1 00:00:00,160 --> 00:00:13,784 Foreign welcome to Emerging excellence and it's our pleasure 2 00:00:13,832 --> 00:00:16,520 today to welcome Sam Malakargalan. 3 00:00:16,600 --> 00:00:19,032 And Sam, you're coming in from the beautiful usa. 4 00:00:19,096 --> 00:00:23,880 You're the general manager at Agent AI, which I think typifies 5 00:00:23,960 --> 00:00:28,178 some of the most interesting and kind of, kind of perhaps paints 6 00:00:28,194 --> 00:00:32,546 a really interesting future picture of what technology is coming 7 00:00:32,618 --> 00:00:35,330 down the pipeline and so opening up some of the opportunities. 8 00:00:35,410 --> 00:00:38,562 So first of all, Sam, a huge welcome and thanks so much for being 9 00:00:38,586 --> 00:00:39,442 on the show. 10 00:00:39,626 --> 00:00:40,434 Thank you for having me. 11 00:00:40,442 --> 00:00:41,630 I'm looking forward to it. 12 00:00:42,090 --> 00:00:46,642 So, Sam, I think for most people agent AI would be a completely 13 00:00:46,706 --> 00:00:47,522 foreign concept. 14 00:00:47,586 --> 00:00:50,050 I certainly am no whiz in this space. 15 00:00:50,170 --> 00:00:53,666 And I think if you could break it down for us for a second. 16 00:00:53,818 --> 00:00:58,630 What, what is agent AI and what is an AI agent more broadly? 17 00:00:59,210 --> 00:01:01,522 Yeah, it's a great example of why we should never let engineers 18 00:01:01,586 --> 00:01:02,482 name things. 19 00:01:02,666 --> 00:01:04,418 ChatGPT is a meaningless thing. 20 00:01:04,474 --> 00:01:08,070 Calling it an AI agent is a technical distinction. 21 00:01:08,730 --> 00:01:12,722 AI we're all kind of familiar with it's, you know, lead scoring. 22 00:01:12,786 --> 00:01:14,310 We can use it to predict the weather. 23 00:01:15,130 --> 00:01:19,378 The difference in AI that you're used to, you know, maybe it's 24 00:01:19,394 --> 00:01:23,582 generating ads or whatever and agents is that agents have gotten 25 00:01:23,646 --> 00:01:25,918 so good at predicting what's going to come next that we started 26 00:01:25,974 --> 00:01:27,582 trusting them to just do it. 27 00:01:27,766 --> 00:01:30,206 So I can say, respond to this email for me. 28 00:01:30,278 --> 00:01:33,326 You've seen autocomplete same general concept, but instead of auto 29 00:01:33,358 --> 00:01:36,542 completing a sentence, you can autocomplete an essay or a book or 30 00:01:36,566 --> 00:01:39,230 a blog article or an entire email. 31 00:01:39,350 --> 00:01:43,934 So an AI agent is just a piece of software that's using what's called 32 00:01:43,942 --> 00:01:48,270 a large language model, usually to try and predict how you 33 00:01:48,310 --> 00:01:50,942 would respond given as much kind of context as you can give it. 34 00:01:51,046 --> 00:01:53,130 And then we call it an agent because it has agency. 35 00:01:53,160 --> 00:01:55,862 Agency not in the philosophical sense, but in like 36 00:01:55,886 --> 00:01:56,534 the literal sense. 37 00:01:56,582 --> 00:01:58,934 Like it can actually go do stuff on your behalf like a buying 38 00:01:58,982 --> 00:02:00,930 agent or a human would. 39 00:02:01,230 --> 00:02:03,958 We now have these armies of little AI agents that go do stuff 40 00:02:03,974 --> 00:02:05,010 for us all the time. 41 00:02:05,870 --> 00:02:07,574 So you've got a very interesting. 42 00:02:07,622 --> 00:02:09,846 So your background, if we just touch on that briefly. 43 00:02:09,878 --> 00:02:13,622 So you're ex HubSpot and you were kind of super early in that 44 00:02:13,646 --> 00:02:16,886 game, in that game that's gone on to become, you know, obviously 45 00:02:16,918 --> 00:02:22,392 a massive company these days in terms of like where did. 46 00:02:22,496 --> 00:02:22,808 What. 47 00:02:22,864 --> 00:02:26,776 What was in your background or your kind of story which had you 48 00:02:26,848 --> 00:02:27,976 focus on this piece. 49 00:02:28,008 --> 00:02:29,784 What, and particularly why are you so. 50 00:02:29,872 --> 00:02:33,620 Why, why are you so kind of excited about the potential here? 51 00:02:34,240 --> 00:02:36,504 So I really like unfair fights. 52 00:02:36,552 --> 00:02:38,540 I like creating unfair advantages. 53 00:02:39,120 --> 00:02:41,720 And this is just such a great, I mean, you think about Google got 54 00:02:41,760 --> 00:02:44,088 sucker punched by a NonProfit Research Company. 55 00:02:44,224 --> 00:02:46,520 OpenAI was originally a nonprofit research company. 56 00:02:46,560 --> 00:02:48,336 ChatGPT was just a demo. 57 00:02:48,488 --> 00:02:51,744 Uh, and if Google can get sucker punched, what does that mean 58 00:02:51,752 --> 00:02:54,460 for like the 800 pound gorilla in Everybody else's industry? 59 00:02:54,760 --> 00:02:58,960 Uh, but what worried me was I, I worried that we would go from having 60 00:02:59,000 --> 00:03:01,712 millions of websites to having like thousands of agents and just 61 00:03:01,736 --> 00:03:05,504 dozens of, you know, kind of orchestration agencies, enterprise 62 00:03:05,552 --> 00:03:06,176 silos. 63 00:03:06,288 --> 00:03:08,624 That would be expensive, you need technical skill, the scope would 64 00:03:08,632 --> 00:03:09,440 be limited. 65 00:03:09,600 --> 00:03:13,152 Uh, and so when Dharmesh had the idea for agent AI, it's like, 66 00:03:13,176 --> 00:03:16,276 what if we could make it so that anybody, regardless of technical 67 00:03:16,308 --> 00:03:19,572 skill, what tools you're using or your ability to afford, you know, 68 00:03:19,596 --> 00:03:22,116 AI tools, what if everybody could be a part of kind of the AI 69 00:03:22,148 --> 00:03:22,724 future? 70 00:03:22,892 --> 00:03:25,476 So not to sound too melodramatic, but we're trying to 71 00:03:25,628 --> 00:03:28,660 avoid the dystopian hellhole that everybody's worried about us 72 00:03:28,700 --> 00:03:32,164 ending up in by making it so that everybody has access to the 73 00:03:32,172 --> 00:03:33,108 same tools. 74 00:03:33,284 --> 00:03:36,120 And the only thing limiting use your creativity, not your budget. 75 00:03:37,020 --> 00:03:38,116 Such a cool idea. 76 00:03:38,148 --> 00:03:40,596 And I think, I think we can definitely can see that happening. 77 00:03:40,628 --> 00:03:40,868 Right? 78 00:03:40,924 --> 00:03:46,690 Especially as they talk about how there's going to be those who 79 00:03:46,730 --> 00:03:49,602 understand AI and those who don't, and it's just going to be 80 00:03:49,626 --> 00:03:51,790 basically a free for all for those who do. 81 00:03:52,570 --> 00:03:54,034 And I think it's very cool. 82 00:03:54,162 --> 00:03:57,058 Some of the examples we just touched on just before we jumped 83 00:03:57,074 --> 00:04:01,026 on in terms of what people are using technology for. 84 00:04:01,098 --> 00:04:02,338 So just talk us through. 85 00:04:02,394 --> 00:04:04,706 Obviously we've got our construction and infrastructure bent 86 00:04:04,738 --> 00:04:05,250 in terms of. 87 00:04:05,290 --> 00:04:07,394 That's where we kind of focus right now. 88 00:04:07,562 --> 00:04:10,114 Tell us from your perspective, what are some of the interesting 89 00:04:10,162 --> 00:04:14,414 applications that you've seen people develop in the space can actually 90 00:04:14,502 --> 00:04:16,850 use agent AI for at the moment? 91 00:04:17,190 --> 00:04:19,438 Yeah, I love people who use it for communication. 92 00:04:19,614 --> 00:04:23,166 One of my favorite stories, I had a friend, if you don't tell AI 93 00:04:23,198 --> 00:04:25,566 to use a certain tone, you can kind of tell it's AI. 94 00:04:25,758 --> 00:04:28,574 So I had a friend who, a couple weeks ago, I was asking, are 95 00:04:28,582 --> 00:04:31,454 you writing, using AI to write your text messages to me? 96 00:04:31,622 --> 00:04:34,750 And he was, he's got some, you know, learning communication disabilities 97 00:04:34,830 --> 00:04:38,606 and he was using AI to turn his, like, long rants into the types 98 00:04:38,638 --> 00:04:41,290 of content that his friends and family would, would want to see. 99 00:04:41,910 --> 00:04:43,438 I love people who use it for learning. 100 00:04:43,534 --> 00:04:46,630 So the other night I couldn't sleep and I spent four hours having 101 00:04:46,670 --> 00:04:51,142 AI teach me some Weird nuance about quantum mechanics and it no 102 00:04:51,166 --> 00:04:51,622 human one. 103 00:04:51,646 --> 00:04:53,910 I couldn't afford to have a human teacher in my room at 4am, 104 00:04:53,950 --> 00:04:55,410 nor would I want to. 105 00:04:55,790 --> 00:04:57,846 Also, no human teacher would have that kind of patience. 106 00:04:57,958 --> 00:05:00,294 But it sat there, it's like, eventually it's like, oh, I see why 107 00:05:00,302 --> 00:05:01,410 you're not getting this. 108 00:05:01,710 --> 00:05:03,510 So using it to learn new skills. 109 00:05:03,590 --> 00:05:04,250 Wow. 110 00:05:04,750 --> 00:05:09,318 And then using it to give you skills like AI is very good at being 111 00:05:09,374 --> 00:05:10,410 mediocre at things. 