Latest / The Adults In The AI Room / Tom Davenport on Competing on AI - Episode #15

Transcript

Tom Davenport on Competing on AI - Episode #15 · The Adults In The AI Room

• 0:04
  SPEAKER_00: Welcome to University of the Early AI Podcast.
• 0:07
  SPEAKER_00: Here we talk with thought leaders and practitioners on what's possible and practical with AI and the enterprise as we move from the current state into a future of cognitive assistance and embedded AI in all aspects of our personal and professional lives.
• 0:22
  SPEAKER_00: Thanks so much for listening.
• 0:23
  SPEAKER_00: If this is your first time, we invite you to subscribe wherever you listen to podcasts.
• 0:28
  SPEAKER_00: And now here are your hosts, Seth Early and Chris Featherstone.
• 0:36
  SPEAKER_01: Well, welcome to today's podcast.
• 0:38
  SPEAKER_04: I'm Seth Erwey.
• 0:40
  SPEAKER_04: And I'm Chris Featherstone.
• 0:41
  SPEAKER_04: It's good to be with you again.
• 0:43
  SPEAKER_01: Yeah, what's going on, Chris?
• 0:44
  SPEAKER_01: What's what's uh new and exciting in your world?
• 0:47
  SPEAKER_04: I think last time we spoke, I was living in Salt Lake and now we are relocated.
• 0:52
  SPEAKER_04: Now I live in Montana.
• 0:53
  SPEAKER_01: Oh, Montana.
• 0:54
  SPEAKER_01: Okay.
• 0:55
  SPEAKER_01: Any any uh dental floss in your world?
• 0:58
  SPEAKER_04: No, just just Frank Zappa.
• 1:00
  SPEAKER_01: No, no, no reference to Frank Zappa.
• 1:02
  SPEAKER_01: Gonna Montana soon gonna be a dental floss tycoon.
• 1:06
  SPEAKER_01: No, oh man.
• 1:07
  SPEAKER_04: I wish I wish I could be.
• 1:08
  SPEAKER_04: I wish I could be, honestly.
• 1:10
  SPEAKER_04: But no, but we uh we live near Flathead Lake and it's awesome, it's beautiful.
• 1:15
  SPEAKER_01: It's great.
• 1:15
  SPEAKER_01: You know, it's funny.
• 1:16
  SPEAKER_01: I I used that the other day when somebody was from Montana and I got crickets as well there.
• 1:20
  SPEAKER_01: So I'll stop using that Frank Zappa, uh, arcaneist Frank Zappa reference, but it's from uh Weasels Rip My Flesh uh album.
• 1:28
  SPEAKER_01: So I don't remember what year, but anyway.
• 1:30
  SPEAKER_01: Well, let's jump in and let's uh get let's get started.
• 1:34
  SPEAKER_01: So today's guest is the author of many, many books.
• 1:38
  SPEAKER_01: Uh, one recent one is the AI Advantage How to Put Artificial Intelligence, the Artificial Intelligence Revolution to Work.
• 1:44
  SPEAKER_01: He also wrote Only Humans Need to Ply, the winners and losers in the age of smart machines.
• 1:48
  SPEAKER_01: He's written uh Competing on Analytics, and I think about 20 other books.
• 1:53
  SPEAKER_01: And that's not an exaggeration.
• 1:54
  SPEAKER_01: I think uh 20 plus uh additional books that uh way back when uh competing was it uh the knowledge, the knowledge management, the knowledge advantage.
• 2:03
  SPEAKER_01: That's right.
• 2:03
  SPEAKER_01: That was back in the olden days.
• 2:05
  SPEAKER_01: Working knowledge.
• 2:06
  SPEAKER_01: Working knowledge, sorry.
• 2:09
  SPEAKER_01: I'm conflating the new with the old.
• 2:10
  SPEAKER_01: That's right, working knowledge.
• 2:12
  SPEAKER_01: Um, and uh of course, his articles have appeared in HBR, Forbes, Loan Management Review, the Wall Street Journal, other publications.
