Latest / The Adults In The AI Room / The Power of AI - Glenn Gow - Earley AI Podcast with Seth Earley - Episode # 032

Transcript

The Power of AI - Glenn Gow - Earley AI Podcast with Seth Earley - Episode # 032 · The Adults In The AI Room

• 0:01
  SPEAKER_01: This is the Early AI Podcast, hosted by Seth Early and Chris Featherstone.
• 0:06
  SPEAKER_01: Join us as we delve deep into the passions, expertise, and experiences of thought leaders and practitioners to talk about what's possible in artificial intelligence.
• 0:15
  SPEAKER_01: Early AI Podcast is sponsored by Early Information Science, your digital transformation journey with the design and deployment innovative technology solutions.
• 0:27
  SPEAKER_02: So good morning, good afternoon, good evening.
• 0:30
  SPEAKER_02: Welcome to today's podcast.
• 0:32
  SPEAKER_02: I'm super excited about our guest today.
• 0:35
  SPEAKER_02: He is known as the AI guy and the CEO therapist.
• 0:40
  SPEAKER_02: I want to hear more about that.
• 0:41
  SPEAKER_02: His direct experience and background as a CEO makes him uniquely qualified to help CEOs grow their company's valuation.
• 0:48
  SPEAKER_02: He's an advisor of the Technology Ventures program at Stanford University, a board member at Soteria Intelligence.
• 0:56
  SPEAKER_02: He's a CEO coach at the Peak Performance CEO Coach Program.
• 1:00
  SPEAKER_02: Len Gal, welcome to the program.
• 1:03
  SPEAKER_00: Beth, it's great to be here.
• 1:05
  SPEAKER_00: Thank you for having me.
• 1:06
  SPEAKER_02: I'm really excited.
• 1:07
  SPEAKER_02: And thank you so much for your time.
• 1:09
  SPEAKER_02: So I wanted to start out with you know, what are the things that you're seeing in the marketplace that are uh typical misconceptions or myths?
• 1:18
  SPEAKER_02: What are what are CEOs kind of walking around with that that you see that uh that really needs kind of a dis this uh dispelling?
• 1:25
  SPEAKER_02: What kind of uh issues are they operating under that that may not be accurate today?
• 1:31
  SPEAKER_00: Well, um there there are quite a few actually that I see.
• 1:35
  SPEAKER_00: One, if uh when when CEOs enter the world of AI and they start thinking of it as Chat GPT, uh the first thing I point out is well, ChatGPT is just a tool.
• 1:46
  SPEAKER_00: It represents one area and it only represents an area of generative AI.
• 1:52
  SPEAKER_00: What may be even more valuable to CEOs is what I call predictive AI or analytical AI, which is the more historical world of AI.
• 2:02
  SPEAKER_00: And what I'm excited about is that at least ChatGPT has brought to the forefront the power of AI.
• 2:11
  SPEAKER_00: And so what I think it's important for a CEO to do is say, help me understand both sides of the equation.
• 2:17
  SPEAKER_00: And I'm gonna say one side is generative AI, and the other is all the rest of the AI that people have been working on for the last 50 or more years.
• 2:25
  SPEAKER_00: And so it opens a door for CEOs to now recognize how this can impact their company and their industry.
• 2:34
  SPEAKER_00: So that's I think that's a bit of a myth that ChapTPT is a just you know how they think about AI alone.
• 2:42
  SPEAKER_00: Um, but the other thing it's doing is it's helping CEOs recognize something that is a theme that Seth, I'm sure you've said to your own clients, which is this is happening faster than we expected it to.
• 2:57
  SPEAKER_00: And the curves are starting to look exponential.
• 3:02
  SPEAKER_00: And so for a CEO, what that means is I really need to pay attention to this because I don't understand yet the implications in my industry.
• 3:13
  SPEAKER_00: And and uh there's there's a concept in AI that I call the winner takes all concept, which is and it's more it's less on the generative AI side, which is if you become better at AI than your direct competitor, and you continually collect more data about your customers and you become smarter about those customers.
• 3:37
  SPEAKER_00: And the AI, because it can learn every single day, or sometimes even all by itself, it creates a flywheel effect where you have the opportunity to dominate a market if your competitors aren't on that same flywheel.
• 3:54
  SPEAKER_00: And if you're left behind, you could reach a point where you're never going to catch up to a competitor that's diving into the AI opportunity.
• 4:04
  SPEAKER_00: It's that transformative.
• 4:06
  SPEAKER_02: Yeah, that's a great, that's a great point.
• 4:08
  SPEAKER_02: And um, I sometimes tell the story of uh a customer many years ago, uh, and this is related to AI, but it's also foundational uh IA information architecture, which of course that I focus on a lot.
• 4:23
  SPEAKER_02: And what was happening is they were using uh at the time, they were using uh uh traditional method for building textbooks, and they were doing K through 12 uh textbook manufacturing.
• 4:36
  SPEAKER_02: And essentially there's a lot of variation depending upon state and local uh requirements and grade requirements and so on.
• 4:43
  SPEAKER_02: And they found that their competitors were getting to market much faster than they were, they were beating them to market by six months, and it turned out that they were using componentized content, they were using uh machine learning at the time.
• 4:56
  SPEAKER_02: This is many years ago, they're using machine learning to auto-categorize and componentize their content, and they were able to come up with a um a draft of a textbook six months faster than this company was, and they had a million objects in the repository, and the CEO said, Well, why aren't we doing that?
• 5:14
  SPEAKER_02: And the staff said, Well, that's the project we've been trying to get off the ground for the last three years, and they said, Great, let's do it now.
• 5:21
  SPEAKER_02: And how long will it take?
• 5:22
  SPEAKER_02: Well, it's gonna take at least a year or two, and the competition is gonna be moving ahead.
• 5:26
  SPEAKER_02: And guess what?
• 5:27
  SPEAKER_02: They lost that market, it was too late for them, and I think we're gonna see the same thing.
• 5:33
  SPEAKER_02: We're gonna see organizations that just are not going to adapt fast enough and they're going to lose their market uh positioning.
• 5:40
  SPEAKER_00: Well, well, Seth, I am so interested that you told that particular story because I don't know if you followed this, but in sort of an adjacent industry to what you just described is CEG.
• 5:53
  SPEAKER_00: And Cheg helps students gather information in order to pass their tests more effectively.
• 6:01
  SPEAKER_00: And their stock dropped 45% in a single day because the CEO said Chat GPT is where all our students are going now.
• 6:09
  SPEAKER_00: They're being disintermediated.
• 6:12
  SPEAKER_00: Wow.
• 6:12
  SPEAKER_00: Yeah.
• 6:13
  SPEAKER_00: And it's a stunning 45% in a single day is pretty stunning.
• 6:18
  SPEAKER_00: And that they're in trouble.
• 6:19
  SPEAKER_00: That that that company is in trouble because of this new technology that's just shown up.
