Description of the video:
WEBVTT 1 00:00:00.460 --> 00:00:03.400 Good afternoon and welcome. I'm Kim Goad, executive director 2 00:00:03.410 --> 00:00:08.000 at Kelley Executive Education. Today, we're excited to share an insightful 3 00:00:08.080 --> 00:00:11.660 conversation between Professor Carolyn Goerner and the recently appointed 4 00:00:11.800 --> 00:00:14.920 faculty chair of Kelley Direct Programs, Erik Gonzalez-Mulé. 5 00:00:15.700 --> 00:00:19.550 In this discussion, Erik and Carolyn explore how leaders can use AI to 6 00:00:19.600 --> 00:00:23.340 become more effective while still focusing on what matters most: 7 00:00:23.700 --> 00:00:24.080 people. 8 00:00:24.920 --> 00:00:28.640 From practical use cases to managing team adoption, this conversation 9 00:00:28.780 --> 00:00:31.990 highlights how AI isn't replacing leadership, 10 00:00:32.000 --> 00:00:34.120 it's reshaping how great leaders show up. 11 00:00:34.980 --> 00:00:38.020 If you're interested in exploring more, the Managing People course in our 12 00:00:38.030 --> 00:00:41.440 Leading with AI series will be delivered in person on July 9th 13 00:00:41.460 --> 00:00:45.740 in Indianapolis, followed by a live virtual session on July 16th. 14 00:00:45.750 --> 00:00:49.200 Watch for details at the end of today's discussion and in our follow-up email. 15 00:00:49.210 --> 00:00:51.880 Thank you for being here, and enjoy the conversation. 16 00:01:02.260 --> 00:01:04.630 [Carolyn Goerner] So Erik, let's start with a shameless sales pitch. 17 00:01:05.140 --> 00:01:09.920 Your class is Leading with AI: Leading People, Managing People. 18 00:01:09.940 --> 00:01:12.680 So why should someone attend the class? What's it all about? 19 00:01:13.620 --> 00:01:15.880 So I would say that the class is really about two major things. 20 00:01:16.480 --> 00:01:20.340 The first is how it is that leaders can best leverage AI to make themselves more 21 00:01:20.360 --> 00:01:24.060 effective at their jobs. So this might mean things like using AI to help you write 22 00:01:24.180 --> 00:01:27.000 interview questions when you're hiring a new person to your team. 23 00:01:27.040 --> 00:01:30.950 It could be things like helping AI write performance review letters that reflect 24 00:01:30.980 --> 00:01:34.080 the wonderful notes and all of the information that you've gathered over the course 25 00:01:34.100 --> 00:01:35.470 of a period of time for an employee. 26 00:01:36.280 --> 00:01:40.240 But it also includes things like how do you help your team best adopt 27 00:01:40.280 --> 00:01:40.940 the use of AI? 28 00:01:41.330 --> 00:01:41.330 [Carolyn] Oh. 29 00:01:41.380 --> 00:01:44.880 How do you coach your team along? How do you help develop your team 30 00:01:44.920 --> 00:01:48.500 in a way that has them open to AI as potentially 31 00:01:48.520 --> 00:01:52.400 pretending some opportunities for the work that they do, 32 00:01:52.440 --> 00:01:55.930 as opposed to seeing it as something that causes anxiety or as something that, 33 00:01:56.780 --> 00:02:00.420 the common concern that it's going to replace me. Right? 34 00:02:00.500 --> 00:02:04.220 And so a big part of a leader's job, I think, in today's world, 35 00:02:04.420 --> 00:02:07.360 is holding both of those things together and being able to help employees see the 36 00:02:07.420 --> 00:02:11.360 opportunities in this, as opposed to seeing it as something that's a barrier to 37 00:02:11.380 --> 00:02:12.180 them getting their work done. 38 00:02:12.540 --> 00:02:16.320 Well, not to put you on the spot, but let's say that I am your employee. 39 00:02:16.780 --> 00:02:20.720 Convince me that there's not only fear in AI. 40 00:02:20.840 --> 00:02:21.640 What would you say? 41 00:02:21.740 --> 00:02:26.470 [Erik] Yeah. So that's a great question. So in some of the approach that we've taken, 42 00:02:26.500 --> 00:02:29.660 and let me also back up and say here that research on this is in its 43 00:02:29.700 --> 00:02:32.880 infancy, right? So we have some studies, we have some evidence that we can draw 44 00:02:32.980 --> 00:02:35.720 from to help us say, "Okay, this is a good use of AI. 45 00:02:35.780 --> 00:02:38.300 Maybe this isn't a great use. Maybe we should think about this. 46 00:02:38.360 --> 00:02:39.420 Maybe we shouldn't do that." 47 00:02:40.360 --> 00:02:40.370 [Carolyn] Right. 48 00:02:40.380 --> 00:02:44.260 But still being relative in its infancy, we're really left to 49 00:02:44.320 --> 00:02:46.299 try things out, experiment, and learn. 50 00:02:46.740 --> 00:02:51.300 And I think that's just the state that we are all in right now with generative AI 51 00:02:51.340 --> 00:02:53.540 and the very fast adoption that it's had in the workplace. 