112 00:05:10,510 --> 00:05:12,978 So it's a mediocre software engineer, it's a mediocre marketing 113 00:05:12,994 --> 00:05:14,658 professional, mediocre designer. 114 00:05:14,834 --> 00:05:17,266 But what that means is that all of us have access to mediocre 115 00:05:17,298 --> 00:05:17,874 skills. 116 00:05:18,002 --> 00:05:21,186 So like I was telling you about the, the, my local Irish pub, 117 00:05:21,258 --> 00:05:23,682 they wanted to have like a drunken spelling bee game, which 118 00:05:23,706 --> 00:05:24,626 is cool, right? 119 00:05:24,698 --> 00:05:26,470 Just something different than trivia. 120 00:05:26,890 --> 00:05:29,906 And normally I would have had to go bother my engineering friends 121 00:05:29,978 --> 00:05:32,866 to help me like set everything up, but I just used AI. 122 00:05:32,898 --> 00:05:34,178 I said, this is what I want it to do. 123 00:05:34,234 --> 00:05:37,234 Generate all the words, write all the code, deploy all the code. 124 00:05:37,362 --> 00:05:40,590 And so they have, you know, I think it's drunkenspellingbee.com 125 00:05:40,630 --> 00:05:41,694 you can probably check it out. 126 00:05:41,782 --> 00:05:46,702 It's entirely powered by AI and it's like we're all still T shaped. 127 00:05:46,766 --> 00:05:50,430 We have our area of expertise, but the width has gotten much wider. 128 00:05:50,590 --> 00:05:53,726 Sales reps can generate decks without talking to the marketing 129 00:05:53,758 --> 00:05:54,330 team. 130 00:05:55,270 --> 00:05:58,910 Individual project managers can do data analytics without having 131 00:05:58,950 --> 00:06:00,610 to bother anybody. 132 00:06:01,350 --> 00:06:04,958 I find that to be the most interesting thing, is that every 133 00:06:05,014 --> 00:06:09,236 individual person has access to a far greater number of skills 134 00:06:09,268 --> 00:06:11,892 than they used to have without having to actually learn at all. 135 00:06:12,076 --> 00:06:12,436 Yeah. 136 00:06:12,468 --> 00:06:15,556 And I think it's interesting, like from my side, I'm, I'm dyslexic 137 00:06:15,588 --> 00:06:18,708 and my spelling is average at best. 138 00:06:18,844 --> 00:06:22,708 But now, you know, I can run any post like LinkedIn or something 139 00:06:22,764 --> 00:06:25,700 through, you know, check for spelling, check for words, like, 140 00:06:25,740 --> 00:06:30,084 and it can just bring it up to a level which is significantly better 141 00:06:30,252 --> 00:06:31,764 than what it was before. 142 00:06:31,852 --> 00:06:35,492 And I think that to your point, right, the base level, I was 143 00:06:35,516 --> 00:06:38,164 talking to someone just the other day and she works in PR and 144 00:06:38,172 --> 00:06:41,344 she's like, I'm really worried about AI taking my job in pr and 145 00:06:41,352 --> 00:06:43,072 I'm like, probably will like it. 146 00:06:43,096 --> 00:06:46,048 Probably take all of our, you know, this at this base level, right? 147 00:06:46,104 --> 00:06:50,304 Because then if you don't rise to that level, there is an opportunity 148 00:06:50,352 --> 00:06:54,912 that those who, who leverage it actually could become supreme 149 00:06:54,976 --> 00:06:55,536 in many ways. 150 00:06:55,568 --> 00:06:58,208 Would you, is that, is that a fair assessment, Sam? 151 00:06:58,384 --> 00:06:59,568 Yeah, it's absolutely fair. 152 00:06:59,624 --> 00:07:02,944 It was one of My biggest fears with AI was, for example, if you 153 00:07:02,952 --> 00:07:07,136 replace all the entry level sales reps to do chat on a website 154 00:07:07,208 --> 00:07:10,492 or whatever, where do the senior sales reps of tomorrow come 155 00:07:10,516 --> 00:07:11,080 from? 156 00:07:11,620 --> 00:07:14,012 What's interesting to me is now people are building AIs to help 157 00:07:14,036 --> 00:07:15,068 people practice those. 158 00:07:15,124 --> 00:07:17,964 So instead of trying to replace a sales rep, replace a prospect. 159 00:07:18,092 --> 00:07:20,668 Let the sales rep practice against a fake prospect and learn 160 00:07:20,684 --> 00:07:22,252 the skills that they need to have. 161 00:07:22,436 --> 00:07:22,844 Wow. 162 00:07:22,892 --> 00:07:23,228 So I. 163 00:07:23,284 --> 00:07:28,732 It's definitely something where if you are early to it and 164 00:07:28,756 --> 00:07:31,516 if you can deal with the fact that it's confusing, I mean, chaos 165 00:07:31,548 --> 00:07:32,572 is an opportunity, right? 166 00:07:32,596 --> 00:07:33,532 And this is chaos right now. 167 00:07:33,556 --> 00:07:36,108 This is the most chaotic thing I've ever even studied, much less 168 00:07:36,164 --> 00:07:38,860 actually experienced because it's so fast, right. 169 00:07:38,900 --> 00:07:41,516 The Industrial revolution took a century, didn't happen everywhere 170 00:07:41,548 --> 00:07:42,188 at once. 171 00:07:42,324 --> 00:07:46,012 Even digital transformation took like 20 years before not adopting, 172 00:07:46,076 --> 00:07:49,356 not having a website put you at like a huge competitive disadvantage. 173 00:07:49,468 --> 00:07:53,676 You have like 20 months until not having some level of familiarity 174 00:07:53,708 --> 00:07:56,380 with AI puts you at a competitive disadvantage in your 175 00:07:56,420 --> 00:07:58,200 career as a company. 176 00:07:58,740 --> 00:08:01,180 The good news is everybody else is confused too. 177 00:08:01,300 --> 00:08:04,876 Like nobody really knows what's going to happen next. 178 00:08:04,948 --> 00:08:08,674 We all have suspicions, but much like the early days of the Internet, 179 00:08:08,722 --> 00:08:12,098 the App Store, like when Apple rolled out the App Store, they didn't 180 00:08:12,114 --> 00:08:13,826 anticipate Uber, right? 181 00:08:13,898 --> 00:08:17,522 But it was, they gave a platform for that innovation and 182 00:08:17,546 --> 00:08:20,338 then, you know, the Uber folks made bajillions of dollars and that's 183 00:08:20,354 --> 00:08:22,590 what's going to happen the next 5, 5 ish years. 184 00:08:23,210 --> 00:08:26,082 Tell us about the commercial model behind Agent AI. 185 00:08:26,146 --> 00:08:30,802 So let's say me as a, you know, just someone sat in Australia, 186 00:08:30,906 --> 00:08:33,682 I can jump on to, what can I, what can I do? 187 00:08:33,786 --> 00:08:37,442 What does it cost to kind of set up a little agent for myself 188 00:08:37,506 --> 00:08:39,030 or, or in my company? 189 00:08:40,130 --> 00:08:41,482 What's that look like? 190 00:08:41,666 --> 00:08:43,546 Yeah, so it doesn't cost anything. 191 00:08:43,738 --> 00:08:47,242 And the reason it doesn't cost anything is because the costs to 192 00:08:47,266 --> 00:08:49,418 us are going down dramatically over time. 193 00:08:49,474 --> 00:08:52,186 So the performance of AI models is increasing really fast. 194 00:08:52,298 --> 00:08:55,530 The cost of AI models is decreasing really fast. 195 00:08:55,690 --> 00:08:59,818 And it would be kind of like if Squarespace or a website provider 196 00:08:59,834 --> 00:09:02,314 was still charging you per website visit, even though cloud 197 00:09:02,362 --> 00:09:03,770 hosting is ridiculously cheap. 198 00:09:03,850 --> 00:09:05,870 That would just be a really bad business model. 199 00:09:06,210 --> 00:09:09,394 And then the bigger thing for us is every, we've got all these 200 00:09:09,402 --> 00:09:13,474 people, we've got, I think 3,000, 4,000 people on the site who 201 00:09:13,482 --> 00:09:16,498 are building agents and over 1.1 million people who are using 202 00:09:16,554 --> 00:09:17,270 agents. 203 00:09:18,250 --> 00:09:21,794 And eventually you're going to want to just like the App Store be 204 00:09:21,802 --> 00:09:23,682 able to allow people to make money off of it. 205 00:09:23,786 --> 00:09:27,122 The benefit to us is that AI gets smarter the more data is run 206 00:09:27,146 --> 00:09:27,682 through it. 207 00:09:27,786 --> 00:09:31,442 So the more people create and use AI agents, we're going to be 208 00:09:31,466 --> 00:09:33,560 the best place to help you. 209 00:09:33,730 --> 00:09:36,684 You know, our AI's job is to know you really well in your job, 210 00:09:36,772 --> 00:09:39,228 your company, your competitors, your style. 211 00:09:39,404 --> 00:09:42,092 And then based on what you're trying to do, we're going to tell 212 00:09:42,116 --> 00:09:45,436 you use this agent, give it that information, then use that agent's 213 00:09:45,468 --> 00:09:47,036 output to do something else. 214 00:09:47,148 --> 00:09:50,268 Like our AI agent is learning from everybody who's using it. 215 00:09:50,404 --> 00:09:53,964 And I don't care if you're Google, Microsoft or anybody else. 216 00:09:54,012 --> 00:09:57,612 Like no amount of money can buy time, can buy observations of 217 00:09:57,636 --> 00:09:58,920 humans doing things. 218 00:09:59,300 --> 00:10:04,040 And so that's the commercial model for us is can we, can we. 219 00:10:04,080 --> 00:10:07,224 And can we also just make it like take the wind out of the sales 220 00:10:07,272 --> 00:10:08,664 for these enterprise silos? 