• 6:24
  SPEAKER_02: Right.
• 6:24
  SPEAKER_02: And they're going to have to reinvent themselves in some other way and be very creative about that process.
• 6:30
  SPEAKER_02: You know, I mean, chat GPT, of course, is transforming lots of industries.
• 6:35
  SPEAKER_02: It's getting all the attention, it's sucking the air out of the room, right?
• 6:41
  SPEAKER_02: Uh, but you know, but there's still a big gap between uh using Chat GPT for publicly available information and using it internally in the organization.
• 6:50
  SPEAKER_02: Do you want to speak to that and talk a little bit about some of those hurdles?
• 6:54
  SPEAKER_02: Because again, you have to do things behind your firewall.
• 6:57
  SPEAKER_02: You can't let stuff out into the wild and have it become public.
• 7:02
  SPEAKER_02: I know Samsung made that mistake, but talk a little bit about that.
• 7:06
  SPEAKER_00: So uh there are a few things that are happening.
• 7:08
  SPEAKER_00: Uh, one is that the big cloud providers, Azure, AWS and GCP, um, and even maybe IBM and Oracle to some extent, are saying, hey, come to us because we're gonna create a walled garden for you.
• 7:24
  SPEAKER_00: We're gonna enable large language model to run, but we're going to protect your data.
• 7:30
  SPEAKER_00: And what we're gonna do is we're gonna let you put your corporate data in this walled garden.
• 7:35
  SPEAKER_00: We're gonna augment it with whatever large language model we choose from.
• 7:39
  SPEAKER_00: And by the way, that's rapidly expanding the choices, um, so that you can build your own, so to speak, with our support.
• 7:48
  SPEAKER_00: Now, related to that is what's happening with open source.
• 7:51
  SPEAKER_00: So ChatGPT uh four, I believe, uh, is not open source, whereas the prior versions were, but that, but but they're not the only party in town.
• 8:01
  SPEAKER_00: Right.
• 8:02
  SPEAKER_00: Right.
• 8:02
  SPEAKER_00: And now because there's open source, there's a tremendous movement in the open source world to create many variations of large language models.
• 8:12
  SPEAKER_00: And so now it's gonna become easier for enterprises to say, I want to take that open source, put it in that call it AWS framework, load our data in it, and now I'm gonna have access and the capability to do all this work inside my organization that I couldn't do before.
• 8:31
  SPEAKER_00: So Bloomberg is an example.
• 8:33
  SPEAKER_00: They announced, hey, guess what?
• 8:35
  SPEAKER_00: We have unique data that ChatGPT doesn't have or Claude doesn't have or uh Facebook doesn't have.
• 8:43
  SPEAKER_00: It's all about financial information and data.
• 8:46
  SPEAKER_00: We're gonna bring out our own large language model.
• 8:49
  SPEAKER_00: So come on.
• 8:50
  SPEAKER_02: There's Bloomberg GPT.
• 8:51
  SPEAKER_02: Yeah, exactly.
• 8:53
  SPEAKER_02: The idea is that there will be lots of different models to choose from.
• 8:57
  SPEAKER_02: And it's not just one size fits all, right?
• 9:00
  SPEAKER_02: There's Lambda, there's BERT, there's a lot of other uh language models.
• 9:04
  SPEAKER_02: And then as you start, as you mentioned, there are specialized language models for particular industries like Cyber, you know, Cybike and Med GPT and Chem GPT.
• 9:16
  SPEAKER_02: There's a whole bunch of them that are for the industries themselves.
• 9:19
  SPEAKER_00: Exactly.
• 9:20
  SPEAKER_00: Exactly.
• 9:21
  SPEAKER_00: So I see this uh this explosion of opportunities for anybody who's looking to take advantage of this and to protect their data.
• 9:29
  SPEAKER_00: Although this is very, very early, right?
• 9:31
  SPEAKER_00: We haven't quite seen anybody implement this yet for let's say uh uh uh well there is just going on right now, the implementations are happening, but yeah, that just means opportunities are massive.
• 9:41
  SPEAKER_00: And going back to your first question, Seth, I'm concerned that CEOs don't yet understand the implications of speed in this area and what's going to be required of them to change very, very rapidly so that they are not disintermediated as a company.
• 10:02
  SPEAKER_02: Right.
• 10:02
  SPEAKER_02: So, you know, when you think about things like ChatGPT and GPT-4 and Lambda and BERT and all of these other models, you know, those are, as you say, there's some that are open source now and people can access them.
• 10:17
  SPEAKER_02: But when you think about it, and I've talked about this before, you know, standardization will give you efficiency, but differentiation gives you competitive advantage.
• 10:27
  SPEAKER_02: So there still has to be a mechanism to control your knowledge and have your own versions of uh ontologies and language models that are fine-tuned to your organization.
• 10:41
  SPEAKER_02: You know, we're finding that even the science-based models, we're working with a life sciences firm, and the science model that they're using is too big, it's too bulky.
• 10:51
  SPEAKER_02: It has terminology that's not related to them.
• 10:53
  SPEAKER_02: You know, they they don't deal with animal diseases.
• 10:55
  SPEAKER_02: There's animal models of diseases, but there's not, they're not treating animal diseases, and animal diseases are part of this large language model.
• 11:03
  SPEAKER_02: So you get a lot of terminology that's not relevant, and so that has to be trimmed down.
• 11:08
  SPEAKER_02: And then there's specialized language for their drugs and their treatments and their mechanisms of action and things that are non-public.
• 11:15
  SPEAKER_02: So maybe you can talk a little bit about that in terms of how that differentiation uh versus um efficiency might play out in organizations.
• 11:25
  SPEAKER_00: I I've seen it from a different perspective, Seth.
• 11:28
  SPEAKER_00: So let me share you with uh with you what I've seen.
• 11:31
  SPEAKER_00: Um, and it's in the area of prompting, which I'm sure you're you're deeply familiar with.
• 11:35
  SPEAKER_00: And the way I describe prompting is you take a large language model and you shrink it, you shrink it to the area of focus that you want the model to focus upon.
• 11:47
  SPEAKER_00: And so, for example, I don't know the large language model you're using in the example of the medical technology, but I could literally prompt it to ignore information related to animals.
• 12:00
  SPEAKER_00: And I could specifically have it focus on cardiology or radiology or something of that nature.
• 12:06
  SPEAKER_00: Um, so there's um um one of the nice things about large language models is you can, even if you're not good at prompting, you can ask it how you should prompt it, and it will tell you how to prompt it.
• 12:20
  SPEAKER_00: Um the second thing I see happening is that this whole prompting thing that we're talking about is very ethereal.
• 12:27
  SPEAKER_00: So, for example, in stable diffusion, right now, I don't know if you've had a chance to use that, um it it it can take a simple prompt and it will give you a better prompt.
• 12:38
  SPEAKER_00: Right.
