who's going to 1352 01:04:12.699 --> 01:04:16.400 pay that universal basic income? Like, all those sorts of questions, right? 1353 01:04:16.619 --> 01:04:17.300 Like, those are. 1354 01:04:17.320 --> 01:04:18.480 Well, the robots will because they... No. 1355 01:04:18.579 --> 01:04:23.539 Yeah, they don't... It doesn't... so, so like that to me, 1356 01:04:23.780 --> 01:04:26.340 when people start talking like that, I'm like, show me 1357 01:04:26.619 --> 01:04:28.809 how we get from A to B because I think 1358 01:04:28.849 --> 01:04:30.130 a lot of people are going to be hurt in the, 1359 01:04:30.329 --> 01:04:32.110 in the, you know, I, I'm not, I'm not, I 1360 01:04:32.150 --> 01:04:35.070 don't think everybody should grind themselves in work all day. 1361 01:04:35.280 --> 01:04:37.079 I don't believe that at all. I just saying like, 1362 01:04:39.280 --> 01:04:43.199 No new technology has been widely adopted with a view 1363 01:04:43.239 --> 01:04:45.260 to how you help most people. It has been in 1364 01:04:45.300 --> 01:04:47.679 this country, a history, and in the West, a history 1365 01:04:47.820 --> 01:04:49.840 of how do a few people get a lot of 1366

01:04:49.880 --> 01:04:50.389 money out of it. 1367 01:04:50.400 --> 01:04:53.150 That's not true at all. That's exactly the opposite of 1368 01:04:53.170 --> 01:04:56.610 what I'm saying. The vast majority of people have prospered 1369 01:04:56.650 --> 01:04:59.550 because of technology, because of what we've been able to 1370 01:04:59.610 --> 01:05:04.530 do with medical breakthroughs and with like... Eventually. Every single 1371 01:05:04.550 --> 01:05:07.719 thing has gotten so much, like enabled us to have 1372 01:05:07.739 --> 01:05:12.260 a like an easier life, a much easier, and I'm 1373 01:05:12.320 --> 01:05:15.880 not saying it's more fulfilling. It's an easier life for 1374 01:05:15.920 --> 01:05:21.780 the vast majority of people. Like the 25 percentile income 1375 01:05:21.920 --> 01:05:24.860 in America live a life that is so much better 1376 01:05:24.909 --> 01:05:28.489 than the richest king of England 150 years ago. That's true. 1377 01:05:28.530 --> 01:05:29.610 It's unequivocally true. 1378 01:05:29.630 --> 01:05:32.789 Again, better is you're putting a lot on that word better. 1379 01:05:32.829 --> 01:05:34.210 Like in what way? In comfort. 1380 01:05:34.269 --> 01:05:34.909 Materially. 1381 01:05:35.070 --> 01:05:35.289 Sure. 1382 01:05:35.630 --> 01:05:38.739 In being protected from every single thing like the food 1383 01:05:38.780 --> 01:05:42.519 they eat. Every single thing is so much easier. 1384 01:05:43.099 --> 01:05:44.159 I just go ahead, Matt. 1385 01:05:44.670 --> 01:05:47.510 Well, no, I just something what one of you was 1386 01:05:47.550 --> 01:05:49.699 saying a moment ago helped, I think, crystallize for me 1387 01:05:49.800 --> 01:05:52.699 why I'm feeling more optimistic. And I feel like this 1388 01:05:52.840 --> 01:05:57.570 is sort of a middle way in between the kinds 1389 01:05:57.610 --> 01:05:59.210 of lines that you guys are pushing. I might be 1390 01:05:59.230 --> 01:06:03.059 wrong about that. So maybe I'm just crazy. But. Think 1391 01:06:03.099 --> 01:06:07.699 about things like smoking or, more broadly, physical health and 1392 01:06:07.739 --> 01:06:10.599 the way that the discourse around that and our cultural 1393 01:06:10.659 --> 01:06:14.809 attitudes have shifted in our own lifetimes. These things move slowly. 