52 00:02:53.580 --> 00:02:56.260 So that kind of caveat out of the way. 53 00:02:56.600 --> 00:03:00.520 The way we talk to our folks is think about ways that AI can help make you more 54 00:03:00.540 --> 00:03:04.300 effective and efficient. And we frame it from the point of view of 55 00:03:04.380 --> 00:03:08.049 AI is never going to replace your job, but a person who is really 56 00:03:08.080 --> 00:03:11.540 competent and knows how to really leverage AI well could replace you 57 00:03:11.600 --> 00:03:14.960 because that person will just, frankly, be better at their job. 58 00:03:15.020 --> 00:03:18.680 If they're able to offload some of the things that are rote, that are monotonous, 59 00:03:18.720 --> 00:03:22.480 that aren't the things that are a really a good use of a person's 60 00:03:22.540 --> 00:03:25.410 time in various work roles, 61 00:03:26.220 --> 00:03:28.300 that person could potentially replace you, right? 62 00:03:28.340 --> 00:03:29.200 And so, 63 00:03:29.260 --> 00:03:32.060 and we just are being very honest and transparent with people and saying, "This is just 64 00:03:32.100 --> 00:03:34.540 the world that we're in." It's the same thing 25 years ago. 65 00:03:34.560 --> 00:03:38.410 If you're hiring someone and a group of folks know how to use Microsoft Word and 66 00:03:38.440 --> 00:03:41.560 Excel and the whole suite of Office products, and another group doesn't, and you 67 00:03:41.580 --> 00:03:44.840 need that person to use those tools, you're probably going to hire that person. 68 00:03:44.880 --> 00:03:48.500 Right? And so it just comes down to the skills and competencies that are needed for 69 00:03:49.160 --> 00:03:51.100 2025 roles. 70 00:03:51.360 --> 00:03:51.960 [Carolyn] We love that. 71 00:03:52.000 --> 00:03:53.600 Maybe I should say 2026. 72 00:03:54.400 --> 00:03:58.140 Well, you know what? I think at this point, it's all changing so fast. 73 00:03:58.160 --> 00:03:58.450 [Erik] That's right. 74 00:03:58.500 --> 00:04:02.360 We could probably have another conversation in August of 2026, 75 00:04:02.480 --> 00:04:06.380 and it would be, again, what should we be calling it now? 76 00:04:06.400 --> 00:04:06.549 77 00:04:06.880 --> 00:04:07.140 [Erik] Yeah. 78 00:04:07.200 --> 00:04:11.000 So that, I think, is part of the challenge, that this is like a change 79 00:04:11.060 --> 00:04:12.340 initiative that never dies. 80 00:04:12.640 --> 00:04:12.950 81 00:04:13.380 --> 00:04:16.969 So what kinds of things could a manager be doing to prepare their 82 00:04:17.000 --> 00:04:18.810 folks for that kind of thinking? 83 00:04:19.190 --> 00:04:21.780 [Erik] Mm-hmm. Yeah. Yeah, that's a great question. 84 00:04:21.820 --> 00:04:25.700 And I think part of what I think generative AI is going to force 85 00:04:25.760 --> 00:04:29.100 managers to do, and should be forcing them to do, is to take a step back and 86 00:04:29.160 --> 00:04:33.100 evaluate themselves and what it is that they do 87 00:04:33.140 --> 00:04:38.000 in their role that is uniquely human, that generative AI cannot do for them, right? 88 00:04:38.100 --> 00:04:40.420 And that, to me, always comes down to being empathetic. 89 00:04:40.660 --> 00:04:40.760 [Carolyn] Yeah. 90 00:04:40.860 --> 00:04:44.640 It comes down to leading people through relationships with people. 91 00:04:44.720 --> 00:04:47.920 It comes down to doing the things that leaders have to do that involve 92 00:04:47.940 --> 00:04:50.720 critical thinking and synthesizing information and making really difficult 93 00:04:50.730 --> 00:04:54.270 decisions that can't be outsourced to AI. 94 00:04:54.420 --> 00:04:56.590 Definitely not now, potentially never. Right? 95 00:04:57.020 --> 00:04:58.800 So I would tell leaders to start there. 96 00:04:59.900 --> 00:05:02.980 And then once you've started there, it's okay, how are you talking to people around 97 00:05:03.020 --> 00:05:05.980 generative AI? Are you removing barriers to adoption? 98 00:05:06.040 --> 00:05:09.960 Are you supporting their work? It's one thing to have an AI 99 00:05:10.060 --> 00:05:13.750 strategy that essentially says, "Here's generative AI, go use it," 100 00:05:14.340 --> 00:05:17.580 as opposed to, "Here's generative AI, let's all get together and talk about it. 101 00:05:17.700 --> 00:05:20.680 Let's discuss appropriate use. Let's discuss inappropriate use. 102 00:05:21.100 --> 00:05:23.740 Let's discuss some use cases in our areas. 103 00:05:23.820 --> 00:05:27.800 Let's set some really proximal, measurable goals. 