• 12:39
  SPEAKER_00: Say, no, that's that's that's not sufficient.
• 12:42
  SPEAKER_00: Um, what I'm suggesting is you you prompt me this way, and that's going to be fully automated to where we're just in a conversation with the model until we get to a narrow place where we can create something or search something or evaluate something that's particularly relevant to us.
• 13:02
  SPEAKER_00: So this is all happening as we speak.
• 13:04
  SPEAKER_02: You know, it's it's it is interesting.
• 13:06
  SPEAKER_02: I mean, there's there's a lot of things that are happening that are kind of emerging.
• 13:09
  SPEAKER_02: Do you want to talk more about um an example of that kind of uh prompting where do you have an example off the top of your head?
• 13:16
  SPEAKER_02: If you don't, that's no problem.
• 13:17
  SPEAKER_02: But where you might give it a an ambiguous prompt and it comes back and asks you or suggests a more precise prompt.
• 13:24
  SPEAKER_00: I do a fair amount of writing and I use support from ChatGPT to help me with research and ways of thinking about this and even some of the words I might choose to use, right?
• 13:33
  SPEAKER_00: And I treat it as a almost like an intern that's doing work, but I will go to ChatGPT and I'll say, I'm going to write an article about uh the importance of AI to CEOs and what specifically they should be thinking about.
• 13:50
  SPEAKER_00: What information do you need from me to help you narrow this down for my audience?
• 13:59
  SPEAKER_00: Right.
• 14:00
  SPEAKER_00: And it will come back and say, well, answer this question and this question and this question.
• 14:04
  SPEAKER_00: So essentially it's building the prompt that I need to give it.
• 14:08
  SPEAKER_00: And I'm I I treat it as a conversation.
• 14:12
  SPEAKER_00: So it actually each chat is unique within ChatGPT and it keeps track of our prior conversations.
• 14:21
  SPEAKER_00: And I'm making I want to make sure it deeply understands where I'm going and what I'm trying to accomplish in writing something.
• 14:28
  SPEAKER_00: That's one.
• 14:30
  SPEAKER_00: The second is um in stable diffusion, I used an example and I gave it a prompt.
• 14:38
  SPEAKER_00: I said, show me a logo for a new sneaker line aimed at Gen Z.
• 14:45
  SPEAKER_00: And that's what I call a weak prompt because there's not a lot of richness to it.
• 14:50
  SPEAKER_00: And in stable diffusion, it said um it said, hey, guess what?
• 14:55
  SPEAKER_00: Here's a here's something they call a prettified prompt.
• 14:58
  SPEAKER_00: Like pretty, prettier, prettified prompt.
• 15:02
  SPEAKER_00: And we suggest you you use a lot of other words here, like innovative and uh light and um uh creative and and on and on.
• 15:17
  SPEAKER_00: It gave me so many additional ways to be more specific.
• 15:22
  SPEAKER_00: Now, the beauty of this getting this information, this prompting from stable diffusion, is that the result I received, I wish I could show you, I don't have it in front of me, was stunning.
• 15:32
  SPEAKER_00: It was absolutely stunning.
• 15:34
  SPEAKER_00: I looked at this sneaker that it created a sneaker with this amazing uh logo on it, and I thought, oh my God, if I'm a designer, I'm gonna be able to create fantastic designs with the help of this model.
• 15:49
  unknown: Right.
• 15:49
  SPEAKER_00: And I think that's Seth, I think that's where we really want to go um uh when we're thinking about productivity, is how do we take anything we're gonna be doing in the world uh uh that and become better as humans by using AI as our coach, as our or our our our our recommender or our intern.
• 16:12
  SPEAKER_02: Right.
• 16:12
  SPEAKER_02: No, that's a great point.
• 16:14
  SPEAKER_02: And you know, I I did a little workshop for a school, uh a charter school to help them, their teachers understand how to use this and our administrators.
• 16:22
  SPEAKER_02: And what we did was we compared ambiguous prompts with very specific prompts.
• 16:28
  SPEAKER_00: Yes, yes, yes.
• 16:29
  SPEAKER_02: And so what you're saying is you actually you can ask it to help you with the prompts, you can ask it to enrich prompts, and then it can do that.
• 16:38
  SPEAKER_02: And I also find it it's a great uh source for doing research, for beginning that research process, for building outlines and things like that and finding research topics.
• 16:48
  SPEAKER_02: So I totally agree with you there.
• 16:50
  SPEAKER_02: Um, so it's going to add a lot of productivity.
• 16:52
  SPEAKER_02: What are the other uh areas of the AI that you're seeing uh get some uptake uh in the enterprise?
• 16:59
  SPEAKER_02: And uh, you know, again, ChatGPT is very important.
• 17:02
  SPEAKER_02: It's it's pretty amazing, it's moving very, very quickly.
• 17:05
  SPEAKER_02: I think there's a lot of challenges.
• 17:07
  SPEAKER_02: Maybe we'll come back to that and talk about the challenges.
• 17:09
  SPEAKER_02: But what else are you seeing that's that's getting some good um acceptance and uptake?
• 17:14
  SPEAKER_00: So I want to use a uh I'll tell you a story.
• 17:16
  SPEAKER_00: This is a true story.
• 17:18
  SPEAKER_02: Um about I'll also take made-up stories, by the way.
• 17:24
  SPEAKER_00: Okay, all right.
• 17:25
  SPEAKER_00: I have some of those too.
• 17:28
  SPEAKER_00: Um, and and and the phrase I use is AI can see patterns that humans can't see.
• 17:34
  SPEAKER_00: And we want to take advantage of that.
• 17:37
  SPEAKER_00: So um Frido Lay did this.
• 17:40
  SPEAKER_00: This is a uh a story about what Frida Lay did.
• 17:44
  SPEAKER_00: Um, they're selling lots of different kinds of potato chips, and they're always testing the market with different potato chips, but it's it's a very expensive process to say, hey, would this particular flavor work in this neighborhood or would this uh type of potato chip?
• 18:00
  SPEAKER_00: Because they have all types of potato chips work in this location.
• 18:04
  SPEAKER_00: So that's how they experiment without AI.
• 18:07
  SPEAKER_00: What they they did is they said, you know, why don't we layer on top of our data about who's buying what chips and what locations third-party public data, like the age of our consumers, the economics of our consumers, the the sex of our consumers, the the consumer preferences of our consumers, all of which is available from third parties.
• 18:33
  SPEAKER_00: You can do this based on pure zip codes and say, what's likely to happen in this location?
• 18:39
  SPEAKER_00: So Friday LA discovered that in Frisco, Texas, which is a relatively small town outside of Dallas, there was a large number of ethnic Indians.
• 18:52
  SPEAKER_00: And it turned out that those ethnic Indians were buying this product that was really not very available in the US, or actually actually wasn't available in the US.
• 19:01
  SPEAKER_00: It was curry flavored Cheetos, which was actually only being sold in India.