• 2:20
  SPEAKER_01: He is a world-renowned thought leader on the topics of artificial intelligence, analytics, information and knowledge management, process management, and of course, if he weren't busy enough, he's also the president's distinguished professor of information technology management at Babson College and a fellow of the MIT Center for Digital Business.
• 2:38
  SPEAKER_01: Welcome to the show, Tom Davenport.
• 2:40
  SPEAKER_01: Glad to be here.
• 2:41
  SPEAKER_01: Thanks for having me.
• 2:42
  SPEAKER_01: So uh so it's interesting.
• 2:45
  SPEAKER_01: Um uh you have written uh a lot on all of these really interesting topics, and uh I know you have some other stuff in process, but I do want to plug your uh your inclusion in this issue of uh how uh the HBR uh special topic issue that came out, which is a best of archive edition.
• 3:03
  SPEAKER_01: And uh you wrote about topics near and dear to my heart around uh processes and knowledge management and information architecture.
• 3:11
  SPEAKER_01: So that's terrific.
• 3:13
  SPEAKER_02: Yeah, I was surprised my wife um came across that in the supermarket in Whole Foods and brought it home one day.
• 3:21
  SPEAKER_02: I didn't even know that it existed.
• 3:23
  SPEAKER_01: That's funny.
• 3:24
  SPEAKER_01: That's great.
• 3:25
  SPEAKER_01: And and I'll plug myself, my article's in here too.
• 3:27
  SPEAKER_04: So well, you know, I wanna I want to ask the most important question is why you got a Texas shirt on, right?
• 3:34
  SPEAKER_04: Aren't you?
• 3:35
  SPEAKER_02: Well, I um despite um my virtual background from Oxford, where I'm currently a visiting professor, I'm actually in Austin, and um I was a professor here back in the mid to late 90s, and I realized I didn't have any Austin um shirts, and so I yesterday bought this Longhorn shirt to um reflect my uh background um at the 40 acres, as they call the University Um of Texas campus.
• 4:12
  SPEAKER_04: So that's so you're you're basically just showing that you bleed orange and you've got your uh you know, you got your heart on your on your lapel now.
• 4:19
  SPEAKER_02: I believe burnt, I bleed burnt orange, yes.
• 4:22
  SPEAKER_04: Burnt orange, yeah, yeah.
• 4:24
  SPEAKER_04: That's awesome.
• 4:26
  SPEAKER_01: So Tom, we've known each other for a number of years, and you know, but I never really heard your kind of origin story, how you got into this space.
• 4:35
  SPEAKER_01: Like what made you what what brought you to this world and how did that how did that arrive arise?
• 4:41
  SPEAKER_02: Uh I think you're a young boy thinking about knowledge and I think it had something to do with my parents, how I was brought into this world, but I I won't go back to that.
• 4:52
  SPEAKER_02: So um, yeah, I you know, I was trained as a sociologist, and I um worked for a couple of years as an academic sociologist, and I said, Yeah, this is a little boring, you know, writing papers for a few other sociologists in the world, if I'm lucky.
• 5:12
  SPEAKER_02: And so I decided, well, I want to kind of go into business or consulting was particularly interesting to me.
• 5:21
  SPEAKER_02: And um eventually um I got it was a little bit of a struggle, but I got a job as a PhD in sociology in a consulting firm, and it was an IT.
• 5:32
  SPEAKER_02: I've worked a lot with statistical computing, and I was quite interested in IT.
• 5:37
  SPEAKER_02: And um, I got a job at an IT strategy boutique called Index, and then I've immediately gravitated toward the research side of stuff.
• 5:47
  SPEAKER_02: We had a multi-client program with a bunch of different sponsors and did work on a lot of topics with the late Michael Hammer.
• 5:54
  SPEAKER_02: And um, so then I've ever since I've sort of gone back and forth between business schools and consulting firms, and always focused on, I guess, if there were such a thing as the sociology of business information and technology, that would be my domain, but there is no such topic.