104 00:05:27.840 --> 00:05:30.920 Like within the next month, let's think about a way that we're going to take 105 00:05:30.980 --> 00:05:35.520 a process that used to take a lot of person-hours, but that did not add value, 106 00:05:35.560 --> 00:05:38.740 that was monotonous, repetitive, and let's see if AI can do that for us," right? 107 00:05:38.820 --> 00:05:40.700 And maybe free that person up to do something else, 108 00:05:40.720 --> 00:05:45.500 or free that person to be able to take a proper hour-long lunch break, right? 109 00:05:45.520 --> 00:05:46.219 Whatever that looks like. 110 00:05:46.240 --> 00:05:46.640 [Carolyn] Oh, my goodness. What's that? 111 00:05:47.520 --> 00:05:51.420 Honestly, as you said that, I kind of felt my shoulders relax because 112 00:05:51.480 --> 00:05:54.420 I think one of the things that people have with AI is that 113 00:05:54.500 --> 00:05:57.260 it's completely up to them individually 114 00:05:57.270 --> 00:05:59.920 to be spending hours every evening figuring it out. 115 00:05:59.940 --> 00:06:00.360 [Erik] Mm-hmm. 116 00:06:00.560 --> 00:06:03.050 And when you said, "Let's we do this. 117 00:06:03.100 --> 00:06:05.880 Let's figure out the parameters together," 118 00:06:05.890 --> 00:06:06.090 119 00:06:06.180 --> 00:06:08.220 I kind of took a deep breath and relaxed. 120 00:06:08.700 --> 00:06:11.969 So thank you in that regard. I think that's really important. 121 00:06:12.020 --> 00:06:14.260 [Erik] Well, and I think your take on that, actually, 122 00:06:14.280 --> 00:06:15.500 I think why you had that reaction is because 123 00:06:16.360 --> 00:06:19.340 most organizations, I think, are taking this kind of an approach of, 124 00:06:19.350 --> 00:06:23.200 "Here are the tools, go figure it out." And I suppose there's a logic 125 00:06:23.220 --> 00:06:25.770 to that in that we're all kind of experimenting. 126 00:06:25.990 --> 00:06:30.000 And in many ways, I think these companies don't really know how they should be using it. 127 00:06:30.100 --> 00:06:33.970 And so they're essentially manifesting their own uncertainty 128 00:06:33.980 --> 00:06:36.900 around how to use this stuff and saying, "Okay, let's just push it down 129 00:06:36.920 --> 00:06:39.340 and let people kind of play with it and see what comes about." 130 00:06:40.360 --> 00:06:44.160 So I think treating this as something that is a tool that we could potentially 131 00:06:44.260 --> 00:06:47.540 leverage to make us all better, to make us serve our customers better, 132 00:06:47.580 --> 00:06:50.830 to make us produce whatever it is that we're creating better, right? 133 00:06:51.040 --> 00:06:52.880 How can we use this tool to just make us 134 00:06:52.890 --> 00:06:54.740 a more effective and efficient organization? 135 00:06:55.620 --> 00:06:58.080 And I think I am also somewhat, 136 00:06:59.280 --> 00:06:59.740 I would say, 137 00:07:00.720 --> 00:07:02.230 and I'm an optimistic person by nature, but- 138 00:07:02.260 --> 00:07:02.640 [Carolyn] Yes 139 00:07:02.660 --> 00:07:06.928 even if you look back in time, we've had very rapid disruption 140 00:07:06.968 --> 00:07:10.348 and very rapid technological change for literally hundreds of years. 141 00:07:10.388 --> 00:07:10.688 [Carolyn] Yes. 142 00:07:10.728 --> 00:07:14.278 Right? And what has happened during this time is that we have 143 00:07:14.328 --> 00:07:16.328 generally become better off, right? 144 00:07:16.408 --> 00:07:20.048 The average person lives a much more comfortable 145 00:07:20.388 --> 00:07:23.708 and good life today than they did 100 years ago, right? 146 00:07:24.168 --> 00:07:26.648 And so I'm trying to be optimistic about it and say, 147 00:07:26.668 --> 00:07:31.900 okay, generative AI, there's going to be some short-term shuffling, 148 00:07:31.910 --> 00:07:34.518 there's going to be some short-term challenges, but I think that 149 00:07:34.520 --> 00:07:37.688 it will ultimately lead, if kind of the great promise 150 00:07:37.698 --> 00:07:41.688 that at least Silicon Valley believes that it holds for the future, 151 00:07:41.788 --> 00:07:45.628 if even a fraction of that is realized, it will in some ways raise all boats. 152 00:07:46.328 --> 00:07:49.848 [Carolyn] I like that. One of the things I think I'm hearing as an undercurrent 153 00:07:49.868 --> 00:07:51.848 is that we have to be more thoughtful about our work. 154 00:07:51.868 --> 00:07:52.648 [Erik] Mm-hmm. 155 00:07:52.668 --> 00:07:55.608 That it's not just, well, this is how we've always done it, 156 00:07:56.388 --> 00:07:57.208 which is a common phrase. 157 00:07:57.268 --> 00:07:58.000 [Erik] Yeah. 158 00:07:58.008 --> 00:08:00.948 But instead, it's, let's really think about how we're doing, 159 00:08:00.988 --> 00:08:03.228 and to the ends that you just described, right? 