1394 01:06:14.829 --> 01:06:17.630 You can't wave a magic wand and make them happen overnight. 1395 01:06:18.989 --> 01:06:24.289 there's an understanding today about the dangers and downsides of 1396 01:06:24.349 --> 01:06:26.989 processed foods that didn't exist when we were kids in 1397 01:06:27.010 --> 01:06:29.610 the eighties. Right. And now there's a lot of people, 1398 01:06:29.849 --> 01:06:32.269 maybe not everybody, maybe not even a majority. I don't know, 1399 01:06:32.389 --> 01:06:35.639 but a lot of people for whom the choice to 1400 01:06:35.969 --> 01:06:39.960 walk more and to eat whole foods and blah, blah, blah, blah, blah. Like, 1401 01:06:40.280 --> 01:06:43.219 like that's, that's a lifestyle that has been recognized as like, Oh, 1402 01:06:43.239 --> 01:06:45.519 there's a lot going for this. And it feels like 1403 01:06:45.539 --> 01:06:49.139 what we're starting to see with AI in 2026 are 1404 01:06:49.159 --> 01:06:51.940 the seeds of a similar kind of shift being planted 1405 01:06:52.340 --> 01:06:55.119 where it's like, oh, yeah, we can do this stuff, 1406 01:06:55.519 --> 01:06:58.369 but it kind of sucks. And it makes life less 1407 01:06:58.409 --> 01:07:02.070 human and it makes things less interesting and less diverse. 1408 01:07:02.269 --> 01:07:04.849 And like, yeah, we can do it, but I don't 1409 01:07:05.030 --> 01:07:08.210 want to. I'm opting out. And I think it seems 1410 01:07:08.230 --> 01:07:12.909 like this. There's reason to hope that more and more people, 1411 01:07:12.929 --> 01:07:16.079 a critical mass, if not a majority, are going to 1412 01:07:16.440 --> 01:07:19.760 choose not to let AI be how they live, how 1413 01:07:19.780 --> 01:07:22.380 they interact with the world. And I think that'd be 1414 01:07:22.400 --> 01:07:25.230 pretty good. And it feels like that's in between the... Yeah. 1415 01:07:25.329 --> 01:07:27.769 And my, like I said, the point I have probably 1416 01:07:27.829 --> 01:07:31.659 not eloquently made is that Technological progress allows people to 1417 01:07:31.719 --> 01:07:33.260 decide that, hey, you know what? I don't need to 1418 01:07:33.300 --> 01:07:36.099 eat this processed shit. I can just decide that whole 1419 01:07:36.139 --> 01:07:38.800 foods are good and I can access them because it's 1420 01:07:39.119 --> 01:07:44.449 unbelievably widely accessible for the average person. And that's where 1421 01:07:44.469 --> 01:07:48.780 we are as a species. It's astonishing that humans have 1422 01:07:48.820 --> 01:07:53.820 done this for eons, millennia. We have horrible poverty, like 1423 01:07:54.159 --> 01:07:58.340 just awful, like child mortality, illiteracy, blah, blah, blah, blah, blah, blah. 1424 01:07:58.719 --> 01:08:01.219 And then like now in the last 200 years, this 1425 01:08:01.469 --> 01:08:04.710 astonishing progress. And if you start to zoom out, you go, 1426 01:08:05.130 --> 01:08:08.269 it's amazing what has happened and what we've allowed to 1427 01:08:08.349 --> 01:08:11.710 happen in terms of human prosperity. Again, I think the 1428 01:08:11.730 --> 01:08:15.510 big problem is that it's not made people's lives richer, deeper, 1429 01:08:15.570 --> 01:08:16.109 more meaningful. 