• 6:14
  SPEAKER_02: So um well, it's kind of a late, yeah.
• 6:16
  SPEAKER_02: Go ahead, sorry.
• 6:17
  SPEAKER_02: I'm always interested in you know how people and organizations use these tools effectively.
• 6:24
  SPEAKER_01: Right.
• 6:24
  SPEAKER_01: I I think it that's really the crux of things these days.
• 6:27
  SPEAKER_01: It's you know, we're we're looking at artificial intelligence and machine learning and cognitive applications, but it's really about uh humans and how people are interacting with technology and and reducing the cognitive load on human, right?
• 6:40
  SPEAKER_01: So I think it's all about you know how people are working, how they're interacting, how they're collaborating, and then how these tools kind of help be more tools in the toolkit to serve humans uh in all of that.
• 6:53
  SPEAKER_02: I agree, except that some I've done some recent research that's suggesting that sometimes AI increases the cognitive load on humans because it does the easy part that humans used to do and leaves the hard part.
• 7:06
  SPEAKER_02: But that's all right.
• 7:08
  SPEAKER_01: That's a good that's a good point, right?
• 7:09
  SPEAKER_01: Because again, you're getting rid of the routine and you know the unambiguous use cases and the thing that the things that are kind of the drudgery, but now here's the real problem.
• 7:19
  SPEAKER_01: Here's the hard problem.
• 7:20
  SPEAKER_01: So, you know, you've you're you've done some really great research and and and obviously you're you've been following this field.
• 7:27
  SPEAKER_01: You and I have been talking about many of these things for a number of years.
• 7:30
  SPEAKER_01: And uh I you wrote the forward to my my book, which was very nice of you.
• 7:35
  SPEAKER_01: And had a little bit of a of a of a debate or discussion about uh IA and and cognitive and machine learning and so and all of that.
• 7:44
  SPEAKER_01: Uh, but what tell me what your thoughts are about what's most important for organizations today when they're trying to operationalize these things?
• 7:51
  SPEAKER_01: We've gone through that curve of, you know, this is so new and exciting and and mysterious.
• 7:57
  SPEAKER_01: And I still think there's a bit of that.
• 7:58
  SPEAKER_01: But what's important to organizations trying to operationalize these things?
• 8:03
  SPEAKER_02: Well, yeah, I mean, that's a big problem.
• 8:06
  SPEAKER_02: Uh, getting a little better, but I think for the last couple of years, I was quite focused on this issue of how do we how do we get return on AI, how do we get beyond this, you know, experimentation stage and really get some production deployments in place because we don't get any value unless we actually put something into production deployment.
• 8:31
  SPEAKER_02: And starting to improve, I wrote a piece um uh a few weeks ago on you know, some companies seem to be making some serious money with AI finally, but for still, I think the number of um AI models in a machine learning sense that actually get deployed is quite low.
• 8:54
  SPEAKER_02: I don't know exactly what the overall percentage is, and I don't know what it should be.
• 8:58
  SPEAKER_02: It shouldn't be 100%, but it should be one venture beat survey said it was 13, which seems a little crazy low.
• 9:06
  SPEAKER_02: So how do we put these things in place?
• 9:08
  SPEAKER_02: And there are a whole variety of ways to do that.
• 9:11
  SPEAKER_02: I think um you have to think about it from the beginning as you know, something you're going to deploy, not just play around with, and then you have to work on the people side, as we were discussing.
• 9:22
  SPEAKER_02: You have to invest, it's a lot more money to um to build a deployed AI system than it is um a proof of concept, because you know, you've got to integrate it with your existing systems and get lots of data and um build the APIs so you know get a good answer quickly and all that sort of thing.
• 9:41
  SPEAKER_02: So um we just need to do a lot more of that kind of activity if we're gonna be successful with it economically and you know in terms of productivity and so on.
• 9:53
  SPEAKER_04: So I'd love to get a sense too, Tom.