160 00:08:03.328 --> 00:08:06.000 To customer satisfaction, to profitability- 161 00:08:06.000 --> 00:08:06.001 162 00:08:06.001 --> 00:08:06.958 [Erik] Mm-hmm 163 00:08:06.988 --> 00:08:10.888 ... to those sorts of things. So it almost sounds like the best 164 00:08:10.948 --> 00:08:14.748 way to think about engaging AI 165 00:08:14.808 --> 00:08:18.468 is to do that through a strategic lens on which work is 166 00:08:18.508 --> 00:08:19.208 important. 167 00:08:19.210 --> 00:08:19.928 [Erik] Mm-hmm. 168 00:08:19.948 --> 00:08:21.528 You're nodding. Does that seem- 169 00:08:21.578 --> 00:08:23.308 ... like an appropriate way to think about it? 170 00:08:23.468 --> 00:08:26.948 [Erik] Absolutely, right. And I think some of the things that folks that take this class 171 00:08:26.968 --> 00:08:29.248 will take away from it are some frameworks to help them think about that. 172 00:08:29.488 --> 00:08:29.548 173 00:08:29.628 --> 00:08:32.629 So there's a really simple framework that I love that is evidence-based, that just 174 00:08:32.708 --> 00:08:36.900 came out, that essentially it breaks down AI usage into two dimensions, right? 175 00:08:36.901 --> 00:08:38.568 And so essentially it helps you decide, 176 00:08:38.678 --> 00:08:41.009 okay, is this a good use case for AI or not? 177 00:08:41.200 --> 00:08:45.000 The first dimension is, can I build an AI agent 178 00:08:45.028 --> 00:08:48.268 or custom or chatbot, whatever it is, that is actually good at this, right? 179 00:08:48.308 --> 00:08:50.528 Because we know that these systems have limitations, right? 180 00:08:50.548 --> 00:08:51.888 They're not good for everything, right? 181 00:08:51.898 --> 00:08:54.600 So the first question is, can I build something 182 00:08:54.618 --> 00:08:55.848 that can actually do a good job of this? 183 00:08:55.858 --> 00:08:58.836 If what you build is crappy, you're better off not doing it. 184 00:08:58.837 --> 00:08:58.838 185 00:08:58.848 --> 00:09:02.768 Right? So that's kind of step one. Step two is, is this a thing 186 00:09:02.828 --> 00:09:06.428 that people expect a person to do, right? 187 00:09:06.748 --> 00:09:07.088 So think- 188 00:09:07.098 --> 00:09:07.768 [Carolyn] Tell me more about that. 189 00:09:07.928 --> 00:09:09.328 Yeah. So think, for example, 190 00:09:10.168 --> 00:09:12.428 a performance review conversation, right, with your manager. 191 00:09:12.438 --> 00:09:12.448 192 00:09:12.458 --> 00:09:14.747 That is something that I think most people would say 193 00:09:15.068 --> 00:09:16.628 they really need their manager to be doing that. 194 00:09:16.728 --> 00:09:19.648 That should not be something that comes through and you have a chatbot telling you 195 00:09:19.848 --> 00:09:21.300 whether you're meeting your goals or not, 196 00:09:21.310 --> 00:09:23.100 and then whether you're going to be promoted, 197 00:09:23.110 --> 00:09:24.878 or whether you're going to get a bonus, 198 00:09:24.888 --> 00:09:25.900 or whether you're going to go on a PIP, right? 199 00:09:25.908 --> 00:09:26.358 These are not things 200 00:09:26.358 --> 00:09:27.928 that you want a bot telling you, right? 201 00:09:27.988 --> 00:09:28.348 202 00:09:28.388 --> 00:09:29.377 That's a uniquely human thing. 203 00:09:29.488 --> 00:09:31.558 At the same time, if what you want 204 00:09:31.658 --> 00:09:33.558 is a refund on your Amazon order 205 00:09:33.808 --> 00:09:35.488 that came damaged 206 00:09:35.628 --> 00:09:35.788 207 00:09:35.908 --> 00:09:37.848 you're probably fine just having a chatbot process that. 208 00:09:38.288 --> 00:09:41.377 At the same time, though, if that Amazon situation is a little more complicated, 209 00:09:41.428 --> 00:09:44.388 maybe you got the wrong item, maybe the chatbot isn't helping you, 210 00:09:44.398 --> 00:09:45.848 you need a way to escalate to a person, right? 211 00:09:45.898 --> 00:09:45.898 212 00:09:45.908 --> 00:09:48.308 And so what the research says, really think about those two dimensions, right? 213 00:09:48.428 --> 00:09:51.508 Can I build an AI agent that does a good job of this? 214 00:09:51.568 --> 00:09:51.888 215 00:09:51.948 --> 00:09:55.298 If it doesn't, don't do it. And is this something that people 216 00:09:55.328 --> 00:09:58.438 perceive only other people should really be doing, right? 217 00:09:58.928 --> 00:10:02.848 And if the answer to either of those questions is no, then don't do it. 218 00:10:02.858 --> 00:10:05.428 If the answer is yes, then you can use generative AI. 219 00:10:06.188 --> 00:10:10.028 [Carolyn] What's interesting about that is that I think one of the fears people have, 220 00:10:10.488 --> 00:10:13.868 who don't really understand or have not played with this a lot, 221 00:10:13.928 --> 00:10:16.528 is that what exactly you described will be happening. 