221 00:10:08,712 --> 00:10:12,728 You know, love my dear friends at Microsoft and Salesforce and such, 222 00:10:12,784 --> 00:10:16,472 but I don't want to see a world in which AI is only useful 223 00:10:16,536 --> 00:10:19,896 in the context of specific tools where you have to have technical 224 00:10:19,928 --> 00:10:23,544 skills to use it, or more importantly where you have to have 225 00:10:23,552 --> 00:10:24,440 a lot of budget to use it. 226 00:10:24,480 --> 00:10:24,632 Right. 227 00:10:24,656 --> 00:10:26,216 Like I want my. 228 00:10:26,368 --> 00:10:27,336 I kid you not. 229 00:10:27,408 --> 00:10:31,032 My 16 year old nephew living in rural Tennessee here in the US 230 00:10:31,136 --> 00:10:34,400 is one of the most prolific AI agent builders on our site. 231 00:10:35,020 --> 00:10:38,756 And he doesn't have budget or anything else like that, but he's 232 00:10:38,788 --> 00:10:42,436 able to be competitive with anybody he wants to try and be competitive 233 00:10:42,468 --> 00:10:42,948 with. 234 00:10:43,084 --> 00:10:44,564 Those are the stories I love. 235 00:10:44,732 --> 00:10:48,836 Level, let's say I jump on, I start to create my own thing. 236 00:10:48,908 --> 00:10:50,100 Can I sell that? 237 00:10:50,220 --> 00:10:56,292 Is it possible to sell that as functionality to a client or something 238 00:10:56,316 --> 00:10:56,996 like that? 239 00:10:57,148 --> 00:10:59,092 We will eventually support that. 240 00:10:59,276 --> 00:11:01,640 Keeping in mind that we just launched last May. 241 00:11:01,820 --> 00:11:04,416 Technically we launched in last September at HubSpot's inbound 242 00:11:04,448 --> 00:11:05,140 conference. 243 00:11:05,640 --> 00:11:08,768 But that is the idea is we want to make it so that you can either 244 00:11:08,784 --> 00:11:10,240 make it public, you can keep it private. 245 00:11:10,280 --> 00:11:13,152 If you want to just use it for yourself, that's fine, you can make 246 00:11:13,176 --> 00:11:14,864 it public if you want to just do that. 247 00:11:14,952 --> 00:11:19,376 Or you can charge a monthly fee, you can charge just a lifetime 248 00:11:19,408 --> 00:11:21,648 license just like you would with any app store. 249 00:11:21,824 --> 00:11:24,240 There's all these companies that make tons of money off the app 250 00:11:24,280 --> 00:11:27,456 store and that's what we really want to get to. 251 00:11:27,528 --> 00:11:29,780 It's just quite a complex problem to solve. 252 00:11:30,450 --> 00:11:34,794 So right now it's one of the reasons we're about actually by the 253 00:11:34,802 --> 00:11:36,314 time this comes out we'll have Announced it. 254 00:11:36,402 --> 00:11:39,562 We're launching a $1 million grant fund to subsidize builders 255 00:11:39,626 --> 00:11:42,842 building agents simply because we can't yet figure out how to do 256 00:11:42,946 --> 00:11:45,642 the monetization between, like, somebody uses your agent, you 257 00:11:45,666 --> 00:11:46,378 actually get paid. 258 00:11:46,474 --> 00:11:48,794 So we're just giving people grants to build really cool agents. 259 00:11:48,842 --> 00:11:51,706 And we know that the enterprise value to us will come, 260 00:11:51,778 --> 00:11:52,790 will follow that. 261 00:11:53,170 --> 00:11:53,962 Interesting. 262 00:11:54,026 --> 00:11:54,522 Very cool. 263 00:11:54,586 --> 00:11:56,026 So let's talk about construction. 264 00:11:56,058 --> 00:11:57,018 There's a whole. 265 00:11:57,194 --> 00:12:00,336 Construction is a very tricky space. 266 00:12:00,408 --> 00:12:01,888 There's a lot of moving parts. 267 00:12:01,984 --> 00:12:04,624 There's, you know, heap of scheduling and in the, in the actual 268 00:12:04,672 --> 00:12:05,888 kind of building process. 269 00:12:05,944 --> 00:12:10,160 But then there's a whole bunch of communication going on between 270 00:12:10,280 --> 00:12:10,960 designers. 271 00:12:11,040 --> 00:12:12,816 And so they're getting information when they're digging 272 00:12:12,848 --> 00:12:15,136 holes, as an example, and they're, you know, discovering pipes 273 00:12:15,168 --> 00:12:15,952 which were never there. 274 00:12:15,976 --> 00:12:18,688 And so they have to kind of adjust designs. 275 00:12:18,784 --> 00:12:21,904 There's a whole bunch of procurement which is happening, you 276 00:12:21,912 --> 00:12:24,608 know, even before they kind of try to find the gear. 277 00:12:24,784 --> 00:12:29,692 There's this small markets like tunneling companies selling 278 00:12:29,756 --> 00:12:34,508 dirt to the road building because, you know, like, they're 279 00:12:34,524 --> 00:12:37,116 trying to get rid of it, the tunnel, but they, they need dirt 280 00:12:37,148 --> 00:12:37,452 when they. 281 00:12:37,476 --> 00:12:39,880 To go up the hill in the road project. 282 00:12:40,180 --> 00:12:43,676 What, what do you think of the opportunities within that space or 283 00:12:43,748 --> 00:12:47,052 kind of, what are the practical examples where, as you 284 00:12:47,076 --> 00:12:49,500 said, you can, the project manager is. 285 00:12:49,540 --> 00:12:52,444 No, doesn't have to bother the, the data management team or 286 00:12:52,452 --> 00:12:53,932 the, or the data analysts anymore. 287 00:12:53,996 --> 00:12:56,760 They can kind of just tap into something and have the information. 288 00:12:57,440 --> 00:13:00,936 What do you see just in terms of what you know about construction? 289 00:13:00,968 --> 00:13:02,968 You certainly don't have to be an expert here, Sam. 290 00:13:03,064 --> 00:13:06,120 But what do you think is kind of the obvious examples and maybe 291 00:13:06,200 --> 00:13:07,620 what's coming beyond that? 292 00:13:08,560 --> 00:13:09,864 There's two big buckets. 293 00:13:09,912 --> 00:13:13,512 One is, can you take things that you're already doing and make 294 00:13:13,536 --> 00:13:16,504 them more efficient, make them faster, make it so you don't have 295 00:13:16,512 --> 00:13:17,560 to do it right? 296 00:13:17,600 --> 00:13:20,584 Like, I have AI that attends meetings for me and just lets me 297 00:13:20,592 --> 00:13:22,744 know if there was anything that it thinks I would find interesting 298 00:13:22,792 --> 00:13:25,860 on that meeting rather than me having to join all of them. 299 00:13:26,570 --> 00:13:29,026 There's, you know, and I was having a conversation with somebody 300 00:13:29,058 --> 00:13:29,538 the other day. 301 00:13:29,594 --> 00:13:32,562 They're like, so is it just going to be my AI agent and your 302 00:13:32,586 --> 00:13:33,970 AI agent having a meeting? 303 00:13:34,090 --> 00:13:35,490 I'm like, is that bad? 304 00:13:35,610 --> 00:13:38,690 Is that like something we don't want as long as it has enough 305 00:13:38,730 --> 00:13:39,906 context around us? 306 00:13:40,058 --> 00:13:41,538 Same thing for buying and selling, right? 307 00:13:41,594 --> 00:13:44,450 Like, is it going to be my AI agent whose job is to buy things 308 00:13:44,490 --> 00:13:48,754 I will like and your Your agent's job to sell and they'll do 309 00:13:48,762 --> 00:13:51,810 all the nonsense, they'll do the demo call and you know, looking 310 00:13:51,850 --> 00:13:54,146 at the procurement, all the like, the details that take us tons 311 00:13:54,178 --> 00:13:54,910 of time. 312 00:13:55,370 --> 00:13:59,742 So anytime you can think about what you're doing now, is there a 313 00:13:59,766 --> 00:14:04,942 way to use the AI and use specifically these AI agents to help 314 00:14:04,966 --> 00:14:07,214 you do it better, help you not have to do it at all and help you 315 00:14:07,222 --> 00:14:08,382 do it at greater scale? 316 00:14:08,526 --> 00:14:10,770 The other thing is just net new capabilities. 317 00:14:11,190 --> 00:14:15,966 So can you like, like you said, you know, sales reps can design, 318 00:14:16,118 --> 00:14:22,782 project managers can do data analysis even, you know, AI has this 319 00:14:22,806 --> 00:14:26,862 fascinating feature or bug, depending on your perspective, called 320 00:14:26,886 --> 00:14:29,998 hallucination, where it will make up things that don't actually 321 00:14:30,054 --> 00:14:30,770 exist. 322 00:14:31,270 --> 00:14:33,774 I find that fascinating because as long as you're aware of 323 00:14:33,782 --> 00:14:38,590 it and you don't use AI to write your legal brief and then get 324 00:14:38,630 --> 00:14:40,606 made fun of by the judge because you cited a case that doesn't 325 00:14:40,638 --> 00:14:43,450 exist, which actually did happen all the time. 326 00:14:44,310 --> 00:14:44,926 Yeah. 327 00:14:45,038 --> 00:14:49,470 If the AI is hallucinating a product that your customers might 328 00:14:49,510 --> 00:14:51,610 want and you don't sell it, maybe you should. 329 00:14:52,320 --> 00:14:55,448 Or if it's hallucinating, you know, different types of companies 330 00:14:55,504 --> 00:14:59,272 that you could be selling the excess dirt to and you haven't thought 331 00:14:59,296 --> 00:15:01,320 about it, like that's a feature. 