1430 01:08:16.770 --> 01:08:17.390 Or right back. 1431 01:08:18.149 --> 01:08:19.390 It's made it easier. 1432 01:08:20.069 --> 01:08:20.970 That's well said. 1433 01:08:22.109 --> 01:08:26.289 Yeah. I mean, for some, yeah, for some people, I 1434 01:08:26.329 --> 01:08:30.789 think the food thing is interesting as an analogy because the, um, like, yes, 1435 01:08:30.850 --> 01:08:35.170 like now there's all this information, right? The

internet has 1436 01:08:35.229 --> 01:08:39.270 helped people find a sort of like healthier way to 1437 01:08:39.340 --> 01:08:41.270 live in lots of ways, right? The internet as a technology, 1438 01:08:41.300 --> 01:08:42.619 I'm trying to like pin it when you said it 1439 01:08:42.739 --> 01:08:44.659 astonishing ways for people to be healthier, like with food, 1440 01:08:45.010 --> 01:08:47.109 So I'm pinning it on a technology, let's say the internet, okay. 1441 01:08:48.569 --> 01:08:52.220 But how many generations were there where we were eating 1442 01:08:52.220 --> 01:08:56.079 processed foods? I think my grandparents had TV dinners. So 1443 01:08:56.079 --> 01:08:58.960 like there's generations there of processed foods where you have 1444 01:08:59.060 --> 01:09:03.770 corporations having mass produced processed food, it's technology. And for 1445 01:09:04.090 --> 01:09:08.060 three generations, shoved that down our faces. They did, they 1446 01:09:08.069 --> 01:09:11.420 did careful experiments to see like how exactly to make 1447 01:09:11.460 --> 01:09:15.600 a Dorito taste so that you keep eating them, et cetera, blah, blah, blah. Okay. So, 1448 01:09:15.960 --> 01:09:19.140 so great. Now we can, you can find, um, you know, 1449 01:09:19.199 --> 01:09:22.060 how to eat better and like a great debate between 1450 01:09:22.100 --> 01:09:25.600 canola oil and, and beef tallow, whatever, like how to 1451 01:09:25.619 --> 01:09:25.979 make fries. 1452 01:09:26.239 --> 01:09:26.439 Okay. 1453 01:09:26.699 --> 01:09:28.939 But like, that's great. But like how many generations had, 1454 01:09:29.000 --> 01:09:31.180 you know, how do we get this obesity crisis? 1455 01:09:31.199 --> 01:09:31.340 Right? 1456 01:09:31.359 --> 01:09:32.300 Like there's all these things there. 1457 01:09:32.340 --> 01:09:32.539 Okay. 1458 01:09:32.600 --> 01:09:34.800 It's crazy that that's the problem in America. 1459 01:09:35.060 --> 01:09:36.800 Well, so So, like, that's an example. 1460 01:09:36.819 --> 01:09:39.479 We're so successful that everyone's overweight, and now GLP-1s have 1461 01:09:39.539 --> 01:09:40.220 solved it. Anyway. 1462 01:09:40.260 --> 01:09:42.060 But I also think of, like, going back further to 1463 01:09:42.100 --> 01:09:45.760 these older technologies, like the Industrial Revolution. Like, you know, 1464 01:09:45.840 --> 01:09:48.550 suddenly we could mass produce things. Is that better? 1465 01:09:48.560 --> 01:09:49.590 I don't know. Maybe. 1466 01:09:49.859 --> 01:09:52.770 But, like, how many people's lives, like, were the people who, 1467 01:09:52.989 --> 01:09:55.229 on the first three generations, who moved to cities from 1468 01:09:55.270 --> 01:09:57.829 the farm to work in a factory, were their lives better? 