• 9:56
  SPEAKER_04: So given where we're at, you know, all the you know, the articles and things, you know, are a lot about which we talk about a lot in this podcast too, is the preparation, right?
• 10:04
  SPEAKER_04: Preparation of the data, preparation of the different roles, the right investment to make sure that you're doing building the foundations right to actually drive the outcome that you want to drive.
• 10:14
  SPEAKER_04: A lot of people don't even know what what the initial outcome should be, let alone you know how it evolves, right?
• 10:20
  SPEAKER_04: Because it's we find there there's a lot of organizations think that this is a one-hit wonder and it's gonna solve all problems the first time they actually set a model and go, right?
• 10:30
  SPEAKER_04: And to your point, it's great in the first you know, pieces to find the easy stuff, and then we leave a lot of the next, we'll say, incremental work to a lot of these folks.
• 10:42
  SPEAKER_04: But given you know, we're at in the pandemic, you know, I'd love to get your take too on when we have now um a lot of different um, whether it be political things going on in the world or whether it be, you know, the you know, the the pandemic type of things, how is what we're doing now with AI gonna really help or hinder, or what should we be focused on, in your opinion?
• 11:08
  SPEAKER_02: Um, well, you know, I I did a thing earlier this week on people analytics, which is one area that I sort of um play around in occasionally.
• 11:19
  SPEAKER_02: And uh you would I think this is a really uh exciting time, as I said at that conference, to be in that space because we're really rediscovering how work should be done.
• 11:32
  SPEAKER_02: I mean, you just moved to Montana, um, where I'm assuming there's not a big um Salesforce presence generally.
• 11:42
  SPEAKER_02: Um uh, you know, what's all that gonna mean for how we work effectively and um how we collaborate and how we innovate and so on?
• 11:49
  SPEAKER_02: And fortunately, you know, we have all these tools, analytics and AI, and lots more data in um various transactional systems that we use um to help us answer those kinds of questions.
• 12:02
  SPEAKER_02: So I think it's a it's a great time.
• 12:05
  SPEAKER_02: You know, I'm not sure how AI can prevent uh, you know, invasions to companies like Ukraine and so on, but it can certainly help us figure out how to effectively do our work, at least.
• 12:21
  SPEAKER_02: And I think it will also, you know, we we have some really interesting dynamics going on in our economy right now.
• 12:29
  SPEAKER_02: Um I think it can help us a lot in that regard, too.
• 12:33
  SPEAKER_02: We start, you know, analyzing um a lot of external macroeconomic data to try to make sense of what seems to be a very unusual period.
• 12:41
  SPEAKER_04: Um I I think your I think your point too, especially with our most expensive yet valuable commodity is the human, you know, factor.
• 12:50
  SPEAKER_04: And you're right.
• 12:52
  SPEAKER_04: It's I mean, I'm on a Starlink system right now, right?
• 12:55
  SPEAKER_04: I'm in an area where, you know, 2,000 feet to, you know, to the west of me is fiber, you know, that's being run in more rural areas, but then they won't run it here because of whatever, you know, whatever their the issue is.
• 13:09
  SPEAKER_04: So yeah, so I'm on I'm on a more rural type of a connectivity system, but it works great.
• 13:15
  SPEAKER_04: So it's allowing me to interact with you guys with very little latency and at the same time do um you know product for high complex enterprise-related software and everything else, where we're connecting up everybody with you know all over the world to actually get together and actually do this.
• 13:34
  SPEAKER_04: However, part of this though, to your point though, is is that we didn't, I'm not working from home.
• 13:41
  SPEAKER_04: I took work home with me, and it's a different psychology than just working from home.
• 13:45
  SPEAKER_04: I've worked from home for 20 something years, right?
• 13:48
  SPEAKER_04: Ever since I kind of started at Microsoft in the 90s and was able to remote work, which is really what it was.
• 13:54
  SPEAKER_04: But now we're we brought work home with us, and there's a an expectation to do the same things.