332 00:15:01,400 --> 00:15:04,760 So giving yourself so one, just make everything more efficient. 333 00:15:04,840 --> 00:15:08,504 Like we, you shouldn't do all the things that just take tons of 334 00:15:08,512 --> 00:15:10,808 time to give yourself those net new capabilities. 335 00:15:10,904 --> 00:15:13,976 Every person on your team should be a thousand times more powerful 336 00:15:14,008 --> 00:15:15,380 than they were a few years ago. 337 00:15:15,840 --> 00:15:20,284 But then three, you know, let AI be co creative with you. 338 00:15:20,372 --> 00:15:23,964 I sometimes tell it to hallucinate intentionally, just to 339 00:15:23,972 --> 00:15:27,756 see what it comes up with, either positioning or products or 340 00:15:27,828 --> 00:15:28,700 something else that we should do. 341 00:15:28,740 --> 00:15:32,268 So for the construction industry, obviously efficiency matters 342 00:15:32,284 --> 00:15:32,780 a lot. 343 00:15:32,900 --> 00:15:37,180 But then very importantly is going to be how do you manage all 344 00:15:37,220 --> 00:15:37,708 those processes? 345 00:15:37,724 --> 00:15:41,468 How do you make it so that it knows if it's going to rain during 346 00:15:41,524 --> 00:15:45,692 the project and automatically reschedules everybody and sends a 347 00:15:45,716 --> 00:15:47,004 note to everybody? 348 00:15:47,052 --> 00:15:51,172 Or how does it know that you know, a bridge collapsed and that 349 00:15:51,196 --> 00:15:53,524 means that you know something that you need isn't going to make 350 00:15:53,532 --> 00:15:55,460 it there in time, you need to reschedule things. 351 00:15:55,580 --> 00:15:58,676 All of this kind of information synthesis is incredibly 352 00:15:58,708 --> 00:15:59,476 powerful for anybody. 353 00:15:59,508 --> 00:16:01,716 But in particular, when you've got all these complex moving parts 354 00:16:01,748 --> 00:16:04,324 and you've got all these people you have to talk to, AI is 355 00:16:04,332 --> 00:16:04,916 going to be beautiful. 356 00:16:04,948 --> 00:16:09,652 For that, I feel like, especially if you're working in a, 357 00:16:09,676 --> 00:16:14,468 in a global world and you when it comes to like structural building 358 00:16:14,524 --> 00:16:18,562 up with your trading metal or even currency, if you're buying in 359 00:16:18,586 --> 00:16:22,194 US dollars but your domestic currency's Australian dollars as 360 00:16:22,202 --> 00:16:27,506 an example, then understanding some global shifts and making predictors 361 00:16:27,538 --> 00:16:30,722 where typically that would be the realms of investment banks perhaps, 362 00:16:30,786 --> 00:16:35,970 or you're having to leverage expertise elsewhere which has become 363 00:16:36,010 --> 00:16:39,346 the domain of these huge advisory firms. 364 00:16:39,538 --> 00:16:44,370 What you're saying is that the local baker who's buying flour or 365 00:16:44,410 --> 00:16:48,212 something like is going to have access to very similar information 366 00:16:48,396 --> 00:16:50,132 potentially from, from that regard. 367 00:16:50,196 --> 00:16:51,444 Is that, is that right? 368 00:16:51,612 --> 00:16:56,452 Yeah, I find that fascinating if you, you know, one of the first 369 00:16:56,476 --> 00:16:58,964 things I did was I built a McKinsey consultant in a box as a 370 00:16:58,972 --> 00:17:02,596 joke when, you know, this is a year and a half ago, but last summer 371 00:17:02,628 --> 00:17:05,652 I gave a demo to the executives at a very large bank and 372 00:17:05,676 --> 00:17:08,212 I just asked, I had an AI agent, very simple question. 373 00:17:08,236 --> 00:17:12,212 I said the price of oil went up, should I buy or sell the like 374 00:17:12,236 --> 00:17:14,196 the Canadian dollar yen pair? 375 00:17:14,388 --> 00:17:16,680 And it's told me what to do, gave the expl. 376 00:17:16,860 --> 00:17:19,904 And one of the people in the room was like, damn. 377 00:17:19,952 --> 00:17:24,400 Like that's my job is to prepare reports for people and that 378 00:17:24,440 --> 00:17:26,288 person will figure out something else to do with their life. 379 00:17:26,344 --> 00:17:28,992 I understand that there's that kind of job displacement. 380 00:17:29,056 --> 00:17:31,456 But what's more interesting to me is now Everybody has a McKinsey 381 00:17:31,488 --> 00:17:32,048 consultant. 382 00:17:32,144 --> 00:17:36,336 Everybody has like a logistics consultant in their pocket that can, 383 00:17:36,408 --> 00:17:39,952 you know, pay attention to all things that you can like even for 384 00:17:39,976 --> 00:17:40,512 marketers, right? 385 00:17:40,536 --> 00:17:43,536 It's like, you know, the Federal Reserve kept interest, raised 386 00:17:43,568 --> 00:17:44,096 interest rates. 387 00:17:44,128 --> 00:17:46,496 Here's what that means for your pay per click campaigns. 388 00:17:46,608 --> 00:17:50,146 Like all of this stuff that we have to couldn't possibly synthesize 389 00:17:50,178 --> 00:17:52,690 on our own unless you had a big team or big budgets or access 390 00:17:52,730 --> 00:17:53,634 to consultants. 391 00:17:53,762 --> 00:17:57,602 Now everybody has that from a leadership perspective. 392 00:17:57,666 --> 00:18:00,990 So we train first time leaders, emerging leaders, new leaders 393 00:18:01,450 --> 00:18:04,530 and increasingly just even just any existing leader. 394 00:18:04,570 --> 00:18:08,658 Because we found that a lot of leaders don't have any understanding 395 00:18:08,674 --> 00:18:11,858 about what they really should be doing broadly, which is despite 396 00:18:11,914 --> 00:18:15,730 their seniority, some of their foundations can be somewhat shaky. 397 00:18:15,890 --> 00:18:20,162 From that perspective, I see as we look forward that leaders are 398 00:18:20,186 --> 00:18:22,994 not just going to be leading people directly now, going to be 399 00:18:23,002 --> 00:18:23,986 leading technology. 400 00:18:24,138 --> 00:18:28,098 If you work for Amazon in a distribution sensor, you've already 401 00:18:28,154 --> 00:18:32,178 probably got a group of robots which are currently in your workforce. 402 00:18:32,274 --> 00:18:35,010 I would imagine that's going to be more prolific in terms of, 403 00:18:35,050 --> 00:18:38,386 as we move forward, especially when it comes to construction safety 404 00:18:38,418 --> 00:18:39,074 and things like that. 405 00:18:39,082 --> 00:18:43,178 And also labor here and domestically is extremely Expensive. 406 00:18:43,354 --> 00:18:44,870 And so if you can. 407 00:18:45,250 --> 00:18:46,938 And also we can't find any more people. 408 00:18:46,994 --> 00:18:47,658 That's the other part. 409 00:18:47,714 --> 00:18:48,602 It's not just expensive. 410 00:18:48,666 --> 00:18:49,738 It's like there's no. 411 00:18:49,874 --> 00:18:51,770 There's this huge surplus of. 412 00:18:51,890 --> 00:18:56,346 Of demand and insufficient people to do a bunch of the work. 413 00:18:56,418 --> 00:18:59,802 So if you can displace some of that or replace it with a robot to 414 00:18:59,826 --> 00:19:02,314 do, you know, moving the cement bags just as a real kind of 415 00:19:02,322 --> 00:19:05,450 trivial example, then all of a sudden you free someone up to do 416 00:19:05,490 --> 00:19:06,682 potentially something else. 417 00:19:06,746 --> 00:19:09,278 So what do you see? 418 00:19:09,414 --> 00:19:13,406 What's a leader going to need to be prepared for when it comes 419 00:19:13,478 --> 00:19:17,854 to the workforce of a decade, or even, let's say, two decades ahead? 420 00:19:17,942 --> 00:19:20,414 Maybe it's not even that far, maybe it's next week. 421 00:19:20,502 --> 00:19:25,726 But what's a leader looking like being prepared for? 422 00:19:25,798 --> 00:19:28,526 What kind of languages do they need to understand and be fluent 423 00:19:28,558 --> 00:19:31,090 in outside of, let's say, just English? 424 00:19:32,150 --> 00:19:35,202 The good news is you won't have to learn any new languages and 425 00:19:35,226 --> 00:19:38,230 you'll actually be able to use languages you'll never have to learn. 426 00:19:38,650 --> 00:19:42,706 So the, the T in GPT that it's called a transformer was originally 427 00:19:42,738 --> 00:19:44,710 invented for translation purposes. 428 00:19:45,370 --> 00:19:49,506 And the example I like to use is one time I was working on an experiment 429 00:19:49,538 --> 00:19:51,842 with some German scientists and I was trying to be cool, this 430 00:19:51,866 --> 00:19:55,538 is 20 years ago, and I tried to translate it into German for them, 431 00:19:55,594 --> 00:19:58,114 but I used, instead of the word for computer server, I use the 432 00:19:58,122 --> 00:19:59,270 word for like a waiter. 433 00:20:00,010 --> 00:20:01,852 And context matters, right? 434 00:20:01,876 --> 00:20:02,668 Like, they did not. 435 00:20:02,724 --> 00:20:05,160 That did not boost my credibility with that team. 436 00:20:05,700 --> 00:20:08,124 So, like, they really cracked that problem with the context. 