1469 01:09:57.850 --> 01:10:00.229 I recommend you read some 19th century literature to answer 1470 01:10:00.270 --> 01:10:03.989 that question. Spoiler alert, no. And like that. And then 1471 01:10:04.050 --> 01:10:07.149 also like Eli Whitney, the cotton gin did a little 1472 01:10:07.180 --> 01:10:13.020 research today, 18th century invention, right? It raised the, it made, 1473 01:10:13.260 --> 01:10:17.100 it raised the GDP. It was like 50% of the 1474 01:10:17.119 --> 01:10:21.550 GDP was cotton after the cotton gin by 1850. it 1475 01:10:21.590 --> 01:10:24.630 also increased slavery, right? Because they needed more people. It 1476 01:10:24.670 --> 01:10:27.329 was more efficient to process this cotton, but they needed 1477 01:10:27.350 --> 01:10:30.069 more people to pick, et cetera. My point is that 1478 01:10:30.739 --> 01:10:33.039 I would love it if with AI, we could skip 1479 01:10:33.100 --> 01:10:36.600 those three generations of torture for people and actually get 1480 01:10:36.640 --> 01:10:38.979 to a position where we're making choices and saying, we 1481 01:10:39.020 --> 01:10:42.159 like this technology for this. We reject it here. Let's 1482 01:10:42.199 --> 01:10:45.050 take care of our people. Let's make sure everyone is okay. 1483 01:10:46.470 --> 01:10:48.470 Yeah, I mean, I'm with you on all that. I 1484 01:10:48.489 --> 01:10:50.390 just think that in the long run, the arc is, 1485 01:10:51.140 --> 01:10:55.000 generally towards more prosperity for all, for all humanity. The 1486 01:10:55.039 --> 01:10:58.439 question becomes, how do we get people to focus on 1487 01:10:58.460 --> 01:11:01.600 what really matters, which is what we've known since Time Memoriam, 1488 01:11:01.619 --> 01:11:04.729 which is the reason I did this podcast, which is 1489 01:11:04.850 --> 01:11:07.229 I love hanging out with you two because you're two 1490 01:11:07.250 --> 01:11:10.270 of the smartest, most interesting people I know. Your political 1491 01:11:10.310 --> 01:11:13.229 beliefs might not be perfectly aligned, but you guys get 1492 01:11:13.270 --> 01:11:15.479 along and this is what matters in life. It's like 1493 01:11:15.520 --> 01:11:15.979 this stuff. 1494 01:11:16.229 --> 01:11:16.460 Right? 1495 01:11:16.899 --> 01:11:17.199 Yeah. 1496 01:11:17.399 --> 01:11:20.020 And can you imagine using AI to replace that? 1497 01:11:20.079 --> 01:11:20.380 I can't. 1498 01:11:21.109 --> 01:11:26.560 Yeah in fact coming soon three questions three drinks with 1499 01:11:26.600 --> 01:11:27.680 my two chat pod. 1500 01:11:27.600 --> 01:11:31.380 Oh my god well we had a good run everybody 1501 01:11:31.399 --> 01:11:33.899 um hey how about if we get into a three 1502 01:11:33.939 --> 01:11:36.489 minutes a half sounds good anybody happy i. 1503 01:11:36.430 --> 01:11:38.489 Enjoyed this conversation. 1504 01:11:37.909 --> 01:11:40.590 Today me too i love it i am. 1505 01:11:40.529 --> 01:11:43.909 A relentless optimist and i know that humanity is not 1506 01:11:43.949 --> 01:11:45.930 perfect but i am an optimist in the long run 1507 01:11:46.170 --> 01:11:47.529 me too You are? 1508 01:11:47.600 --> 01:11:49.640 I mean, I don't like the idea