• 19:07
  SPEAKER_02: I kind of would like curry flavored Cheetos, I think.
• 19:09
  SPEAKER_02: There you go.
• 19:11
  SPEAKER_00: So but what you like curry?
• 19:12
  SPEAKER_02: I like Cheetos Exactly.
• 19:15
  SPEAKER_00: Now, here's what's so important about this is that Frito Lay, without humans having to figure this out, said we need to put curry flavored Cheetos, which we're already manufacturing, but we're sending to India.
• 19:28
  SPEAKER_00: We need to put them on the shelves here.
• 19:31
  SPEAKER_00: So, yes, it's a test, but we're pretty sure this test is gonna work.
• 19:35
  SPEAKER_00: And what happened, what really happened if you think about this as a marketer, is that the AI found a micro market.
• 19:43
  SPEAKER_00: Okay, this market was only 200,000 people, it wasn't 200 million people, it's 200,000 people, but Frito Lay now dominates the shelves.
• 19:54
  SPEAKER_00: There you know, lots of other competitors.
• 19:57
  SPEAKER_02: Yes, yes, yes, yes.
• 20:00
  SPEAKER_00: And so their competitors don't even understand what's going on, right?
• 20:04
  SPEAKER_00: Right.
• 20:05
  SPEAKER_00: And so the question I want to ask any CEO is how many micro markets are there out there that you can serve and dominate where your competitor doesn't even know what's happening by using AI to find the patterns that already exist that we as humans can't see.
• 20:24
  SPEAKER_02: Right.
• 20:24
  SPEAKER_02: That's a really great, that's a really great point.
• 20:26
  SPEAKER_02: I want to I want to take a step back a little bit and understand a little bit more about you in your background and how did you kind of uh get into this space?
• 20:34
  SPEAKER_02: Maybe you can just talk a little bit about uh who you are and your career path and and how uh how you got to where you are today.
• 20:41
  SPEAKER_02: Uh so maybe the world according to Glenn.
• 20:44
  SPEAKER_00: Sure, happy to do that.
• 20:45
  SPEAKER_00: Well, um I was fortunate enough to be a CEO for 25 years, and most of my work was working with large technology companies to help them in strategy.
• 20:54
  SPEAKER_00: So Apple, Google, Facebook, Microsoft, Oracle, all the big players were clients of ours.
• 21:02
  SPEAKER_00: Then, however, I got recruited into venture capital.
• 21:06
  SPEAKER_00: And two things happened in venture capital.
• 21:08
  SPEAKER_00: One, I started coaching CEOs and I fell in love with that work.
• 21:12
  SPEAKER_00: Really, it was amazing.
• 21:14
  SPEAKER_00: And that's what's created my my current uh full-time role, is which is coaching CEOs.
• 21:19
  SPEAKER_00: But the other thing that happens in in venture capital, Seth, it's kind of magical, is that you can predict the future.
• 21:28
  SPEAKER_00: And the reason you can predict the future is you have people walking in the door every day telling you what the future is going to look like.
• 21:35
  SPEAKER_00: Right.
• 21:36
  unknown: Right.
• 21:36
  SPEAKER_00: Right.
• 21:36
  SPEAKER_00: And so when you start to watch the patterns that are happening and what people are talking about, I call them waves of technology.
• 21:44
  SPEAKER_00: You start to see these waves all happening at the same time.
• 21:47
  SPEAKER_00: And what I noticed was this tsunami out on the horizon called AI.
• 21:53
  SPEAKER_00: Right.
• 21:54
  SPEAKER_00: And when you start listening to all these very, very intelligent, cutting edge people talk.
• 22:00
  SPEAKER_00: Talking about what they're going to do with AI, you want to make sense of it.
• 22:04
  SPEAKER_00: And so I dove very deeply into this to begin the under to begin to understand the implications of AI.
• 22:11
  SPEAKER_00: Now, I'm not a deep technologist, Seth.
• 22:14
  SPEAKER_00: You know a lot more about technology than I do.
• 22:17
  SPEAKER_00: My expertise is at the board level and at the senior executive level in translating what's happening in the world of AI so that senior executives and board members make the right decisions about AI.
• 22:31
  SPEAKER_00: Right.
• 22:32
  SPEAKER_00: And so now I have coaches, I'm sorry, I have CEOs that I coach that run AI companies.
• 22:36
  SPEAKER_00: I'm on the board of an AI company, and I speak very often to board members on this topic, in addition to a column in Forbes that I write on AI.
• 22:48
  SPEAKER_02: That's terrific.
• 22:49
  SPEAKER_02: And so what do you see as kind of the biggest stumbling blocks?
• 22:54
  SPEAKER_02: And you know, my my sense is um there's still I I don't want to tell the joke again because the listeners of this podcast have heard it many times, uh, but it's about making assumptions.
• 23:06
  SPEAKER_02: Ah, now I have to tell it because uh for those of you, just just turn turn fast forward, but it's about making and it's about economists making assumptions.
• 23:15
  SPEAKER_02: And the old joke is there's an economist, there's a chemist, and there's a mechanical engineer on a desert island, and they have a container of food, but they don't have an implement to open it.
• 23:23
  SPEAKER_02: So the mechanical engineer is going to use rocks and coconuts and break it open.
• 23:27
  SPEAKER_02: The chemical engineer is going to erode it and use seawater and sunlight.
• 23:31
  SPEAKER_02: And the economist says, assume it can open.
• 23:35
  SPEAKER_02: So, what I have found is that most of the time when you talk to vendors, they'll say things like, Yes, uh, well, how do you how do you evolve this, or how do you make these tools better?
• 23:46
  SPEAKER_02: They'll say, Well, the customer has a knowledge base, or you have to go after the knowledge base, the corporate knowledge base.
• 23:52
  SPEAKER_02: And they're assuming the thing that many organizations don't have a good handle on, right?
• 23:57
  SPEAKER_02: There's a lot of organizations that don't have good corporate knowledge sources, right?
• 24:02
  SPEAKER_02: It's very fragmented, it's very distributed, uh, it's not in good shape, it's not well organized.
• 24:08
  SPEAKER_02: And there's this kind of uh prevailing notion that you can simply point uh the AI technology to bad data or incomplete data or poor content or you know, the incorrect content or content that's not well curated, and that it's going to somehow produce a result.
• 24:26
  SPEAKER_02: And I think that that's kind of the challenge that you know a lot of organizations.
• 24:30
  SPEAKER_02: I mean, we're still doing basic blocking and tackling for a lot of enterprises or you know, getting their knowledge house in order, getting their content house in order.
• 24:39
  SPEAKER_02: And then we're building POCs that are leveraging large language models to access that content.
• 24:43
  SPEAKER_02: But again, there's a lot of assumptions about that content being ready, or that these large language models will handle that.
• 24:51
  SPEAKER_02: Do you run into that same thing?