437 00:20:08,252 --> 00:20:10,572 And, you know, you'll be able to talk, whether it's communication 438 00:20:10,636 --> 00:20:14,120 styles or languages, you'll be able to communicate with people and 439 00:20:14,420 --> 00:20:16,060 that you wouldn't have been able to do before. 440 00:20:16,100 --> 00:20:18,460 Like, translation is just going to be absolutely insane. 441 00:20:18,540 --> 00:20:19,596 Including context. 442 00:20:19,708 --> 00:20:22,508 Like, the engineering team will say, oh, like what artifacts 443 00:20:22,524 --> 00:20:23,132 do we need for that? 444 00:20:23,156 --> 00:20:24,860 This is something that actually happened to me today. 445 00:20:24,980 --> 00:20:28,380 And I used Zoom's AI companion to say, like, what do they mean by 446 00:20:28,420 --> 00:20:28,988 artifact? 447 00:20:29,084 --> 00:20:31,468 And it gave me an explanation in context that they were talking 448 00:20:31,484 --> 00:20:33,700 about something that people use on the website. 449 00:20:34,240 --> 00:20:37,340 So you're not going to have to learn any new languages. 450 00:20:37,680 --> 00:20:42,328 You are going to have to learn new concepts in terms of, in particular, 451 00:20:42,384 --> 00:20:45,540 in terms of shifting your ideas about what's possible. 452 00:20:46,080 --> 00:20:50,856 So you can go, you can go launch a startup now with, you know, 453 00:20:51,008 --> 00:20:54,328 three people, can go launch a newborn startup because you can have 454 00:20:54,384 --> 00:20:57,300 an AI marketing person, an AI finance person. 455 00:20:58,270 --> 00:21:04,454 You can also just a lot of people are using it in the obvious 456 00:21:04,582 --> 00:21:06,970 way, which is like, how do I spam more on LinkedIn? 457 00:21:07,310 --> 00:21:10,330 Or how do I replace live chat reps, for example? 458 00:21:10,670 --> 00:21:13,110 I think that's very short sighted. 459 00:21:13,190 --> 00:21:16,246 I think that instead of, you know, one live chat rep, instead 460 00:21:16,278 --> 00:21:18,454 of doing three chats at a time, because the human brain can 461 00:21:18,462 --> 00:21:21,062 only do so much, maybe now they can do 30 because it helps them 462 00:21:21,086 --> 00:21:23,462 with the context and everything else like that, you could 463 00:21:23,486 --> 00:21:26,022 say, cool, I only need 1 10, the number of live chat reps, which 464 00:21:26,046 --> 00:21:27,858 is what most companies are saying right now. 465 00:21:28,014 --> 00:21:31,802 Or you can say, great, my live chat reps can now spend more time 466 00:21:31,986 --> 00:21:34,778 digging into the problems and challenges of each person that they're 467 00:21:34,794 --> 00:21:37,498 talking to and create a competitive advantage for myself. 468 00:21:37,594 --> 00:21:39,418 So you can definitely use it to lower costs. 469 00:21:39,514 --> 00:21:42,602 But I think what's going to be most interesting is, you know, use 470 00:21:42,626 --> 00:21:43,018 construction. 471 00:21:43,034 --> 00:21:45,750 For example, can I have it? 472 00:21:46,130 --> 00:21:49,642 Ask the person a bunch of clarifying questions in advance so 473 00:21:49,666 --> 00:21:50,790 that I don't have to. 474 00:21:51,170 --> 00:21:55,402 Can I have it, you know, translate the jargon that I'm using 475 00:21:55,506 --> 00:21:57,750 into jargon that the other people understand. 476 00:21:58,210 --> 00:22:00,670 So you're not going to have to learn any new languages. 477 00:22:01,250 --> 00:22:05,562 The mindset is going to change and then in particular the way that 478 00:22:05,586 --> 00:22:06,938 you think about your career. 479 00:22:07,114 --> 00:22:08,890 So we were talking before this began. 480 00:22:08,970 --> 00:22:13,210 Like I kind of started my professional career early 2000s, 481 00:22:13,290 --> 00:22:15,510 you know, the Internet was kind of like becoming a thing. 482 00:22:15,890 --> 00:22:19,402 And because of that I got a huge jumpstart in my career, right? 483 00:22:19,426 --> 00:22:21,482 I got to run HubSpot labs. 484 00:22:21,626 --> 00:22:26,246 I had a CMO job when I was, you know, barely, you know, done 485 00:22:26,318 --> 00:22:29,046 not graduating from college, actually got to teach at the college 486 00:22:29,078 --> 00:22:31,174 I dropped out of because they didn't have a program covering the 487 00:22:31,182 --> 00:22:32,130 subject matter. 488 00:22:32,430 --> 00:22:35,014 And for the, for the younger leaders, this is an incredible. 489 00:22:35,062 --> 00:22:38,038 Again, my 16 year old nephew, this is an incredible opportunity 490 00:22:38,214 --> 00:22:42,422 where the only thing separating you from the people who 491 00:22:42,446 --> 00:22:45,558 currently hold the roles that you want to hold is that you don't 492 00:22:45,574 --> 00:22:48,598 have all the baggage you don't have, you don't know the right way. 493 00:22:48,654 --> 00:22:51,238 And so because of that you're, it's much easier for you to figure 494 00:22:51,254 --> 00:22:52,450 out the best way. 495 00:22:52,910 --> 00:22:57,594 All the rest of us, frankly, like I have to, I can't even switch 496 00:22:57,642 --> 00:22:59,002 to Superhuman for email, right? 497 00:22:59,026 --> 00:23:01,834 I'm still using Gmail even though I'm trying to like use some 498 00:23:01,842 --> 00:23:04,218 of these AI tools just because, you know, three decades 499 00:23:04,234 --> 00:23:07,434 of habits, professional habits have trained me. 500 00:23:07,522 --> 00:23:11,002 So it's a huge opportunity for anybody who's willing to incur the 501 00:23:11,026 --> 00:23:13,310 work to learn new skills and new mindsets. 502 00:23:13,730 --> 00:23:14,298 Yeah. 503 00:23:14,394 --> 00:23:15,130 Extraordinary. 504 00:23:15,210 --> 00:23:17,738 You used some interesting examples before we jumped on that. 505 00:23:17,794 --> 00:23:22,074 You've got AI summarizing your diary, using it, doing detailed research 506 00:23:22,162 --> 00:23:25,836 to, to kind of tell you about who you're meeting with, what they've 507 00:23:25,868 --> 00:23:28,700 said recently, what you might want to, you know, interesting things 508 00:23:28,740 --> 00:23:29,580 to ask them. 509 00:23:29,700 --> 00:23:32,732 Like, you know, it's kind of worrying when you, when you say that 510 00:23:32,756 --> 00:23:33,116 to someone. 511 00:23:33,188 --> 00:23:33,932 Oh, wow, okay. 512 00:23:33,996 --> 00:23:36,268 That's like way more research than I was. 513 00:23:36,404 --> 00:23:38,780 You know, I looked at your LinkedIn before we jumped on. 514 00:23:38,820 --> 00:23:39,564 It's like, oh, wow. 515 00:23:39,612 --> 00:23:43,132 Like it's, it seems so pathetic in many ways. 516 00:23:43,196 --> 00:23:43,800 Right. 517 00:23:44,100 --> 00:23:46,920 And that your, your approach is. 518 00:23:47,380 --> 00:23:50,524 Did you say it summarizes it into audio and tells you that some 519 00:23:50,532 --> 00:23:51,100 of the key things. 520 00:23:51,140 --> 00:23:52,428 Just tell us about, like some of your. 521 00:23:52,484 --> 00:23:55,042 Because you're the kind of the cutting edge in many anyways. 522 00:23:55,106 --> 00:23:55,762 Right? 523 00:23:55,946 --> 00:23:56,306 Yeah. 524 00:23:56,338 --> 00:23:59,378 And by the way, to clarify for everybody listening, I am not technical. 525 00:23:59,554 --> 00:24:02,098 I cannot write code that is usable by anybody. 526 00:24:02,194 --> 00:24:06,386 I can think logically and methodically and I can herd a bunch 527 00:24:06,418 --> 00:24:10,066 of tech cats, which is my actual job, is keeping everybody 528 00:24:10,098 --> 00:24:10,790 focused. 529 00:24:11,370 --> 00:24:13,858 But one of the things that, you know, I hate showing up to meetings 530 00:24:13,874 --> 00:24:14,738 unprepared. 531 00:24:14,914 --> 00:24:17,538 And so I have it look at who I'm meeting that day. 532 00:24:17,594 --> 00:24:21,702 My Google calendar, it uses, I think it's Clearbit to give me all 533 00:24:21,726 --> 00:24:22,870 the information about your company. 534 00:24:22,990 --> 00:24:25,670 Then it goes to LinkedIn and looks up you and your company's last, 535 00:24:25,710 --> 00:24:27,798 I think, hundred LinkedIn posts. 536 00:24:27,894 --> 00:24:31,078 And then I have it synthesize based on what it knows about me, 537 00:24:31,134 --> 00:24:33,254 what are interesting things that we might be able to talk about. 538 00:24:33,342 --> 00:24:37,702 So, for example, with you teed in on the young leaders thing because 539 00:24:37,726 --> 00:24:40,454 it's something I'm very passionate about and also just part 540 00:24:40,462 --> 00:24:41,526 of my origin story. 541 00:24:41,678 --> 00:24:43,942 And then I have a generate audio because I go for a walk every 542 00:24:43,966 --> 00:24:47,414 morning and I could, yeah, sure, I could listen to the Economist 543 00:24:47,462 --> 00:24:47,942 podcast. 