• 13:59
  SPEAKER_04: However, my water cooler talk is with the dogs and my wife, right?
• 14:04
  SPEAKER_04: And you know, how we take breaks and all that kind of stuff.
• 14:07
  SPEAKER_04: And I think there's definitely some areas of you know, psychology that, you know, when I build the um the human engagement systems that we do that either you know enhance call centers or whatever, this aspect of what we are able to do to provide more information in a very you know specific way is super critical now, more than ever.
• 14:32
  SPEAKER_04: You know, like we also used to do this, you know how how this goes in the in the call center space.
• 14:36
  SPEAKER_04: What's your next next best action?
• 14:39
  SPEAKER_04: Well, what's the context of what I'm talking about?
• 14:41
  SPEAKER_04: Right.
• 14:42
  SPEAKER_04: And do we take that context in?
• 14:43
  SPEAKER_04: Do I look at the conversation from a different lens than just you know leaning on somebody who's maybe not educated in these topics to solve these problems?
• 14:55
  SPEAKER_04: And so there's a lot of this kind of stuff like companies want to keep this high level of loyalty and brand awareness and everything else at the same time.
• 15:02
  SPEAKER_04: Their people are now effectively at home workers, not in an office, you know, type of scenario.
• 15:08
  SPEAKER_04: So, you know, I think that's you know, part and parcel to what you you wrote about is how this is gonna change and how it needs to change.
• 15:16
  SPEAKER_04: But I love the perspective though that you give on we've, you know, I hate that the you know, when you hear people say, well, I've I you know we've done this before and this is what we're gonna do again.
• 15:27
  SPEAKER_04: We've never seen this before and never seen these types of things.
• 15:30
  SPEAKER_04: And you know, so I think it's gonna, like you said, cause a lot of companies to kind of take a step back and say, hmm, how do we approach this?
• 15:38
  SPEAKER_04: Right.
• 15:39
  SPEAKER_01: And to that point, I think that's a great point, Chris.
• 15:41
  SPEAKER_01: And to that point, it's it's like, what are we doing to enable that remote work and that collaboration?
• 15:47
  SPEAKER_01: Because we don't have the same ability to kind of connect with colleagues.
• 15:50
  SPEAKER_01: And what are the big stumbling blocks to leveraging what are people trying to leverage to fill that gap?
• 15:56
  SPEAKER_01: And it seems like you know, Semantic Search or you know, helper box or knowledge retrieval box, or what what else is happening in that kind of human collaboration space, that knowledge space, uh, in your opinion, and what what what's holding organizations back from really being able to capitalize on the tools to support things like remote work and collaboration?
• 16:17
  SPEAKER_02: Yeah, well, you know, I think a lot of that still has to be figured out and invented even.
• 16:24
  SPEAKER_02: I mean, I um this thing I was doing earlier this week, somebody mentioned the importance of tacit knowledge to innovation.
• 16:33
  SPEAKER_02: Um, and I brought up um this work um at MIT a number of years ago by this guy, Tom Allen, uh, who's a professor, did work on um how do the physical surroundings of work affect um innovation.
• 16:52
  SPEAKER_02: And he, you know, for years I've been a big believer in this that you know, he had something called the Allen curve, or some people refer to it as the 25-meter rule, suggesting that if you're more than 25 meters away from someone else, your chances of um exchanging ideas with them are effectively zero.
• 17:15
  SPEAKER_02: Um so how do we translate that kind of um belief or finding into an environment today where we're more than 25 meters away from everybody?
• 17:30
  SPEAKER_02: I mean, Chris is more than 25,000 meters away from everybody.
• 17:35
  SPEAKER_02: Um and I clearly, you know, technology can help, but we don't yet know um under what circumstances it can help.
• 17:45
  SPEAKER_02: I mean, um uh Seth is a big fan of getting a bunch of people around a fire pit to um exchange ideas.
• 17:56
  SPEAKER_02: So what's the technological equivalent of that?