• 24:53
  SPEAKER_02: And when you when it comes to obstacles in the organization, do you believe it's more cultural, technical?
• 24:59
  SPEAKER_02: Do you run into this issue where people are still wrestling with and struggling with data quality issues?
• 25:05
  SPEAKER_02: And what's what's your point of view or perspective on that type of thing?
• 25:11
  SPEAKER_00: So, Seth, I would say that the vast majority of organizations have large to significant issues with their data.
• 25:20
  SPEAKER_00: And it's because uh more often than not, the data isn't the the data systems are not designed to take advantage of a uh and be treated as a corporate-wide asset.
• 25:36
  SPEAKER_00: They're they're usually functional.
• 25:38
  SPEAKER_00: Oh, we know a lot about our customers over here in the sales organization, or maybe within client success, or maybe within support.
• 25:46
  SPEAKER_00: But oh, and we did some research over here in marketing and engineering talks to them every once in a while, but none of that is brought together in a way that's designed from the corporate level.
• 25:57
  SPEAKER_00: So, what I say to CEOs is look, if you are interested at all in doing significant work in the area of AI, long pull in your tent is getting your data wrangled.
• 26:09
  SPEAKER_00: Yeah, yeah.
• 26:09
  SPEAKER_00: And guess what?
• 26:10
  SPEAKER_00: That can take you a year or more before you feel like we have a true asset here.
• 26:17
  SPEAKER_00: So before you go playing more with Chat GPT this evening, I would focus on what does my organization look like?
• 26:25
  SPEAKER_00: That what is the team that's managing my data?
• 26:28
  unknown: Right.
• 26:28
  SPEAKER_00: And how do we treat that as a deep corporate asset?
• 26:32
  SPEAKER_02: Right.
• 26:33
  SPEAKER_02: Yeah, no, that's a really great point.
• 26:34
  SPEAKER_02: And you know, there's so many organizations that are kind of have this holy grail of the 360-degree view of the customer, and yet that is so distributed.
• 26:44
  SPEAKER_02: There may be 50 or 100 applications that touch that customer.
• 26:49
  SPEAKER_02: And that's right, very infrequently, you know, and all of those representations of the customer are different across those different systems.
• 26:56
  SPEAKER_02: That's right.
• 26:56
  SPEAKER_02: So pulling those together is extremely challenging.
• 27:00
  SPEAKER_02: And then you have lots of, you know, organizational silos, political silos, right?
• 27:04
  SPEAKER_02: Responsibility silos, you know, uh governance silos, funding.
• 27:09
  SPEAKER_02: And so it really takes that uh that that uh C level view of this to say we need to stop the turf wars and we need to to bring people together to look for the common uh uh good and look for the greater good and give up your fiefdom, right?
• 27:26
  SPEAKER_02: Are you seeing that?
• 27:27
  SPEAKER_02: Yeah, go ahead.
• 27:29
  SPEAKER_00: I'll give you another story here.
• 27:31
  SPEAKER_00: Um so I know of a company, a very large company, Fortune 500 company, that said, we want to do what is arguably the most effective marketing campaign, which is we want to market to our current customers to enable them to upgrade what they've purchased from us.
• 27:48
  SPEAKER_00: The problem is we don't know who our customers are.
• 27:53
  SPEAKER_00: Now, granted, it they're sold through a complex set of channels, and there are a lot of reasons why it's not easy to collect that information, but nobody nobody drove the desire to ensure we know who our customers are, and therefore we have lots of data points about our customers, but we might not know who the how to differentiate between the user uh versus the buyer versus the purchasing agent versus the economic decision maker.
• 28:24
  SPEAKER_00: We probably and this company didn't know any of that.
• 28:27
  SPEAKER_02: Yeah, yeah.
• 28:28
  SPEAKER_02: We just went through that process uh with a software uh subscription company that sells to both consumer and business segments.
• 28:36
  SPEAKER_02: And one of the things they were trying to do is they were trying to optimize that customer lifecycle and that customer experience.
• 28:42
  SPEAKER_02: And because they didn't have the correct customer data model, they weren't, they weren't uh describing the customers in terms of those attributes, and they weren't able to identify what stage they're in, they weren't able to impact the conversions, right?
• 28:58
  SPEAKER_02: Because they didn't know how to target, you know, who is in you know, uh the middle of the funnel, what does that mean?
• 29:04
  SPEAKER_02: And what are the descriptors that we can identify that will allow us to give them the content and the information they need to move them to the next stage, right?
• 29:13
  SPEAKER_02: And then how well is our program working?
• 29:16
  SPEAKER_02: Do we have the metrics that can uh determine what content is most effective?
• 29:21
  SPEAKER_02: And we talk about doing things like personalization.
• 29:24
  SPEAKER_02: Well, it requires that we have those customer attribute models, it requires that we have the product data models, it requires that we have the content models, right?
• 29:34
  SPEAKER_02: And all of those things need to be aligned so that we can experiment, we can do multivariate uh testing, we can understand, you know, what the potential uh avenues are that they can bring people through this life cycle and accelerate, you know, from trial to to subscriber, right?
• 29:52
  SPEAKER_02: And so that was a huge problem.
• 29:54
  SPEAKER_02: And again, beginning to get, you know, a lot of organizations don't even know how to describe the problem, right?
• 30:01
  SPEAKER_02: And so starting with the customer lifecycle is a great place to to begin.
• 30:05
  SPEAKER_02: You know, what are you advising organizations when it comes to uh beginning this process or when it comes to you know starting to sort out and get that data house in order?
• 30:16
  SPEAKER_02: You know, what are what are you suggesting?
• 30:18
  SPEAKER_02: Where do they start that process?
• 30:20
  SPEAKER_02: Again, you know, we think of the customer as being a great place to start.
• 30:24
  SPEAKER_02: What are your what are your thoughts on that?
• 30:26
  SPEAKER_00: I think we start with the CEO.
• 30:29
  SPEAKER_00: So let's let's let's pretend I'm in charge of uh data at a big company and I've got a I've got to wrangle data, right?
• 30:38
  SPEAKER_00: Because I don't, I may not own all the data, first of all, right?
• 30:41
  SPEAKER_00: It might some might sit in in different functions, different divisions, different geographies.
• 30:46
  SPEAKER_00: Unless I have the CEO saying to the organization, this is one of the cornerstones of success for this company.
• 30:56
  SPEAKER_00: And we as an organization, as a whole, need to bring together mechanisms to create assets out of that data.
• 31:04
  SPEAKER_00: Unless I have that coverage from on high, it may never happen.
• 31:10
  SPEAKER_00: It may never happen.
• 31:11
  SPEAKER_00: I need I need support.
• 31:12
  SPEAKER_00: And the same thing is true, by the way, of AI.
• 31:16
  SPEAKER_00: This isn't, it isn't so much about uh I'm not gonna say it's AI first, Seth.