544 00:24:48,006 --> 00:24:50,582 I could listen to, you know, the various podcasts that I used 545 00:24:50,606 --> 00:24:53,886 to listen to, or I can have it just also listen to those podcasts 546 00:24:53,918 --> 00:24:56,926 for me and only synthesize the things that it thinks I'll find interesting 547 00:24:57,038 --> 00:24:58,510 and just be really efficient in my day. 548 00:24:58,550 --> 00:24:59,070 So I get a. 549 00:24:59,110 --> 00:25:02,046 I get a President's daily briefing on everybody I'm meeting. 550 00:25:02,118 --> 00:25:04,782 And if there's any interesting news about your industry or your 551 00:25:04,806 --> 00:25:07,630 company without me having to go do that. 552 00:25:07,750 --> 00:25:11,822 That light amount of research before every meeting is a lot of 553 00:25:11,846 --> 00:25:12,686 time on our part. 554 00:25:12,758 --> 00:25:15,166 And it's still, like I said, it's very, very shallow. 555 00:25:15,278 --> 00:25:19,322 So now I can have these deep, you Know, briefings in context at 556 00:25:19,346 --> 00:25:20,910 scale, which is nice. 557 00:25:21,250 --> 00:25:21,946 Phenomenal. 558 00:25:21,978 --> 00:25:24,202 And I think the context piece is the most interesting part for 559 00:25:24,226 --> 00:25:24,410 me. 560 00:25:24,450 --> 00:25:24,666 Sam. 561 00:25:24,698 --> 00:25:24,890 Right. 562 00:25:24,930 --> 00:25:27,402 Because you can put. 563 00:25:27,426 --> 00:25:29,242 There's a podcast which you might want to listen to, and then 564 00:25:29,266 --> 00:25:31,802 you have to filter through and then you go, okay. 565 00:25:31,946 --> 00:25:35,242 Like a friend of mine at McKinsey, funnily enough, sent me 566 00:25:35,266 --> 00:25:38,250 a podcast yesterday about developing leaders and their take 567 00:25:38,290 --> 00:25:38,618 on it. 568 00:25:38,674 --> 00:25:41,482 And I was like, you know, I think it's a 20 minute episode. 569 00:25:41,546 --> 00:25:43,642 I was like, oh, I kind of got a little bit stuck after the first 570 00:25:43,666 --> 00:25:47,702 few minutes because it wasn't directly relevant to me immediately. 571 00:25:47,766 --> 00:25:50,934 But it's not to say the rest of it wasn't and there wouldn't be 572 00:25:50,942 --> 00:25:51,622 some gold in there. 573 00:25:51,646 --> 00:25:55,142 But I think if I could have something listen to it and then tell 574 00:25:55,166 --> 00:25:56,518 me just like an assistant in many ways. 575 00:25:56,534 --> 00:25:56,854 Right. 576 00:25:56,942 --> 00:25:58,742 I guess is ultimately what we're talking about. 577 00:25:58,846 --> 00:25:59,974 Listen to it contextually. 578 00:26:00,022 --> 00:26:02,854 I well, given what you're doing right now, here's how it might 579 00:26:02,862 --> 00:26:04,742 be relevant and here's what they talked about and here's what 580 00:26:04,766 --> 00:26:05,190 you hear. 581 00:26:05,230 --> 00:26:08,086 Some, you know, five quotes or something like that you could pump 582 00:26:08,118 --> 00:26:08,550 out. 583 00:26:08,670 --> 00:26:12,022 To have that capability at your fingertips for free is just 584 00:26:12,126 --> 00:26:14,468 absolutely bonkers that. 585 00:26:14,604 --> 00:26:17,076 It'S a literal agent, a literal assistant. 586 00:26:17,108 --> 00:26:18,680 Everybody has their own ea. 587 00:26:19,100 --> 00:26:19,652 Everybody. 588 00:26:19,716 --> 00:26:22,708 I mean, I had IT draft a contract for me for the bank earlier 589 00:26:22,764 --> 00:26:27,172 this week and I still had the actual lawyer legal nerds bless it 590 00:26:27,196 --> 00:26:28,788 because that's more of a liability issue. 591 00:26:28,844 --> 00:26:31,236 Like if it, if there's something in there is wrong, I want 592 00:26:31,308 --> 00:26:36,772 to be able to sue somebody who's not a non human AI, but they 593 00:26:36,796 --> 00:26:37,332 just set it back. 594 00:26:37,356 --> 00:26:38,612 They're like, yeah, this is fine. 595 00:26:38,756 --> 00:26:39,156 Right? 596 00:26:39,228 --> 00:26:43,040 And so it's, it's, it's, it's giving you those capabilities. 597 00:26:43,120 --> 00:26:43,504 It. 598 00:26:43,592 --> 00:26:46,192 And like you said, most business books are the same too, 599 00:26:46,296 --> 00:26:46,512 right? 600 00:26:46,536 --> 00:26:49,472 Like how much of the content we spend all of our time consuming. 601 00:26:49,616 --> 00:26:52,784 Maybe 20% of the average business book is actually interesting 602 00:26:52,832 --> 00:26:53,984 and or new information. 603 00:26:54,072 --> 00:26:58,224 For me, one of the things that I'm really interested in is this 604 00:26:58,232 --> 00:27:00,912 concept called information gain, where I want it to only give 605 00:27:00,936 --> 00:27:02,544 me things I don't already know. 606 00:27:02,712 --> 00:27:04,256 I hate that, especially with the news. 607 00:27:04,328 --> 00:27:04,656 Right. 608 00:27:04,728 --> 00:27:06,800 It's just the same news over and over again. 609 00:27:06,840 --> 00:27:09,024 But I have to listen to all these different podcasts just in 610 00:27:09,032 --> 00:27:10,460 case there's something I don't know. 611 00:27:11,880 --> 00:27:14,688 We're quickly getting to where it'll listen to everything for you 612 00:27:14,824 --> 00:27:19,072 and turn 400 page business book into the 30 pages you'll find 613 00:27:19,096 --> 00:27:19,648 interesting. 614 00:27:19,744 --> 00:27:23,072 Turn a 90 minute podcast into the 15 minutes you'll find interesting. 615 00:27:23,216 --> 00:27:25,952 And then you can either like, I don't know, go play chess with 616 00:27:25,976 --> 00:27:28,816 the rest, you know, your free time, or you can just do more. 617 00:27:28,888 --> 00:27:31,632 Like, I now listen to more podcasts than I ever listened to 618 00:27:31,656 --> 00:27:35,824 before because I don't actually have to listen to the entire 619 00:27:35,872 --> 00:27:36,304 podcast. 620 00:27:36,352 --> 00:27:38,460 I can just get the bits that are interesting to me. 621 00:27:39,020 --> 00:27:40,980 Wow, that is extraordinary. 622 00:27:41,060 --> 00:27:44,692 I think one thing we talk about is the power of having mentors. 623 00:27:44,836 --> 00:27:46,276 And, you know, and. 624 00:27:46,348 --> 00:27:49,060 And often the junior people get stuck in the. 625 00:27:49,100 --> 00:27:53,732 With a middle mentor that being like a very mediocre typically 626 00:27:53,796 --> 00:27:58,276 mentor at this level where, you know, the people they're surrounded 627 00:27:58,308 --> 00:28:01,412 by are people in the middle of their career who haven't necessarily 628 00:28:01,476 --> 00:28:02,532 reached the higher realms. 629 00:28:02,596 --> 00:28:05,890 Not to say that the higher realms are superior in any way, but 630 00:28:05,930 --> 00:28:08,898 they have a very different perspective often about the right 631 00:28:08,954 --> 00:28:12,270 advice that you might get from a middle manager versus from an executive. 632 00:28:12,890 --> 00:28:16,350 And yeah, I'm just thinking, wow, this is a real. 633 00:28:16,650 --> 00:28:18,834 You don't even need the right network anymore. 634 00:28:18,882 --> 00:28:20,354 In some ways, you could get. 635 00:28:20,442 --> 00:28:22,194 The network is an AI. 636 00:28:22,322 --> 00:28:24,818 You could have an advisory board every morning that you could 637 00:28:24,954 --> 00:28:27,778 interact with about, okay, here's the challenges I'm dealing 638 00:28:27,794 --> 00:28:27,922 with. 639 00:28:27,946 --> 00:28:29,150 What do you think we should do? 640 00:28:29,770 --> 00:28:32,098 What would be your different perspectives from around the table? 641 00:28:32,194 --> 00:28:37,920 Is it that advanced yet in terms of could you have a CEO almost 642 00:28:37,960 --> 00:28:41,420 have like a conversation or an advisory board meeting? 643 00:28:41,800 --> 00:28:44,656 It is very much, especially for the more senior folks. 644 00:28:44,688 --> 00:28:47,860 So, like Dharmesh, who founded HubSpot, also founded Agent AI. 645 00:28:48,280 --> 00:28:52,912 He fed it all of his keynote speeches, all the articles he's written, 646 00:28:52,976 --> 00:28:55,344 everything else like that, and created kind of a digital clone of 647 00:28:55,352 --> 00:28:56,096 himself. 648 00:28:56,288 --> 00:28:59,136 So I actually do this on occasion. 649 00:28:59,168 --> 00:29:01,980 I'm like, do I need to ask his permission for something? 650 00:29:02,450 --> 00:29:05,322 And it'll be like, it'll give me some advice on that, right? 651 00:29:05,346 --> 00:29:06,586 I'll be like, yes, you should ask for permission. 652 00:29:06,618 --> 00:29:07,370 Or like, no. 653 00:29:07,410 --> 00:29:07,786 Like he's. 654 00:29:07,818 --> 00:29:10,474 He wants you to move fast and not bother with, you know, permission 655 00:29:10,522 --> 00:29:11,270 on things. 656 00:29:11,810 --> 00:29:15,786 So being able to have kind of this counsel where you can get advice. 