222 00:10:16.568 --> 00:10:20.018 They will have bots telling them what their bonus is, 223 00:10:20.128 --> 00:10:23.268 and there's a lot of fear associated with that. 224 00:10:23.478 --> 00:10:25.088 And so hearing you say, 225 00:10:25.098 --> 00:10:25.100 226 00:10:26.001 --> 00:10:28.328 "No, we need humans to be part of the process," 227 00:10:28.408 --> 00:10:31.100 especially when people expect them to be there, 228 00:10:31.110 --> 00:10:32.000 I think is another thing that'll 229 00:10:32.100 --> 00:10:35.108 make people take a deep breath and [grimaces] 230 00:10:35.148 --> 00:10:36.500 [Erik] Yeah. [Carolyn] Exactly. 231 00:10:36.510 --> 00:10:38.468 Well, that's an aha moment for me. 232 00:10:38.488 --> 00:10:40.448 What other aha moments do you think 233 00:10:40.488 --> 00:10:42.128 folks will experience in the class with you? 234 00:10:42.208 --> 00:10:46.428 [Erik] Yeah. You know, I think instead of aha, 235 00:10:46.448 --> 00:10:48.558 what I hope people will come out is more of a 236 00:10:49.208 --> 00:10:50.048 "hmm." 237 00:10:50.108 --> 00:10:50.308 238 00:10:50.388 --> 00:10:52.208 I don't know how satisfying that is. 239 00:10:52.218 --> 00:10:53.388 [Carolyn] That's lovely. No, I like it. 240 00:10:53.468 --> 00:10:57.268 A "hmm." And to me, the big "hmm" is really something I referenced earlier. 241 00:10:57.328 --> 00:11:00.958 This is a grand experiment that we're all taking part in 242 00:11:00.988 --> 00:11:03.708 together, and we don't know how it's going to end up. 243 00:11:03.748 --> 00:11:06.768 We don't even know in a month what's going to change, potentially. 244 00:11:07.168 --> 00:11:10.998 And I think what I'm going to hope to get across to folks is the importance to 245 00:11:11.088 --> 00:11:14.608 give each other grace, the importance to lean into the 246 00:11:14.668 --> 00:11:16.728 uniquely human things that leaders do. Right? 247 00:11:16.798 --> 00:11:20.528 And that is empathy, that's building relationships with folks, that's helping 248 00:11:20.548 --> 00:11:24.100 people develop their career, develop towards the accomplishment of 249 00:11:24.110 --> 00:11:26.000 a shared group of goals. 250 00:11:26.048 --> 00:11:29.068 None of that's going away from the role of a leader. 251 00:11:29.308 --> 00:11:32.808 Generative AI is just this thing that we can use to help us 252 00:11:32.848 --> 00:11:36.208 accomplish those goals and help our people and our 253 00:11:36.548 --> 00:11:39.548 teams do a better job, right? And that's how we should be thinking about it, 254 00:11:39.588 --> 00:11:42.928 and how we can leverage it for those ultimate goals. 255 00:11:42.938 --> 00:11:46.448 So I'm hoping that "hmm" is really that notion of, okay, let's 256 00:11:46.468 --> 00:11:50.448 give each other grace. Let's realize that me, as a leader, I might be a lot 257 00:11:50.468 --> 00:11:54.048 more excited about this thing than some of the rank-and-file folks at my company. 258 00:11:54.068 --> 00:11:56.588 And there's some research that shows that that is generally the case, 259 00:11:56.608 --> 00:11:58.000 that leaders are much more excited about generative AI 260 00:11:58.100 --> 00:12:00.908 than the rank-and-file folks for a variety of reasons. 261 00:12:01.000 --> 00:12:02.978 And how can I hold space for that, 262 00:12:02.988 --> 00:12:04.968 and how can I support people and bring them along? 263 00:12:04.988 --> 00:12:08.948 Because ultimately, without bringing people along, you're not going to 264 00:12:08.988 --> 00:12:12.918 see a financial benefit or any sort of benefit from 265 00:12:12.928 --> 00:12:15.628 generative AI, right? If you create these tools, you incorporate them, and people 266 00:12:15.648 --> 00:12:17.708 aren't using them or using them effectively, 267 00:12:17.728 --> 00:12:20.000 it's just not going to move the needle. 268 00:12:20.100 --> 00:12:23.188 So it's in everybody's best interest to approach this in that way. 269 00:12:23.228 --> 00:12:27.168 It's ironic, but the best practice is in change initiatives. 270 00:12:27.228 --> 00:12:30.268 You're saying this is our change initiative, and the best practice in those 271 00:12:30.308 --> 00:12:32.708 initiatives is to make people part of the process. 272 00:12:32.778 --> 00:12:33.258 273 00:12:33.388 --> 00:12:37.348 And so to just come in and say, "Here is AI, use it," would actually fly in the 274 00:12:37.388 --> 00:12:40.968 face of not just the logic you articulated, but of just good change 275 00:12:41.008 --> 00:12:41.458 management- 276 00:12:41.808 --> 00:12:42.278 [Erik] Yeah. Absolutely 277 00:12:42.278 --> 00:12:43.388 ... in the same way. 278 00:12:43.428 --> 00:12:45.138 [Erik] Yeah. And I love that you brought that up, right? 