• 17:59
  SPEAKER_02: Um I did a thing uh with Red Hat a couple of weeks ago called a lean coffee, where you um, you know, it's a mechanism for discussing ideas for a while, and then people vote and decide do I go for five minutes more?
• 18:14
  SPEAKER_02: I was so excited when I proposed a topic around AI and that got extended for an extra five minutes.
• 18:19
  SPEAKER_02: Uh vote of confidence for you.
• 18:22
  SPEAKER_01: Yeah.
• 18:22
  SPEAKER_02: You know, we have to figure out what are those new ways of working under those circumstances in almost every area, whether it's innovation or call centers or um consulting or education.
• 18:35
  SPEAKER_02: I mean, I it seems highly unlikely to me that we'll continue to have to bring everybody to a single campus in order to be educated, although you know, we're we're not too far beyond that now.
• 18:49
  SPEAKER_02: Yeah.
• 18:50
  SPEAKER_04: Well, uh, you know, I was I was actually reading that, I read that article that you uh had written, you know, about that exact topic, right?
• 18:57
  SPEAKER_04: Around your lean coffee discussion.
• 18:59
  SPEAKER_04: And it's interesting because, you know, the one thing that that I kind of took away from it too is obviously like you know, like it states there in Georgia's law around, you know, or you know, technology moves fast, organizations move slow.
• 19:15
  SPEAKER_04: Now, is that you know, in your mind indicative that executives move slow and their people are trying to move fast?
• 19:23
  SPEAKER_04: Or is it that, you know, because we also have like organizations don't just move slow.
• 19:29
  SPEAKER_04: I mean, maybe you move as fast as your slowest person or something, but by and large, a lot of it seems to be, you know, almost the Peter principle where you know folks can't rival rise above the level of their own incompetence.
• 19:42
  SPEAKER_04: And is it because you know the organizations can't move faster than the actual leaders that don't understand where to go or what decisions to make?
• 19:52
  SPEAKER_02: Well, there's certainly some of that going on in the um, you know, return to work kind of environment where you have some of these, I would um almost want to call them dinosaurs who say, by God, you know, the only place to really get work done is in the office, and you're gonna come in at least five days a week, you know, maybe more, and and work all day long there.
• 20:16
  SPEAKER_02: Um, no data to support that assertion, no um analytics, no evidence at all.
• 20:22
  SPEAKER_02: And I think people are gonna vote with their feet and say, I don't want to do that, and then I'll see you later.
• 20:30
  SPEAKER_04: That seems to me the great resignation right there.
• 20:33
  SPEAKER_04: You typified the reason why.
• 20:35
  SPEAKER_04: But anyway, one of the reasons why.
• 20:37
  SPEAKER_04: Yeah.
• 20:37
  SPEAKER_02: Well, although it's interesting at this um conference there was some data suggesting that there's not really a great resignation.
• 20:46
  SPEAKER_02: It this was a trend that was happening even before COVID, and it's just continuing.
• 20:51
  SPEAKER_02: So I think people were slowly figuring out, you know, life is too short to spend it all on commuting and airplanes and so on.
• 21:00
  SPEAKER_02: I mean, God knows.
• 21:01
  SPEAKER_02: I I have seven million miles on various airlines.
• 21:06
  SPEAKER_02: I don't know why I did it, except that I I don't mind it so much if I get to sit in first class and you're going to be able to do that.
• 21:15
  SPEAKER_01: Back of the bus, yeah.
• 21:17
  SPEAKER_01: Yeah.
• 21:18
  SPEAKER_01: So one of the questions, you know, that we had uh talked about is is when you start thinking about what's slowing organizations down from really leveraging these technologies, you know, is it is it infrastructure or is it understanding?
• 21:34
  SPEAKER_01: And you know, to Chris's point, the the the executive level, is it is it that the legacy systems are so difficult to evolve and the data and all the infrastructure things that I talk about around architecture, et cetera?
• 21:47
  SPEAKER_01: Or do you think it's more on the cultural side and more executive understanding or misconceptions?