657 00:29:15,898 --> 00:29:18,222 And there are some companies that, that just do this, by the way, 658 00:29:18,236 --> 00:29:22,762 that are much, much better at it than just a generic AI agent where 659 00:29:22,786 --> 00:29:25,594 you can feed it all of this stuff that I've written, all the 660 00:29:25,602 --> 00:29:27,418 lectures that I've given, etc. 661 00:29:27,514 --> 00:29:29,834 And then if you're a startup and you're telling me you've got 662 00:29:29,842 --> 00:29:33,002 a retention problem, I'm pretty sure you actually have a pricing 663 00:29:33,066 --> 00:29:34,490 problem, not a retention problem. 664 00:29:34,530 --> 00:29:37,626 If you're like a software startup and. 665 00:29:37,698 --> 00:29:40,730 But I have to give that advice over and over and over again and 666 00:29:40,770 --> 00:29:45,230 because of that I only mentor a couple dozen startups and entrepreneurs. 667 00:29:45,650 --> 00:29:48,682 If you clone me, I can mentor an infinite number of startups and 668 00:29:48,706 --> 00:29:52,442 entrepreneurs and also elevate those middle managers faster. 669 00:29:52,586 --> 00:29:53,270 Right. 670 00:29:53,890 --> 00:29:55,850 Tell it to give you a different perspective. 671 00:29:55,930 --> 00:29:59,370 I always tell it to argue against me writing the contract. 672 00:29:59,410 --> 00:30:01,946 For example, I'm like now pretend you're the bank and you're 673 00:30:01,978 --> 00:30:05,062 worried about me doing, doing something bad or weird like how would 674 00:30:05,086 --> 00:30:06,214 you change the agreement? 675 00:30:06,342 --> 00:30:09,398 And it came up with a bunch of stuff that I didn't, hadn't even 676 00:30:09,454 --> 00:30:11,158 thought would be a concern for them. 677 00:30:11,294 --> 00:30:13,142 So having it be multi perspective. 678 00:30:13,286 --> 00:30:13,734 Right. 679 00:30:13,822 --> 00:30:15,810 We always joke about explain it like I'm five. 680 00:30:16,190 --> 00:30:18,214 There's also explain it like I'm the VP of sales. 681 00:30:18,262 --> 00:30:19,590 Explain it like I'm the cfo. 682 00:30:19,670 --> 00:30:22,902 Like the same piece of content should customize itself to the, to 683 00:30:22,926 --> 00:30:25,990 the person consuming it, but then also the same piece of advice. 684 00:30:26,150 --> 00:30:26,518 Right. 685 00:30:26,574 --> 00:30:31,316 I, I want to get advice from somebody who's Gen Z and actually 686 00:30:31,388 --> 00:30:36,116 understands how, I don't know, TikTok works or whatever. 687 00:30:36,148 --> 00:30:36,292 Right. 688 00:30:36,316 --> 00:30:39,400 Like I used to be cool and then time happened. 689 00:30:40,220 --> 00:30:43,556 But you know, like can I, can I get like advice from very senior 690 00:30:43,588 --> 00:30:46,852 people, very educated and can I ask it to have a perspective where 691 00:30:46,876 --> 00:30:49,412 it's like, you know, what would somebody from Gen Z say about 692 00:30:49,436 --> 00:30:49,572 this? 693 00:30:49,596 --> 00:30:51,840 Or are there other networks that I should look into? 694 00:30:52,620 --> 00:30:55,396 Especially with some of the new stuff like the deep research 695 00:30:55,468 --> 00:30:56,120 models. 696 00:30:56,420 --> 00:30:58,412 It's just, it's just fascinating. 697 00:30:58,476 --> 00:31:00,844 I have it writing research papers for me on everything you can 698 00:31:00,852 --> 00:31:04,076 imagine and I learned so much that I hadn't thought to ask. 699 00:31:04,228 --> 00:31:06,636 And those are always the most interesting questions and the hardest 700 00:31:06,668 --> 00:31:09,820 for mentors and why more experienced mentors are generally, 701 00:31:09,980 --> 00:31:12,732 I don't want to say better, but they just have more experience 702 00:31:12,836 --> 00:31:14,092 to do pattern recognition with. 703 00:31:14,116 --> 00:31:15,640 Like they've seen this before. 704 00:31:16,740 --> 00:31:22,780 Being able to have it go know what questions you haven't thought 705 00:31:22,820 --> 00:31:23,228 to ask. 706 00:31:23,284 --> 00:31:24,556 I think that's, that's really interesting. 707 00:31:24,588 --> 00:31:27,484 That's gonna be a big advantage for people who need mentors 708 00:31:27,612 --> 00:31:29,920 and not everybody can be mentored by the CEO. 709 00:31:30,420 --> 00:31:32,700 Yeah, yeah, it's a great point. 710 00:31:32,740 --> 00:31:33,228 I love that. 711 00:31:33,284 --> 00:31:34,044 Really, really powerful. 712 00:31:34,092 --> 00:31:36,204 I think if you can tap into it, maybe. 713 00:31:36,252 --> 00:31:38,044 Let's finish with the last bit of advice. 714 00:31:38,092 --> 00:31:38,556 Sam. 715 00:31:38,668 --> 00:31:45,388 Let's say you've got a, let's say, let's say an engineer 2027. 716 00:31:45,524 --> 00:31:49,030 They're early in their career interested about developing the right 717 00:31:49,070 --> 00:31:51,462 skills to put them ahead in the future. 718 00:31:51,566 --> 00:31:55,622 What would be some of your guidance to them, let's say in construction 719 00:31:55,686 --> 00:31:58,390 infrastructure, if they're interested in being a senior leader 720 00:31:58,430 --> 00:32:03,846 down the track and being able to take advantage of the opportunities 721 00:32:03,878 --> 00:32:06,246 that we've got here at our fingertips. 722 00:32:06,438 --> 00:32:06,806 Yeah. 723 00:32:06,838 --> 00:32:09,238 So we're still obviously in the very early days of AI. 724 00:32:09,334 --> 00:32:12,214 It feels like it's moving very fast and it is moving much faster 725 00:32:12,262 --> 00:32:13,830 than any other technology change. 726 00:32:13,950 --> 00:32:15,206 The, the most important thing. 727 00:32:15,278 --> 00:32:17,478 And I try to force myself to do this as well. 728 00:32:17,534 --> 00:32:18,982 Just make it a part of your day. 729 00:32:19,086 --> 00:32:20,502 Like, I don't use love. 730 00:32:20,526 --> 00:32:21,958 My friends at Google don't use Google anymore. 731 00:32:21,974 --> 00:32:25,222 I use Perplexity because I can ask it questions, it'll cite its 732 00:32:25,246 --> 00:32:26,490 answers, et cetera. 733 00:32:28,350 --> 00:32:31,702 I the first thing you should do in addition to just like doing 734 00:32:31,726 --> 00:32:34,214 something dumb like I built a chess coach agent right where it 735 00:32:34,222 --> 00:32:37,926 just looks at my games and looks at my, the YouTube playlist 736 00:32:37,958 --> 00:32:40,358 of my actual chess coach and says like, what's he going to yell 737 00:32:40,374 --> 00:32:40,982 at me about? 738 00:32:41,086 --> 00:32:41,798 That's fine. 739 00:32:41,894 --> 00:32:44,790 But also just ask it to explain to you the lives of the people 740 00:32:44,830 --> 00:32:45,496 around you. 741 00:32:45,638 --> 00:32:47,812 So, and you don't have to build your own AI agent for this. 742 00:32:47,916 --> 00:32:50,900 You can use ChatGPT, perplexity, anything you want. 743 00:32:51,020 --> 00:32:52,996 Just say like, here are the people you work with and interact 744 00:32:53,028 --> 00:32:56,052 with on a day to day basis and like explain their background. 745 00:32:56,116 --> 00:32:58,676 Explain like, what is it like to be a project manager? 746 00:32:58,708 --> 00:33:00,580 What is it like to be a comptroller? 747 00:33:00,660 --> 00:33:05,780 What is it like to be a risk and compliance analyst and start 748 00:33:05,820 --> 00:33:10,148 to understand that AI can give you different perspectives and then 749 00:33:10,204 --> 00:33:12,338 start to think about how it can give you different skills. 750 00:33:12,524 --> 00:33:15,318 There's a great site I love called there's an AI for that.com 751 00:33:15,454 --> 00:33:18,902 and if you go there you can just click into, you know, whatever 752 00:33:18,966 --> 00:33:21,830 job or task or industry you have and there's an AI for that, 753 00:33:21,870 --> 00:33:26,534 it turns out, starting to then say, what skills can I give myself 754 00:33:26,582 --> 00:33:27,366 that I don't have? 755 00:33:27,438 --> 00:33:29,382 So I am not a good designer. 756 00:33:29,446 --> 00:33:31,770 The last design tool I used was Microsoft Paint. 757 00:33:32,270 --> 00:33:34,730 And so I, which doesn't even exist anymore. 758 00:33:35,310 --> 00:33:37,490 So I use it to like help me design. 759 00:33:37,870 --> 00:33:39,102 I am not. 760 00:33:39,246 --> 00:33:41,930 I can write a thousand words, I can't write a hundred. 761 00:33:42,310 --> 00:33:45,902 So I use it to edit things so that my essays, you know, to the 762 00:33:45,926 --> 00:33:48,610 team are actually just, you know, 50 word emails. 763 00:33:49,510 --> 00:33:52,542 Figure out what it is that you wish you were good at and then go 764 00:33:52,566 --> 00:33:56,462 find an AI that can help you with that, whether it's data analysis 765 00:33:56,606 --> 00:33:59,742 or anything else like that, and then use it to Learn like, I 766 00:33:59,766 --> 00:34:04,110 have it just like the weirdest stuff, I have it going down. 767 00:34:04,150 --> 00:34:06,606 Like I think I said earlier, you know, learning about quantum 768 00:34:06,638 --> 00:34:10,020 mechanics, learning about, you know, the. 769 00:34:10,060 --> 00:34:14,020 The nuances of finance, like all these things that they're just 770 00:34:14,060 --> 00:34:17,780 helpful for you to have, like a broader context, but you would 771 00:34:17,820 --> 00:34:19,492 normally not have time to learn all of that. 772 00:34:19,516 --> 00:34:20,692 And you need like a teacher. 