because it's passive, 1509 01:11:49.739 --> 01:11:52.180 but I believe in hope. Can I just quote the 1510 01:11:52.239 --> 01:11:54.960 great Rebecca Solnit who says, I'm going to misquote her, 1511 01:11:55.020 --> 01:11:58.079 but it's generally this. Hope is not a lottery ticket 1512 01:11:58.119 --> 01:12:00.119 you clutch and sit on the couch. Hope is an 1513 01:12:00.260 --> 01:12:03.840 ax you break down doors with. In other words, take action. 1514 01:12:03.960 --> 01:12:05.760 Make the thing you want to happen happen. If you 1515 01:12:05.779 --> 01:12:08.100 just sit there and say, oh, that looks great, it's 1516 01:12:08.119 --> 01:12:08.619 not going to happen. 1517 01:12:09.739 --> 01:12:16.409 I want more people to have an easier... life what 1518 01:12:16.430 --> 01:12:18.840 they do with it it's up to them And they 1519 01:12:18.859 --> 01:12:21.020 need people like you two to figure out how to 1520 01:12:21.039 --> 01:12:21.619 have a deeper. 1521 01:12:21.859 --> 01:12:24.979 Oh, you're so kind. Thank you. So three minutes, three 1522 01:12:25.000 --> 01:12:26.060 minutes of happiness. Yeah. 1523 01:12:26.239 --> 01:12:26.550 All right. 1524 01:12:26.560 --> 01:12:28.470 What are you guys happy about? What's going on? 1525 01:12:28.479 --> 01:12:32.590 I'll start. Um, so Dolly Parton died a few days ago. Um, 1526 01:12:32.770 --> 01:12:33.289 and I feel like. 1527 01:12:33.229 --> 01:12:37.109 We should, we should go more universally beloved. I know. Right. And, um, 1528 01:12:37.270 --> 01:12:38.930 and everyone should live their life. Totally. 1529 01:12:38.970 --> 01:12:40.649 And so many people have said such wonderful things about her, 1530 01:12:40.670 --> 01:12:42.529 but I just want to say like just a little, 1531 01:12:42.750 --> 01:12:45.750 little personal journey of mine. We came up in the 1532 01:12:46.050 --> 01:12:48.050 age in the eighties and nineties where Dolly Parton was 1533 01:12:48.090 --> 01:12:48.449 a joke. 1534 01:12:48.850 --> 01:12:49.449 I think. Right. 1535 01:12:49.510 --> 01:12:52.350 If you were outside of country music, like she was 1536 01:12:52.390 --> 01:12:55.270 a joke on Johnny Carson because of her boobs. She 1537 01:12:55.350 --> 01:13:00.140 was like, she had boobs. My favorite tweet after her 1538 01:13:00.159 --> 01:13:05.899 death was ladies, lower your titties to half staff. Sorry. 1539 01:13:05.920 --> 01:13:06.180 Sorry. 1540 01:13:06.319 --> 01:13:07.699 And there's that E rating again. 1541 01:13:07.960 --> 01:13:12.279 Obviously not my way of expressing things, but hats off 1542 01:13:12.300 --> 01:13:18.420 to that. Anyway, or something like that. Anyway, what I 1543 01:13:18.439 --> 01:13:20.539 was going to say was I grew, you know, I 1544 01:13:20.579 --> 01:13:23.539 was a young feminist in high school. I read Susan 1545 01:13:23.560 --> 01:13:27.439 Faludi's Backlash, The War Against American Women. I modeled myself 1546 01:13:27.500 --> 01:13:30.159 in the sort of version of feminism that was like 1547 01:13:30.180 --> 01:13:33.979 Gloria Steinem, whatever. Um, so I was like deeply skeptical 1548 01:13:34.020 --> 01:13:36.649 of Dolly Parton as a woman. Okay. For a long time. 1549 01:13:36.989 --> 01:13:40.010 It wasn't until, because she was everything, not that right. 1550 01:13:40.109 --> 01:13:42.670 Like she seemed at least in what I had consumed 1551 01:13:42.710 --> 01:13:45.449 as a kid, seeing her perform is like, is like 1552 01:13:45.470 --> 01:13:47.149 the opposite of those things. 