• 21:52
  SPEAKER_01: So maybe you could talk about some of those misconceptions.
• 21:55
  SPEAKER_01: So where's the where's the the uh stumbling block here?
• 21:59
  SPEAKER_02: Yeah, well, I think.
• 22:00
  SPEAKER_02: You know, I think things are changing in that regard, and historically it was um uh I think a more of an issue of um the technology was kind of ahead of what people could do with it, and um people were content to let centralized IT groups tell them, you know, what technology they they needed to do their work.
• 22:25
  SPEAKER_02: But I was um interviewing a guy this morning um at a relatively small company.
• 22:32
  SPEAKER_02: Um, and I said, you know, do you need to really um bring the people?
• 22:37
  SPEAKER_02: He was talking about some great new manufacturing applications with sensor data and put it pulling it all together.
• 22:43
  SPEAKER_02: Um and I said, You you need to educate them.
• 22:47
  SPEAKER_02: And he said, Well, no, they're really leading the charge.
• 22:50
  SPEAKER_02: And the guys in charge of manufacturing is the primary advocate of this stuff in the organization, and we're just trying to keep up.
• 22:58
  SPEAKER_02: And I think as as we get this generational change of people who've had technology all their lives that is available in real time and all the time, and um uh can give you answers to basically any any question um that centralized IT people and old technologies will sort of become the biggest problem that we that we have.
• 23:26
  SPEAKER_02: And even you know, I was um I'd done some work in the past with a guy who lives in the Boston area named Gary Loveman.
• 23:33
  SPEAKER_02: He was the he was an interesting guy, Harvard professor, um, MIT trained economist who um left Harvard and went to be CEO eventually of Caesars, the biggest gaming company.
• 23:47
  SPEAKER_02: And um he left there.
• 23:50
  SPEAKER_02: Probably didn't help that they went bankrupt.
• 23:52
  SPEAKER_02: Uh that's another another issue.
• 23:55
  SPEAKER_02: Part of them went bankrupt.
• 23:56
  SPEAKER_02: Um, but he went to a large health care, health insurance company, um, and they were gonna try to do some interesting work in consumer health.
• 24:06
  SPEAKER_02: And he said, you know, I propose um that if we're gonna do that, we need to add to our customer or member database um people's email and their um cell phone number because you know, we're gonna be sending them out reminders of, you know, get your exercise or um eat better tonight or whatever.
• 24:27
  SPEAKER_02: And he said the cost of doing that was estimated at$30 million.
• 24:33
  SPEAKER_02: And so eventually he said, I can't take this anymore.
• 24:36
  SPEAKER_02: I'm gonna create a startup to do it because startups don't have all that technical debt and they can build it from the beginning that way.
• 24:44
  SPEAKER_02: So I think that's one of the reasons I find it really interesting to have to change a large organization in that regard.
• 24:52
  SPEAKER_02: I'm talking to another big health insurance company that really has some fantastic aspirations for what they want to do with analytics and AI and um, you know, increasing the health of the improving the health of their members, but um, it makes her some really interesting organizational change problems.
• 25:09
  SPEAKER_03: Yeah.
• 25:10
  SPEAKER_04: I was actually um, I love the the clinical side and the fact that you've got clinical outcomes that um have been on a parallel track next to, let's say, mental health outcomes, and they haven't been blending that data in the background.
• 25:26
  SPEAKER_04: And so therefore, they don't really know the clinical outcomes if they truly influence the medical outcomes.
• 25:32
  SPEAKER_04: I mean, they're starting to become more of this, you know, cross uh pollinizing and stuff.
• 25:37
  SPEAKER_04: But, you know, I worked with some clinical or some uh neuropsychologists around this exact topic, like how do we blend the data?