• 31:21
  SPEAKER_00: It's more strategy first.
• 31:23
  SPEAKER_00: It's here's a strategic opportunity for us to take a dominant market lead in this space.
• 31:29
  SPEAKER_00: And oh, by the way, um AI is gonna be one way we do that.
• 31:34
  SPEAKER_00: And oh, by the way, the long pole in the tent is data, at least on the on the predictive uh analytics side.
• 31:42
  SPEAKER_00: So when the CEO deems this critical, that's when the organization begins to shift.
• 31:50
  SPEAKER_00: This is um, you know, one of the biggest obstacles to adopting AI in a company is culture.
• 31:56
  SPEAKER_00: It isn't, I mean, yes, okay, they we talked about the data structures, et cetera, but it's it's it's a cultural question of who are we as a company?
• 32:07
  SPEAKER_00: And are we gonna be in fiefdoms saying, no, this is my data, you can't have access to my data.
• 32:13
  SPEAKER_00: Or are we gonna be a company that says, wow, if we can leverage all the rich data we have, we can enable AI to give us incredible insights into what we're gonna be doing going forward, who, what suppliers we work with, what products we build, what consumers we go after, and on and on.
• 32:35
  SPEAKER_02: And that's so true that you know, the vision does have to come from the top.
• 32:40
  SPEAKER_02: And it we have to have leadership that not only understands this, but is ready to allocate the resources and the organizational time and it uh and so some of these things, you know, of course, leaders have been uh burned before, right?
• 32:56
  SPEAKER_02: And they've been burned on big projects that may look to get the data house in order, big master data initiatives, big digital transformations.
• 33:06
  SPEAKER_02: I like to say a digital transformation is a data transformation, but many of those fail.
• 33:11
  SPEAKER_02: And many of those top-down initiatives are not successful.
• 33:15
  SPEAKER_02: And so the question becomes you know, and even a lot of the large global integrators, the large consulting firms, the large uh uh uh the large uh uh uh creative agencies, a large digital agencies, you know, don't necessarily have the incentive to solve the problem at the root, right?
• 33:35
  SPEAKER_02: Because they these are annuities for them, right?
• 33:38
  SPEAKER_02: These are evergreen problems and they're annuities, and they keep staying in the organization and keep doing the same things over and over again.
• 33:46
  SPEAKER_02: So, you know, what are your thoughts about um how to kind of get past that, you know, in terms of making sometimes they're very large investments and you know, directing those investments, making sure that gaining uh the right return on those investments and maintaining the organizational attention?
• 34:04
  SPEAKER_02: Because again, it's very difficult for these long-term engagements to retain the traction, retain the attention, retain the funding that they need.
• 34:14
  SPEAKER_02: And again, part of it is they uh they've been burned, right?
• 34:19
  SPEAKER_02: So at what point do you, you know, do you say, yes, I'm gonna keep supporting this initiative?
• 34:25
  SPEAKER_02: Like what do they have to do to retain their uh uh and again, we look quarter to quarter a lot of times, and sometimes uh tenures of chief digital officers are not very long, right?
• 34:37
  SPEAKER_02: So if they don't get those results very, very quickly, but at the same time, many of these programs require longer ongoing investment.
• 34:45
  SPEAKER_00: Sure.
• 34:45
  SPEAKER_00: Well, let me let me speak to um the AI specific part of this.
• 34:50
  SPEAKER_00: Um and and and I think the best place to start is from a strategy, as I mentioned earlier.
• 34:56
  SPEAKER_00: And what and let's use an example.
• 34:58
  SPEAKER_00: So let's say uh the strategy is we want to uh expand from let's say we're a B2B company or we're selling to the enterprise, we want to expand into the middle market.
• 35:10
  SPEAKER_00: And we believe there's a huge opportunity for us in the middle market.
• 35:14
  SPEAKER_00: That's the strategy.
• 35:15
  SPEAKER_00: Now, the next question is how can AI help us do that?
• 35:19
  SPEAKER_00: Now, if AI can actually help us do that, and I believe in this particular example it'd be fantastic, tool to help us move there, then that's the initiative that's being driven is move to the middle market.
• 35:32
  SPEAKER_00: AI is a tool to help us get there.
• 35:36
  SPEAKER_00: Now, notice I'm not I'm not saying we have to clean up all the data in the company in the next year to make all the data accessible for all different potential applications.
• 35:47
  SPEAKER_00: I'm saying we're choosing one initiative, and that's move to the middle market.
• 35:52
  SPEAKER_00: So I've I've shrunken down um it's it's it's not an experiment, it's not really a test, but I've shrunken down the focus.
• 36:01
  SPEAKER_00: Okay, because uh we it needs to be just the right size.
• 36:05
  SPEAKER_00: There's a sizing question of what are we going to invest in time and energy and training and change management to make something happen.
• 36:12
  SPEAKER_00: That's what I would call a bite-size initiative.
• 36:15
  SPEAKER_00: So that's one place to go.
• 36:17
  SPEAKER_02: Um, the next is we want to think about narrow the scope, narrow the scope, get the right baselines, experiment, see what you can impact in terms of those metrics.
• 36:28
  SPEAKER_00: Exactly.
• 36:29
  SPEAKER_00: Exactly.
• 36:30
  SPEAKER_00: Um, and it's in the context of all of our data needs to be accessible and managed correctly, but we're not tackling that elephant at the moment.
• 36:40
  SPEAKER_00: We might be just taking a bite off the elephant to get this one initiative going into the mid market done.
• 36:46
  SPEAKER_00: The next question is uh build or buy.
• 36:50
  SPEAKER_00: Do I try to build this internally or do I buy it?
• 36:53
  SPEAKER_00: Well, here's here's the the reality is that in the world of AI, uh, most companies, even Fortune 500 companies, can't afford to hire the AI people they need.
• 37:06
  SPEAKER_00: First of all, there are very few of them.
• 37:09
  SPEAKER_00: And and secondly, you're competing with Google and Amazon and Apple and Meta and Microsoft for those people, and that's very hard to get them.
• 37:18
  SPEAKER_00: Now you can get some, but I'm saying is you can't build an entire team of highly talented AI people very easily.
• 37:26
  SPEAKER_00: Rarely do you see that maybe in hedge funds or or the largest banks, you'll you're gonna see that, but but that's because they can pay.
• 37:34
  SPEAKER_00: So, what I recommend companies do, I'm talking about fairly large companies now, is they build a center of excellence of AI.
• 37:42
  SPEAKER_00: They have talent who can run that center of excellence.
• 37:45
  SPEAKER_00: And then they're gonna rely on two parties.
• 37:48
  SPEAKER_00: One is external consultants who who might think more in the about themselves in the system integration world and have some talent and experience in their industry who can help them solve that data problem.
• 38:04
  SPEAKER_00: And the second constituent that's going to help them is vendors.