279 00:12:45.208 --> 00:12:48.788 Because the approach that I've taken in my own teaching over the past several 280 00:12:48.848 --> 00:12:51.548 years is to essentially go back to first principles, right? 281 00:12:52.308 --> 00:12:56.301 We don't know a lot about generative AI and how best to incorporate it into 282 00:12:56.352 --> 00:12:58.592 work systems and the like, we're still figuring that out. 283 00:12:58.632 --> 00:13:02.512 We know a lot about change management, and we know a lot about good 284 00:13:02.532 --> 00:13:05.432 change management principles. And those are what they are. 285 00:13:05.512 --> 00:13:08.912 It doesn't matter if the change, if it's generative AI or any number of other things. 286 00:13:08.932 --> 00:13:11.922 And we can go back to those principles of things that we know about 287 00:13:11.952 --> 00:13:15.532 leadership and management, and then just think about generative AI as a context. 288 00:13:15.772 --> 00:13:19.482 So generative AI is just a context for change, and we can still leverage a lot of 289 00:13:19.512 --> 00:13:21.792 those same things that we know towards that end. 290 00:13:22.362 --> 00:13:25.952 I love that. And so the idea that I know how to structure work, this is just 291 00:13:26.012 --> 00:13:29.792 another tool I have to do that. I know how to do customer satisfaction. 292 00:13:29.832 --> 00:13:31.572 This is just another tool I have to do that. 293 00:13:31.712 --> 00:13:31.902 294 00:13:31.902 --> 00:13:35.352 I think that perspective is one that's really helpful for managers as they're 295 00:13:35.392 --> 00:13:39.788 trying to share that this is a safe place to experiment with folks, 296 00:13:39.788 --> 00:13:39.800 297 00:13:39.871 --> 00:13:39.872 298 00:13:39.892 --> 00:13:41.752 that, kind of, reinforcing what they already know. 299 00:13:41.762 --> 00:13:42.492 [Erik] Absolutely. 300 00:13:42.732 --> 00:13:46.212 So you just referenced something that I'd like to dig a little deeper on. 301 00:13:47.152 --> 00:13:51.142 This has taken all of us by storm. So you, as someone who is a 302 00:13:51.572 --> 00:13:54.322 manager of people and someone who is 303 00:13:54.732 --> 00:13:58.412 rapidly adopting whatever AI seems to make sense for you, 304 00:13:59.232 --> 00:14:03.052 what have been your favorite tricks, tips, and 305 00:14:03.652 --> 00:14:05.872 if there's any place that's gone woefully wrong, 306 00:14:05.972 --> 00:14:07.732 that would be good to know as well. 307 00:14:07.952 --> 00:14:11.532 So I love that you ended with that element of woefully wrong because I do think a 308 00:14:11.792 --> 00:14:15.682 big part of what leaders have to do in this context is 309 00:14:15.712 --> 00:14:19.112 model like, "Hey, I tried it for this and I screwed it up. 310 00:14:19.332 --> 00:14:23.192 "And this is what I learned from that experience," to show people that we're all 311 00:14:23.202 --> 00:14:24.812 doing this together. 312 00:14:24.872 --> 00:14:28.952 So some of my tools that I like. So I love building customs in ChatGPT. 313 00:14:28.992 --> 00:14:32.352 So I have a whole list of customs that I use for various tasks, 314 00:14:32.412 --> 00:14:36.032 whether it's giving students feedback on assignments, whether it's helping me plan 315 00:14:36.092 --> 00:14:39.232 a lesson, whether it's helping me look for company examples to illustrate a point 316 00:14:39.272 --> 00:14:42.000 I'm trying to make in class. And then I love using 317 00:14:42.100 --> 00:14:43.692 Google's Nano Banana tool. 318 00:14:43.702 --> 00:14:43.802 319 00:14:43.802 --> 00:14:47.692 Amazing tool to generate images for my decks that I use in class and the like. 320 00:14:47.952 --> 00:14:50.892 And I always tell students this was generated by generative AI. 321 00:14:51.432 --> 00:14:55.012 I often have some sort of joke because there's always, or typically, something just 322 00:14:55.072 --> 00:14:58.212 slightly off about the image, which also is also good modeling. 323 00:14:58.252 --> 00:14:59.322 Like, "Hey, this 324 00:15:00.372 --> 00:15:02.212 was kind of a crappy job that this thing did." 325 00:15:03.172 --> 00:15:05.412 And that always makes for a fun conversation with students as well. 326 00:15:05.472 --> 00:15:08.472 But yeah, I use customs in ChatGPT for any number of things because that helps you 327 00:15:08.512 --> 00:15:12.432 kind of train almost like a pseudo personal assistant that's kind 328 00:15:12.472 --> 00:15:15.992 of bracketed in what it is that they are going to help you with and what they know, 329 00:15:16.432 --> 00:15:20.222 while still kind of pulling on the great processing power that ChatGPT has compared 330 00:15:20.232 --> 00:15:24.902 to building something at home or in an office that pulls from a local server. 