1553 01:13:47.430 --> 01:13:47.850 Yeah. 1554 01:13:50.619 --> 01:13:53.609 Obviously, that's so not true. And I'm so happy that 1555 01:13:54.109 --> 01:13:56.010 in my own life, I came to have a much 1556 01:13:56.069 --> 01:13:59.109 broader understanding of what feminism could look like, what women 1557 01:13:59.149 --> 01:14:01.789 look like, and just sort of seeing the nuances and 1558 01:14:01.810 --> 01:14:03.750 the complications of that. And I came to it through 1559 01:14:03.770 --> 01:14:07.829 in the last 10, 15 years of really liking folk 1560 01:14:07.869 --> 01:14:11.029 and the sort of bluegrass side of country and just 1561 01:14:11.100 --> 01:14:13.859 first of all what a genius musician she was both 1562 01:14:13.899 --> 01:14:16.079 in terms of songwriting but also her her playing of 1563 01:14:16.159 --> 01:14:19.739 various instruments she is picking and pure total genius in 1564 01:14:19.779 --> 01:14:22.880 addition to being obviously a show woman and like brilliant 1565 01:14:22.899 --> 01:14:25.180 in all these ways so i'm really happy um that 1566 01:14:25.260 --> 01:14:28.149 i didn't that she she that i was able to 1567 01:14:28.159 --> 01:14:30.770 sort of open up my own small mind about those 1568 01:14:30.810 --> 01:14:32.630 things and and uh appreciate. 1569 01:14:32.289 --> 01:14:34.270 Her one maybe that's how you play with ai someday 1570 01:14:36.149 --> 01:14:36.960 I just ruined it. 1571 01:14:38.510 --> 01:14:38.850 I quit. 1572 01:14:41.989 --> 01:14:43.189 Next up, AI Amanda. 1573 01:14:43.210 --> 01:14:49.170 Literally, Amanda's about to shiv me right now. Next fan. 1574 01:14:49.210 --> 01:14:52.789 It's going to be a Max Kedrum version of Amanda Adams. 1575 01:14:53.869 --> 01:14:54.350 Oh, man. 1576 01:14:56.210 --> 01:14:56.670 Sorry. 1577 01:14:56.689 --> 01:15:00.680 I have one. So we used to, I feel like... 1578 01:15:01.779 --> 01:15:03.899 Excuse me, my microphone just fell down. I feel like 1579 01:15:03.920 --> 01:15:06.680 in the early days of 3Q3D, we would recommend like 1580 01:15:06.779 --> 01:15:09.579 books and concerts and things like that. And I thought about, 1581 01:15:10.079 --> 01:15:12.770 I am going to recommend a musical artist and a 1582 01:15:12.930 --> 01:15:16.630 song in particular that I really have gotten into. And 1583 01:15:16.829 --> 01:15:18.810 it's thanks to my kids. My kids are much cooler 1584 01:15:18.850 --> 01:15:20.970 than

me. They're younger. I have four kids and they're 1585 01:15:21.029 --> 01:15:22.359 all in their teens or 20s. 1586 01:15:23.800 --> 01:15:25.680 Wait, did you say my kids are much younger than 1587 01:15:25.720 --> 01:15:27.609 me a minute ago? Was that the fourth one? Yes, 1588 01:15:27.680 --> 01:15:27.960 it did. 1589 01:15:28.000 --> 01:15:29.159 That's the third drink. 1590 01:15:30.779 --> 01:15:33.210 That's good, right? Because it would be weird if they weren't. 1591 01:15:33.569 --> 01:15:36.130 My kids are much younger than me, folks. 1592 01:15:37.329 --> 01:15:38.069 You're so wise. 1593 01:15:38.090 --> 01:15:45.649 Yes. Anyway, the kids these days are into Noah Kahn. 1594 01:15:45.930 --> 01:15:46.649 Do you know Noah Kahn? 1595 01:15:46.770 --> 01:15:49.760 I know of him. My students wear t-shirts. I should listen. 