• 38:08
  SPEAKER_00: So vendors who are incorporating AI into their products and bringing those vendors in, again, in the context of help us enter the mid-market.
• 38:20
  SPEAKER_02: Right, right.
• 38:21
  SPEAKER_02: And that's really interesting.
• 38:22
  SPEAKER_02: I think you know, you you make a really good point.
• 38:25
  SPEAKER_02: And it it really is dependent upon where they're trying to create that value or competitive advantage, because it's less about building the foundational elements of the deep learning models and more about applying them to the problem.
• 38:42
  SPEAKER_02: Right.
• 38:43
  SPEAKER_02: So, you know, many times organizations have challenges going from pilot or POC to production, not because of the uh the uh AI model, the machine learning model, but because of their environment and because of their data and because of their processes.
• 38:56
  SPEAKER_02: So if they get that right, it's much easier to leverage a vendor offering who has already done the heavy lifting and yes, use some consulting uh resources and build that center of excellence that understands how to operationalize uh that AI.
• 39:12
  SPEAKER_02: So I totally agree with you.
• 39:13
  SPEAKER_02: I think it's a really great point.
• 39:15
  SPEAKER_02: And it's really about understanding your business process and where you want to have that intervention, right?
• 39:21
  SPEAKER_02: Like you say, you can't automate a mess, you can't automate what you don't understand.
• 39:25
  SPEAKER_02: So understanding that process and then having that very precise intervention and then looking for the tools and technologies and uh vendor applications that will help you with that intervention.
• 39:37
  SPEAKER_02: So that's a really great point.
• 39:39
  SPEAKER_00: That's right.
• 39:39
  SPEAKER_00: And there's another element to this, um, which is change management.
• 39:45
  SPEAKER_00: And I want to talk about that for a moment.
• 39:46
  SPEAKER_00: True.
• 39:47
  SPEAKER_00: So, in our example that I'm using about moving from the enterprise to the mid-market, um, we have to be very careful to choose something that we can succeed at.
• 39:58
  SPEAKER_00: Because we're gonna move beyond a proof of concept here.
• 40:00
  SPEAKER_00: We're actually going to roll out this strategy.
• 40:04
  SPEAKER_00: And it's important to be able to show some success as it relates to what we're doing with AI because the rest of the organization is watching.
• 40:15
  SPEAKER_00: Right.
• 40:16
  SPEAKER_00: And if this if this fails in in the context that we can't get AI to work correctly, uh then um that's gonna slow down everything else when we try to roll it out elsewhere.
• 40:28
  SPEAKER_00: And so the most important part, well, there's a couple of important parts.
• 40:31
  SPEAKER_00: One, it has to have the support from the most senior management.
• 40:35
  SPEAKER_00: That's how change management is is gonna uh be rolled out.
• 40:40
  SPEAKER_00: Uh, but secondly, we need to manage expectations.
• 40:43
  SPEAKER_04: Right.
• 40:44
  unknown: Yeah.
• 40:44
  SPEAKER_00: Right.
• 40:44
  SPEAKER_00: So to your your point earlier, if we if we enter into this effort and we discover we have a terrible mess in several areas related to our data that we need to clean up, and that's gonna take a long time, then we need to be very clear about that.
• 40:59
  SPEAKER_00: And it has nothing to do with AI in this case, right?
• 41:02
  SPEAKER_00: It has to do with the way we've managed data and governed data.
• 41:06
  SPEAKER_02: It's it's the IA, you know.
• 41:08
  SPEAKER_02: I I code years ago, there's no AI without IA.
• 41:12
  SPEAKER_02: There's no inform, there's no artificial intelligence without information architecture in that's right.
• 41:17
  SPEAKER_02: Kind of that that foundational uh uh piece of that, yeah.
• 41:20
  SPEAKER_00: So in my example, if I'm the CEO, I might look at that and say, oh wow, uh Seth just told us that uh this is gonna take a year longer than I thought it was.
• 41:30
  SPEAKER_00: Why don't we put that on the back burner and find something that doesn't have some of these fundamental problems?
• 41:37
  SPEAKER_00: We're gonna keep working on it, but we're gonna lower its profile.
• 41:40
  SPEAKER_00: We're gonna say it's a long-term project.
• 41:42
  SPEAKER_00: Let's go find another place where we can find a near-term success.
• 41:46
  SPEAKER_02: Right, right.
• 41:47
  SPEAKER_02: And the quick wins are very important.
• 41:49
  SPEAKER_02: So whatever it is that you're doing when you're building a transformation roadmap, and we just do this frequently, we'll have three-year roadmaps with organizations and say, you know, this is a long-term initiative, but along the way, here are your wins, right?
• 42:05
  SPEAKER_02: Here are your successes, right?
• 42:06
  SPEAKER_02: Because you can't have, you know, something we can't promise something three years down the road and not show any progress, right?
• 42:12
  SPEAKER_03: Yeah, that's right.
• 42:13
  SPEAKER_02: Show you know quarterly uh uh uh wins that are going to show how the organization can improve the work today, the process today, you know, partially solve the problem or related problem.
• 42:26
  SPEAKER_02: Uh, I also like to talk about the excuse case, right?
• 42:29
  SPEAKER_02: The use case that's the excuse to do something bigger, right?
• 42:33
  SPEAKER_02: Something really you can really measure and you know there's a baseline, you know you can hit that, but while you're doing that, you know, you're actually enabling other types of initiatives.
• 42:43
  SPEAKER_02: We also did something very similar to what you're talking about with uh a current client.
• 42:47
  SPEAKER_02: We'd be going to be doing a webinar in a couple of weeks uh for Silicon Labs.
• 42:51
  SPEAKER_02: They did a content management, content knowledge management initiative that started with a very narrow focus on their training content.
• 42:58
  SPEAKER_02: And that was precisely to have that win, right?
• 43:02
  SPEAKER_02: Keep that narrow focus, have that win.
• 43:05
  SPEAKER_02: It was a huge success.
• 43:06
  SPEAKER_02: And then the rest of the organization going, hmm, what did you do over there?
• 43:10
  SPEAKER_02: Wow, that looks pretty cool.
• 43:11
  SPEAKER_02: Can I have some of that?
• 43:12
  SPEAKER_02: Right.
• 43:12
  SPEAKER_02: And now it's going from one part of the organization to the next because they saw that that was a success, that that was a win.
• 43:20
  SPEAKER_02: And that inspired another part of the organization to say, let's leverage those lessons learned and let's start to deploy this elsewhere.
• 43:27
  SPEAKER_02: So that's another way to think about it.
• 43:28
  SPEAKER_02: You know, there can be a top-down mandate to do things, but then there can be a bottom-up, you know, uh execution.
• 43:35
  SPEAKER_02: And for large global enterprises, many times that's a process.
• 43:38
  SPEAKER_02: It's get those wins at that functional level, department level, business unit level, and then shampoo, rinse, repeat.