331 00:15:24.932 --> 00:15:27.592 It's kind of like, to me, best of both worlds that you can get. 332 00:15:28.132 --> 00:15:30.002 So I've used those extensively and then, 333 00:15:30.010 --> 00:15:32.761 yeah, huge fan of Nano Banana for pictures. 334 00:15:32.800 --> 00:15:33.652 [Carolyn] Nano Banana. 335 00:15:33.692 --> 00:15:33.702 336 00:15:33.712 --> 00:15:36.892 I have not tried that one, but you better bet that's where I'm heading when we're done. 337 00:15:36.912 --> 00:15:38.852 That sounds like way too much fun. 338 00:15:39.512 --> 00:15:43.172 I think one of the things that as I'm listening, at least this is what's 339 00:15:43.192 --> 00:15:46.892 happened with me, is as I do try to use Chat 340 00:15:46.992 --> 00:15:51.000 or Gemini for giving students feedback, for example, 341 00:15:51.100 --> 00:15:54.152 it means that I have to be so much more explicit than 342 00:15:54.252 --> 00:15:58.072 I ever have about what my expectations are, about what, 343 00:15:58.572 --> 00:16:00.412 especially telling them what they've done well. 344 00:16:01.000 --> 00:16:03.392 Because I'm not that nice. 345 00:16:03.402 --> 00:16:05.972 It's really easy for me to say, "Oh, here's why your points are coming off." 346 00:16:05.992 --> 00:16:08.512 I'm not very good at giving it positive feedback. 347 00:16:08.892 --> 00:16:11.532 So I've been using AI to give positive feedback 348 00:16:11.602 --> 00:16:14.152 for a while now because that's my character flaw. 349 00:16:14.572 --> 00:16:14.672 350 00:16:15.252 --> 00:16:18.942 But is there a translation there, too, in terms of 351 00:16:19.432 --> 00:16:21.792 we have to think about that evaluation academically? 352 00:16:22.292 --> 00:16:26.041 Do managers also need to go through that process if they're asking 353 00:16:26.072 --> 00:16:27.932 AI to help in evaluating their people? 354 00:16:27.972 --> 00:16:31.472 I mean, do we have to go deeper in our standard development? 355 00:16:31.552 --> 00:16:33.712 [Erik] Yeah I would say absolutely. 356 00:16:34.000 --> 00:16:36.512 And I think where generative AI can really help is 357 00:16:36.572 --> 00:16:40.032 in just doing the kind of the repetitive work of evaluations. 358 00:16:40.072 --> 00:16:43.992 Whether that's crafting the letters that you can provide it with templates 359 00:16:44.012 --> 00:16:45.952 and let it know what it is that you want it to focus on. 360 00:16:45.981 --> 00:16:46.882 361 00:16:46.912 --> 00:16:49.552 I think it can also be really helpful in terms of helping you gather notes. 362 00:16:49.632 --> 00:16:53.192 So one of the best practices in performance management is that if you wait until 363 00:16:53.232 --> 00:16:56.932 the end of the year and you save all your kind of juice over the whole year of 364 00:16:56.992 --> 00:17:00.192 everything that you've seen this person do or not do, and then just deliver it in 365 00:17:00.252 --> 00:17:04.142 one blah moment, first of all, you're going to have forgotten half the things, 366 00:17:04.212 --> 00:17:07.172 if not more than half the things. The person's not going to hear it because you're 367 00:17:07.212 --> 00:17:10.572 telling them in December about something they did in February, and they don't know, 368 00:17:10.672 --> 00:17:12.633 even if it's a good thing or a bad thing, whatever it is. 369 00:17:13.042 --> 00:17:13.042 370 00:17:13.052 --> 00:17:16.302 And so generative AI can be really great in helping you be more methodical about 371 00:17:16.333 --> 00:17:20.312 taking those notes and helping you synthesize them into something cohesive at the 372 00:17:20.333 --> 00:17:21.512 end of the year that you can use. 373 00:17:22.032 --> 00:17:25.000 [Carolyn] And maybe I could even take a custom and say, 374 00:17:25.032 --> 00:17:28.272 "Here's this employee," and just keep my notes as long as 375 00:17:28.312 --> 00:17:30.092 it's all under the right data protection, et cetera. 376 00:17:30.332 --> 00:17:30.652 377 00:17:30.692 --> 00:17:32.312 And then synthesize at the end of the year. 378 00:17:32.592 --> 00:17:36.232 [Erik] Absolutely, yeah. And generative AI is so good at taking 379 00:17:36.312 --> 00:17:40.192 unstructured, messy information, like the notes that you 380 00:17:40.232 --> 00:17:44.002 might jot down right after a meeting with someone, or in a meeting, it's 381 00:17:44.012 --> 00:17:46.262 something you've observed, and you just kind of scribble it down. 382 00:17:46.272 --> 00:17:50.012 It's really great at taking that information and putting it in 383 00:17:50.092 --> 00:17:53.232 something that's more legible and more cogent, frankly. 