1596 01:15:49.819 --> 01:15:52.000 His last name is spelled K. If you don't know 1597 01:15:52.060 --> 01:15:59.420 Noah Kahn, K-A-H-A-N. It's like Kahan. Noah Kahn. K-A-H-A-N. Look 1598 01:15:59.460 --> 01:16:03.359 up the song All Them Horses. It's called All Them Horses. 1599 01:16:04.180 --> 01:16:06.390 If you want to, the better version is the live 1600 01:16:06.399 --> 01:16:09.729 version on YouTube where he's sitting around a campfire playing 1601 01:16:09.750 --> 01:16:15.050 with his bandmates. It is really great. It's such a 1602 01:16:15.149 --> 01:16:18.710 great song. Especially the live version when the woman picks 1603 01:16:18.789 --> 01:16:23.420 up the little violin and it kicks in. It's awesome. 1604 01:16:24.159 --> 01:16:26.600 Noah Kahn, he's a throwback. Amanda, I think you'd really 1605 01:16:26.640 --> 01:16:28.880 like him. I know I would. Because he's a singer-songwriter, 1606 01:16:28.920 --> 01:16:29.279 a little folk. 1607 01:16:29.319 --> 01:16:29.859 No, it's my style. 1608 01:16:29.979 --> 01:16:34.210 It's your style. You'd love him. He's got such great stuff. 1609 01:16:34.810 --> 01:16:37.829 But all them horses, Noah Kahn, listen to it. 1610 01:16:38.579 --> 01:16:40.779 Love it. Uh, and I'm going to, I'm going to 1611 01:16:40.800 --> 01:16:44.770 exercise hosting privileges and add a recommendation to my moment 1612 01:16:44.779 --> 01:16:46.510 of happiness because it's something Amanda and I were talking 1613 01:16:46.520 --> 01:16:48.850 about during one of the, when Chris was making his drink. Um, 1614 01:16:48.909 --> 01:16:51.470 there's a terrific article in the Atlantic people. If you 1615 01:16:51.529 --> 01:16:54.670 are interested in this stuff, um, the end of reading 1616 01:16:54.750 --> 01:16:57.670 is here. Um, totally worth looking up, check it out. 1617 01:16:58.090 --> 01:16:59.750 We'll get Chris to put a free link in the, uh, 1618 01:16:59.829 --> 01:17:03.229 the show notes for this episode. And, um, but what 1619 01:17:03.250 --> 01:17:05.130 I wanted to recommend or sorry, what my moment of 1620 01:17:05.149 --> 01:17:07.390 happiness is, I guess it's also

a recommendation. Um, We've 1621 01:17:07.409 --> 01:17:10.100 been trying in our house for quite some time to 1622 01:17:10.720 --> 01:17:14.460 do things that are more human. And that's obviously relevant 1623 01:17:14.500 --> 01:17:17.670 to the topic of our conversation today. And one thing 1624 01:17:17.689 --> 01:17:20.449 we started a couple of months ago, we're only a 1625 01:17:21.060 --> 01:17:22.899 third or so of the way through it, but there's 1626 01:17:22.960 --> 01:17:27.159 an Anglican poet named Malcolm Gite. This guy is fascinating. 1627 01:17:27.220 --> 01:17:32.079 He's taught literature at Cambridge University. He is a blues musician, 1628 01:17:32.939 --> 01:17:37.430 and he's a really brilliant poet and theologian and just 1629 01:17:37.470 --> 01:17:41.619 a remarkable human being. And he has recently started publishing 1630 01:17:41.640 --> 01:17:47.279 a multi-volume retelling of the Arthurian. That is the story 1631 01:17:47.319 --> 01:17:49.199 of King Arthur and the Holy Grail and all that. 1632 01:17:50.220 --> 01:17:51.939 And so the first volume was published a couple of 1633 01:17:51.970 --> 01:17:54.409 months ago. And it's intended