773 00:34:20,756 --> 00:34:22,356 And teachers aren't going to be replaced. 774 00:34:22,468 --> 00:34:25,604 To be clear, TAs might be replaced, teacher's assistants might 775 00:34:25,612 --> 00:34:26,212 be replaced. 776 00:34:26,276 --> 00:34:29,796 Because do we actually want human humans sitting there grading 777 00:34:29,828 --> 00:34:30,520 essays? 778 00:34:31,660 --> 00:34:35,412 But I think that's my number one piece of advice is one, use it 779 00:34:35,436 --> 00:34:37,894 to figure out how everything you hate in your day automate it. 780 00:34:37,972 --> 00:34:40,658 Like it's probably automatable unless it's physically picking stuff 781 00:34:40,674 --> 00:34:42,390 up where you actually do need robots. 782 00:34:42,730 --> 00:34:45,762 Two, have it help you understand everybody around you. 783 00:34:45,946 --> 00:34:49,790 And then three, have it help you learn and master skills that 784 00:34:50,250 --> 00:34:54,674 master skills without actually having to learn, you know, indesign 785 00:34:54,722 --> 00:34:57,522 in canva and design tools or data analytics. 786 00:34:57,586 --> 00:34:57,810 Right. 787 00:34:57,850 --> 00:34:59,202 Just be able to ask a question. 788 00:34:59,306 --> 00:35:02,690 The people who win the future are going to be the people who are 789 00:35:02,730 --> 00:35:05,830 really, really good at asking AI the right questions. 790 00:35:07,820 --> 00:35:08,340 Wow. 791 00:35:08,420 --> 00:35:09,092 Super powerful. 792 00:35:09,156 --> 00:35:11,268 So I'm scribbling like mad down here. 793 00:35:11,404 --> 00:35:15,040 So that was asking for a friend, but mainly for me. 794 00:35:15,340 --> 00:35:17,172 So, Sam, awesome work. 795 00:35:17,196 --> 00:35:18,660 Thank you so much for your time. 796 00:35:18,700 --> 00:35:20,196 Sam, I just want to maybe just. 797 00:35:20,268 --> 00:35:22,004 What's one question around security? 798 00:35:22,092 --> 00:35:23,348 Because I think it's. 799 00:35:23,524 --> 00:35:25,444 We're giving a lot of information to the machines. 800 00:35:25,492 --> 00:35:27,156 We're giving a lot of information to AI. 801 00:35:27,268 --> 00:35:30,360 You're basically, you know, quote unquote, replicating yourself. 802 00:35:32,210 --> 00:35:33,450 What are your thoughts around that? 803 00:35:33,490 --> 00:35:38,186 Like, you know, the kind of the AI, Sam, being better than you, 804 00:35:38,258 --> 00:35:39,990 knowing more even. 805 00:35:40,290 --> 00:35:44,586 I don't know, just especially in a free context, giving that amount 806 00:35:44,618 --> 00:35:46,950 of detailed information, quote, unquote, away. 807 00:35:47,650 --> 00:35:50,390 How do you think about security and safety around that? 808 00:35:52,370 --> 00:35:55,402 I have like, a specific punch list of ethical concerns that we're 809 00:35:55,466 --> 00:35:57,012 working on researching our way through. 810 00:35:57,146 --> 00:36:01,344 One of them is just the consent of data, right? 811 00:36:01,432 --> 00:36:04,288 So the big models, they vacuumed up the entire Internet, 812 00:36:04,384 --> 00:36:06,544 which is first of all unnecessary. 813 00:36:06,592 --> 00:36:09,232 Like, you don't need the published works of Tolkien and the 814 00:36:09,256 --> 00:36:11,232 Lord of the Rings to write good LinkedIn posts. 815 00:36:11,296 --> 00:36:12,500 It was unnecessary. 816 00:36:13,080 --> 00:36:16,336 But then two, I also just think, like, you know, if you're 817 00:36:16,368 --> 00:36:18,944 creating content and putting it out there on the Internet, you 818 00:36:18,952 --> 00:36:20,460 should get to decide how it's used. 819 00:36:21,080 --> 00:36:24,288 There are different language models that are being trained on 820 00:36:24,344 --> 00:36:27,530 things that, for example, only people have specifically opted in, 821 00:36:27,570 --> 00:36:29,626 or the data, the people who created the data have been. 822 00:36:29,698 --> 00:36:30,910 Have been compensated. 823 00:36:31,250 --> 00:36:36,394 So I think that's one bit is we're going to see a significant 824 00:36:36,482 --> 00:36:36,858 backlash. 825 00:36:36,874 --> 00:36:38,842 And it's already, at least in the US it's working its way through 826 00:36:38,866 --> 00:36:39,818 our court system. 827 00:36:39,954 --> 00:36:43,706 You know, if your AI is using, you know, somebody's book or movie 828 00:36:43,738 --> 00:36:45,690 or something like that, do you owe them something? 829 00:36:45,810 --> 00:36:47,950 I would argue yes, but not a lawyer. 830 00:36:48,770 --> 00:36:52,890 In terms of the data security, there's a huge move in all of computing 831 00:36:52,970 --> 00:36:55,070 to do what's called moving stuff to the edge. 832 00:36:55,410 --> 00:37:00,244 So your phone has enough processing power to do most of the 833 00:37:00,252 --> 00:37:03,320 stuff that AI does without the data ever leaving your phone. 834 00:37:04,620 --> 00:37:06,292 I think probably overkill, right? 835 00:37:06,316 --> 00:37:08,164 Like, data security is actually pretty good. 836 00:37:08,252 --> 00:37:10,388 We're pretty good at making sure credit card numbers and stuff 837 00:37:10,404 --> 00:37:10,740 like that. 838 00:37:10,780 --> 00:37:12,228 Like, yeah, there's leaks. 839 00:37:12,404 --> 00:37:16,580 But I think that in order for AI to gain adoption, the number one 840 00:37:16,620 --> 00:37:18,740 thing has to be trust, right? 841 00:37:18,780 --> 00:37:20,756 So if I'm going to ask it for advice, if I'm going to be vulnerable 842 00:37:20,788 --> 00:37:24,172 to the AI, I kind of want to know that it's not going to, you 843 00:37:24,196 --> 00:37:26,860 know, that my boss isn't going to be able to log in and see that. 844 00:37:26,900 --> 00:37:28,060 You know, I'm just not feeling it today. 845 00:37:28,100 --> 00:37:31,692 And I need some motivation, which I actually use AI for a lot, 846 00:37:31,796 --> 00:37:33,640 motivating me to get out of bed and go run. 847 00:37:34,580 --> 00:37:38,652 And I think while we can solve that as a technology problem, I just 848 00:37:38,676 --> 00:37:41,516 think that because it's such a human interaction and experience, 849 00:37:41,668 --> 00:37:44,892 like, if you were in the room with me, I would say anything to 850 00:37:44,916 --> 00:37:48,348 you, but we're talking through a computer, you know, kind of anybody. 851 00:37:48,364 --> 00:37:50,876 Well, we're recording, so by definition, everybody's listening. 852 00:37:50,988 --> 00:37:53,068 But even if we weren't, like kind of anybody could be listening 853 00:37:53,084 --> 00:37:53,260 in. 854 00:37:53,300 --> 00:37:56,964 So I think that that is going to be big over the next few years 855 00:37:57,132 --> 00:37:59,172 as all of these AI capabilities. 856 00:37:59,236 --> 00:38:00,772 And Apple is doing a very good job of this. 857 00:38:00,796 --> 00:38:03,812 Actually not a huge Apple fan, but in this I am. 858 00:38:03,916 --> 00:38:06,980 They're moving everything to the edge so that the AI models themselves 859 00:38:07,140 --> 00:38:08,388 never actually see the data. 860 00:38:08,444 --> 00:38:12,020 Can you have the benefits of AI without any of the data that you 861 00:38:12,060 --> 00:38:15,364 rightfully own ever leaving a device that only you control? 862 00:38:15,452 --> 00:38:18,052 I think that's going to be important and we're heading in that 863 00:38:18,076 --> 00:38:18,760 direction. 864 00:38:19,100 --> 00:38:19,798 Interesting. 865 00:38:19,884 --> 00:38:20,530 Very cool. 866 00:38:20,610 --> 00:38:21,650 Sam, we've got. 867 00:38:21,690 --> 00:38:22,418 We've come to time. 868 00:38:22,474 --> 00:38:25,970 Thank you so much for your input here. 869 00:38:26,090 --> 00:38:28,450 We're going to link to Agent AI. 870 00:38:28,610 --> 00:38:28,994 It's. 871 00:38:29,042 --> 00:38:31,410 If you're watching the video, it's on your T shirt, which is very 872 00:38:31,450 --> 00:38:32,150 helpful. 873 00:38:32,570 --> 00:38:34,386 And you've said it a few times here. 874 00:38:34,458 --> 00:38:37,186 So, yeah, huge thanks for your, for your help and your input 875 00:38:37,218 --> 00:38:37,602 on this one. 876 00:38:37,626 --> 00:38:40,370 I think it's going to be a fascinating topic, and I hope people 877 00:38:40,410 --> 00:38:44,194 listening to this are logging in as we speak and checking out what 878 00:38:44,202 --> 00:38:46,290 you've got and how it can utilize them and benefit them. 879 00:38:46,330 --> 00:38:48,390 So, yes, Sam, huge thanks for being here. 880 00:38:48,570 --> 00:38:49,374 Thanks for having me. 881 00:38:49,382 --> 00:38:51,662 And just everybody keep in mind that everybody else is just 882 00:38:51,686 --> 00:38:54,450 as confused as you are, but chaos is an opportunity. 883 00:38:55,910 --> 00:38:57,166 That's a great way to finish. 884 00:38:57,198 --> 00:38:57,590 I love it.