• 43:45
  SPEAKER_02: You know, a lot of times for large enterprises with, you know, you know, hundreds of thousands or millions of products, you know, across multiple divisions, that's kind of the way to do it, right?
• 43:55
  SPEAKER_02: You you take you you eat the elephant one you know bite at a time, like.
• 44:00
  SPEAKER_02: And then you show that you're able to do this, and then that builds on itself, and that creates that flywheel and that uh that virtuous cycle of success.
• 44:10
  SPEAKER_02: Yeah.
• 44:11
  SPEAKER_02: That's I I mean, I hear of companies spending you know tens of billions of dollars on AI initiatives, and it's like, how can you spend billions of dollars like at that scale?
• 44:22
  SPEAKER_02: It's just incredible.
• 44:23
  SPEAKER_02: Where you know you can spend several million dollars at a smaller scale and show really great improvements and then start building those capabilities.
• 44:32
  SPEAKER_02: So um, so the next time you run into a company spending billions of dollars on on AI, just let uh let us know.
• 44:40
  SPEAKER_02: So um I know we're coming up to the top of our our uh time here.
• 44:44
  SPEAKER_02: And I wanted to uh just give you an opportunity to kind of uh finish with your your uh final thoughts.
• 44:50
  SPEAKER_02: And you know, one of the things that I I like to ask people is if you could go back in time and you know, when you you you just graduated from college or from graduate school and you could give yourself some advice, what what what might you say to yourself back then?
• 45:03
  SPEAKER_02: What kind of advice would you give yourself?
• 45:06
  SPEAKER_00: I I I would say, look, um be more innovative, don't be afraid of making mistakes, um, and and grab on to the changes you see happening.
• 45:22
  SPEAKER_00: So one of my one of my favorite quotes is from Neil Stevenson, uh a science fiction author.
• 45:29
  SPEAKER_00: Yeah, he said, the future is here, it's just unevenly distributed.
• 45:37
  SPEAKER_00: Right, right.
• 45:39
  SPEAKER_00: And so what I think about when I hear that quote is I can lean in to learning the latest things that are changing the world, or I can wait to see what happens.
• 45:55
  SPEAKER_00: But in particular, right now, at this moment in time, we're seeing some massive change that we don't fully understand the implications.
• 46:05
  SPEAKER_00: The example I use are many examples that have been used, but I think this is like 1994, when Netscape, this little company called Netscape, popped up on the radar screen and said, we're taking the internet from what used to be just a communication mechanism in academia, we're putting in consumer layer on top of it.
• 46:27
  SPEAKER_00: And then the world exploded with everything that we live in today.
• 46:33
  SPEAKER_00: Right.
• 46:34
  SPEAKER_00: This moment or the moment in uh November when ChatGPT came out is a similar moment.
• 46:41
  SPEAKER_00: And I'm not even convinced ChatGPT is going to be the answer or or the most important thing.
• 46:47
  SPEAKER_00: What's important is now we've put a consumer layer on top of AI.
• 46:53
  SPEAKER_00: And now every single person on the planet can be impacted in a positive way.
• 46:59
  SPEAKER_00: Now there will be negative impacts, no question about that, because this is just a tool and and and a tool can be used for good or bad.
• 47:07
  SPEAKER_00: But I I want to uh I want to tell you a story.
• 47:11
  SPEAKER_00: So one of my CEOs uh sold his company a little while ago, uh, successful exit.
• 47:19
  SPEAKER_00: And now I'm still coaching the CEO, and and that CEO is um advising the company he sold on technical issues.
• 47:27
  SPEAKER_00: It turns out he's very technical.
• 47:30
  SPEAKER_00: So the other day the CEO was was advising the company he sold on technical issues, and said to them, Hey, would you like me to write an algorithm to solve that problem?
• 47:44
  SPEAKER_00: Because he knows more than anybody about the code.
• 47:49
  SPEAKER_00: And he got a note back from a junior engineer that said, Yeah, no, uh, we don't need you to write, we don't need you to write an algorithm.
• 47:57
  SPEAKER_00: I asked ChatGPT and it did it for me.
• 48:00
  SPEAKER_00: And my CEO was devastated, said he said, I've always been known as the person who could solve problems no one else could solve.
• 48:12
  SPEAKER_00: Wow.
• 48:12
  SPEAKER_00: And now suddenly I'm not that person anymore.
• 48:16
  SPEAKER_02: Interesting.
• 48:17
  SPEAKER_02: Yeah, wow.
• 48:18
  SPEAKER_00: And and and and that is an existential crisis.
• 48:21
  SPEAKER_00: It is, it is, it is an existential crisis.
• 48:24
  SPEAKER_00: And I don't have the the perfect answer, but uh my one answer I I'm encouraging everybody, whether you're a CEO or a board member or um uh a first line worker, is to lean in to what's happening now, become good at all the tools that are being made available to us because that's going to create opportunity for you.
• 48:46
  SPEAKER_00: People who don't lean in are going to be losing a lot of opportunity and maybe certainly tasks in their job, maybe satisfaction in their jobs, and for some of them, jobs themselves.
• 48:59
  SPEAKER_02: Absolutely.
• 49:00
  SPEAKER_02: Well, I really want to thank you.
• 49:02
  SPEAKER_02: Uh, we've been with uh Glenn Gao, who is known as the uh AI guy.
• 49:07
  SPEAKER_02: And uh Glenn, where can people find you?
• 49:09
  SPEAKER_02: We're gonna have some of your contact information in the show notes, but where can people find you?
• 49:14
  SPEAKER_00: Great.
• 49:14
  SPEAKER_00: It's easy.
• 49:15
  SPEAKER_00: You just need to know how to spell my name, and that's the name of my website.
• 49:18
  SPEAKER_00: It's Glenn G-L-E-N-N-G-O-W.com.
• 49:23
  SPEAKER_00: Wonderful.
• 49:24
  SPEAKER_00: Thank you to us.
• 49:25
  SPEAKER_02: Excuse me, thank you so much.
• 49:27
  SPEAKER_02: That's been wonderful.
• 49:28
  SPEAKER_02: So thank you again.
• 49:29
  SPEAKER_02: Thank you to our audience.
• 49:31
  SPEAKER_02: Uh, this has been another episode of the Early AI Podcast, and we will see you all next time.
• 49:37
  SPEAKER_02: Thank you again.
• 49:39
  SPEAKER_01: Thank you for joining us on another deep dive into AI innovation.
• 49:43
  SPEAKER_01: Join us next time when we introduce another expert and discuss how to maximize AI in your world.
• 49:48
  SPEAKER_01: Early AI Podcast is sponsored by Early Information Science, CMS Wire, and Marketing AI Institute.
• 49:55
  SPEAKER_01: To learn more about us, visit Early.com.
• 49:58
  SPEAKER_01: At a C A R L A Y.com.
• 50:02
  SPEAKER_01: See you next time.