to be read out loud. 1634 01:17:55.130 --> 01:17:57.710 And we as a family have been occasionally, not as 1635 01:17:57.770 --> 01:17:59.770 often as we'd like, but we've been working our way 1636 01:17:59.869 --> 01:18:03.149 through Galahad and the Grail. And it is just a 1637 01:18:03.369 --> 01:18:08.050 deeply human thing to be reading the story of Sir 1638 01:18:08.069 --> 01:18:10.970 Galahad and his quest for the Holy Grail out loud 1639 01:18:11.069 --> 01:18:13.010 as a family. It's beautiful. You should do it, everybody. 1640 01:18:13.850 --> 01:18:18.260 Get a book and read it. You'll like it. That's 1641 01:18:18.300 --> 01:18:20.039 my moment of happiness. It's the time I've spent this 1642 01:18:20.060 --> 01:18:22.399 summer reading Galahad and the Grail with my family. But 1643 01:18:22.439 --> 01:18:25.380 now let's go to our final toast, shall we? Yes. 1644 01:18:25.840 --> 01:18:30.159 So tonight's toast, today's toast, this morning's toast, whenever you 1645 01:18:30.180 --> 01:18:35.609 listen to this, is a quote. Again, relevant to our 1646 01:18:35.649 --> 01:18:41.010 topic today, the current Pope, Leo XIV, recently issued an 1647 01:18:41.050 --> 01:18:45.319 encyclical called Magnifica Humanitas, which just means. 1648 01:18:45.420 --> 01:18:46.750 As everyone knows, man, of course. 1649 01:18:46.770 --> 01:18:51.979 Which just means magnificent humanity. And the subtitle is something like, 1650 01:18:53.800 --> 01:18:57.300 on remaining human in the age of artificial intelligence, something 1651 01:18:57.319 --> 01:19:00.760 like that. Safeguarding humanity in the age of artificial intelligence. 1652 01:19:01.640 --> 01:19:05.560 And there's a line from early in that encyclical that 1653 01:19:05.579 --> 01:19:07.220 I want to use as our toast, where the Holy 1654 01:19:07.260 --> 01:19:12.119 Father exhorts us, and I quote... Ours is the pressing 1655 01:19:12.199 --> 01:19:17.420 duty to remain profoundly human. Ah. Cheers. 1656 01:19:17.439 --> 01:19:18.199 I can drink to that. 1657 01:19:18.439 --> 01:19:21.890 I'll drink to that. Cheers, Matt. Well done. 1658 01:19:23.550 --> 01:19:26.890 Well, folks, if you enjoy this kind of thought-provoking conversation, 1659 01:19:27.329 --> 01:19:29.630 if you find us as charming and witty as we do, 1660 01:19:30.130 --> 01:19:33.090 then please be sure to like, share, and subscribe. Give 1661 01:19:33.109 --> 01:19:36.170 us that five-star review. Tell a friend. Tell the world 1662 01:19:36.529 --> 01:19:42.399 that 3Q3D is... The podcast for, I don't know, whatever 1663 01:19:42.439 --> 01:19:42.840 this is. 1664 01:19:43.680 --> 01:19:44.199 For humans. 1665 01:19:44.319 --> 01:19:44.779 For humans. 1666 01:19:44.840 --> 01:19:45.779 And it's made by humans. 1667 01:19:45.880 --> 01:19:47.319 It's made by humans. All of it's human. 1668 01:19:47.560 --> 01:19:50.300 It's 100% authentically made by human beings. 1669 01:19:52.390 --> 01:19:54.689 Robots would never make this kind of repetitive error that 1670 01:19:54.710 --> 01:19:57.189 we're making. So thanks so much for listening, everybody. We 1671 01:19:57.229 --> 01:19:59.670 hope we'll see you or hear you. Talk to you 1672 01:19:59.710 --> 01:20:02.689 next time. Great. Thanks, Matt. Well done. 1673 01:20:02.710 --> 01:20:03.329 That was great.