384 00:17:53.312 --> 00:17:55.612 And I use it extensively for those kinds of tasks. 385 00:17:55.972 --> 00:17:58.282 I love that you brought up giving the students feedback example, 386 00:17:58.312 --> 00:18:00.971 that I built this custom for that. 387 00:18:00.982 --> 00:18:01.082 388 00:18:01.112 --> 00:18:03.542 It took me a ton of trial and error 389 00:18:03.822 --> 00:18:03.822 390 00:18:03.852 --> 00:18:05.022 for the reason you just described. 391 00:18:05.252 --> 00:18:08.732 Like being able to define what success looks like is so important 392 00:18:09.312 --> 00:18:09.322 393 00:18:09.322 --> 00:18:13.082 for the custom, but it's also important for us as leaders. 394 00:18:13.292 --> 00:18:17.212 So I think it pushes us to be more explicit, and I've noticed that when 395 00:18:17.272 --> 00:18:20.812 I've assigned a project to my students and I'm not super clear about what I expect 396 00:18:20.832 --> 00:18:23.572 from them, what I get back from the generative AI is frankly garbage. 397 00:18:23.612 --> 00:18:25.000 It's all over the place. 398 00:18:25.100 --> 00:18:27.532 It's not understanding what it is that I expect because I 399 00:18:27.552 --> 00:18:29.672 have not been clear with those expectations to it. 400 00:18:30.432 --> 00:18:34.092 Whereas where it does a really nice job is when I've articulated a really 401 00:18:34.252 --> 00:18:37.052 specific set of expectations for that work. 402 00:18:37.172 --> 00:18:37.232 403 00:18:37.952 --> 00:18:39.852 It does a really wonderful job if you do it that way. 404 00:18:40.712 --> 00:18:43.332 I actually even go back a little further and just say, 405 00:18:43.372 --> 00:18:46.332 "Here's the assignment I'm thinking of. Tell me what's not clear about this." 406 00:18:46.572 --> 00:18:46.832 407 00:18:46.942 --> 00:18:49.452 And it'll even help craft the directions 408 00:18:49.692 --> 00:18:49.782 409 00:18:49.812 --> 00:18:51.312 in a way that becomes really helpful. 410 00:18:52.000 --> 00:18:53.832 Oh my gosh. This is fun. 411 00:18:53.888 --> 00:18:54.991 My pleasure. 412 00:18:55.000 --> 00:18:58.332 And I think when folks come into a conversation in the classroom, they're going to 413 00:18:58.372 --> 00:19:01.632 have the day with you, then they'll have the online half day with you, 414 00:19:01.692 --> 00:19:06.552 and I get the sense they're going to walk away like I just did with my shoulders down. 415 00:19:06.612 --> 00:19:10.212 But if you were to say to someone, "You should take this class because," 416 00:19:10.292 --> 00:19:13.332 What would the final pitch be? 417 00:19:13.342 --> 00:19:13.352 418 00:19:13.712 --> 00:19:17.592 Yeah. So my final pitch would be, take this class 419 00:19:17.692 --> 00:19:21.472 if you want to take some of the things you already know 420 00:19:21.482 --> 00:19:24.832 about being a leader and put them in the context of an 421 00:19:24.892 --> 00:19:28.352 ever-changing world where you've got to meet people where they are, where you 422 00:19:28.432 --> 00:19:32.112 yourself also are facing these pressures of needing to upskill and needing to 423 00:19:32.132 --> 00:19:36.272 understand, okay, what does my team really need to be able to succeed in this world? 424 00:19:36.332 --> 00:19:38.712 That to me is the ideal person for this role. 425 00:19:38.732 --> 00:19:41.582 Someone that has some past leadership experience. You don't need to have it. 426 00:19:41.832 --> 00:19:45.482 If you don't and you're an aspiring people leader, I think you'll still get a lot 427 00:19:45.512 --> 00:19:48.452 out of this course. But someone that has some of this experience and can bring it 428 00:19:48.492 --> 00:19:51.852 to class, and we'll spend quite a bit of time digging into that and thinking 429 00:19:51.872 --> 00:19:55.752 through how it is that you can use these tools in a way that really 430 00:19:55.792 --> 00:19:56.972 benefits you and your team. 431 00:19:57.432 --> 00:20:01.282 So rather than just, "Hey, AI, it's something shiny," it's, "Here is your actual 432 00:20:01.292 --> 00:20:04.412 practical use case that I want you to walk away being able to do." 433 00:20:04.452 --> 00:20:05.452 [Erik] Yep, absolutely. 434 00:20:05.592 --> 00:20:07.000 I want to be there. 435 00:20:07.001 --> 00:20:07.002 436 00:20:07.012 --> 00:20:07.802 Just selfishly. 437 00:20:07.812 --> 00:20:08.412 I'd love to have you. 438 00:20:08.652 --> 00:20:11.498 I would love to do that. Eric, thank you. 439 00:20:11.500 --> 00:20:11.502 440 00:20:11.512 --> 00:20:11.971 Appreciate this. 441 00:20:11.982 --> 00:20:13.112 Of course. My pleasure. Thank you.