• 25:44
  SPEAKER_04: And the problem was, as you can imagine, was getting access to the data to actually then map the outcomes and looking for the pattern so that they could actually, you know, know that in a setting where they have, you know, somebody who's let's say a traumatic brain injury or post-traumatic stress or something, they can actually have a clin, you know, have a panel of folks interviewing them, but at the same time doing some of the more clinical related stuff in a you know a setting like this, and um, whether it be EMDR or something like that, and then trying to grab those outcomes to desensitize all those things and then mapping them.
• 26:19
  SPEAKER_04: It was you know, just fascinating.
• 26:21
  SPEAKER_04: But part of that in my mind too is you know, this is the this is the antithesis for why we need you know machine learning models to actually look for those patterns together in order to create at least some kind of an outcome, because there's no way you know it'll it'll take decades, you know, for you know, to put let's say a bunch of analysts on that to try to come up with these things and we need machines just pounding on all the time.
• 26:47
  SPEAKER_01: But anyway, and the privacy the privacy issue on the data is really uh very very significant.
• 26:55
  SPEAKER_01: Of course.
• 26:56
  SPEAKER_04: I I think there's also this notion too of teasing out the the you know the velocity that humans can actually work is as good as let's say the systems that enable them to do it, but also we get into the scenario too where we've got digital immigrants and digital natives.
• 27:13
  SPEAKER_04: And what I see a lot going on is that the digital immigrants have a lot of the incumbent business experience and just experience at life in general, and they see a lot of the maybe the higher level um patterns of how to you know get things, you know, get things done.
• 27:30
  SPEAKER_04: And at the same time, the digital natives are quick to adopt, quick to adopt, quick to adopt.
• 27:36
  SPEAKER_04: And yet I see a big gap in terms of trust between the two, right?
• 27:41
  SPEAKER_04: These people, the digital natives, don't trust the digital immigrants to know how to move faster, and that yet the digital immigrants, and they know that they don't have the experience to move to do these things, you know, but we also see that opposite effect going on where the digital immigrants don't trust the digital natives to be able to get things done, you know, in the quote common, you know, sense of business type of thing, too.
• 28:04
  SPEAKER_04: So your buddy actually creating a new startup makes a ton of sense.
• 28:06
  SPEAKER_04: However, you know, how much trust does he give to those folks that are coming in to say, no, let's do it faster, better, cheaper, stronger, whatever it is, because we can do it this way, right?
• 28:15
  SPEAKER_04: I'd love to get your take on that.
• 28:17
  SPEAKER_02: Yeah, well, I think that all of us, um, if we want to continue to be, you know, productive and effective at our work, really have to push ourselves to um explore and adopt some of those technologies.
• 28:30
  SPEAKER_02: I mean, I could do better at it.
• 28:32
  SPEAKER_02: I was talking to a friend of mine um recently.
• 28:35
  SPEAKER_02: We were laughing about the fact that um we still create things and send around Microsoft Word documents.
• 28:45
  SPEAKER_02: And I I don't know, I I find it comforting sort of to have that.
• 28:50
  SPEAKER_02: And um now I send them to a lot of companies and they immediately convert them into Google Docs and I say, where'd my file go?
• 28:57
  SPEAKER_02: You know, but we we have to we have to get over that, I think.
• 29:02
  SPEAKER_02: Uh um we, you know, we all have these areas where we're comfortable.
• 29:08
  SPEAKER_02: I'm sure that will be true of the digital natives as well eventually, but we've got to press ourselves to try try the new stuff and take some time to you know invest in learning it and using it.
• 29:21
  SPEAKER_01: I I get ridiculed in my dentist's office for bringing a newspaper.
• 29:25
  SPEAKER_01: A newspaper, my I haven't seen one of those in the same the the first the hygienist did it, and then the following week that the dentist did it.
• 29:35
  SPEAKER_01: I don't know if they're in cahoots, they were both flabbergasted that I would actually bring a physical newspaper.
• 29:40
  SPEAKER_01: I like it.
• 29:40
  SPEAKER_04: Well as my kids say, you know, that's you know, dad, that's so cool because it's vintage.
• 29:47
  SPEAKER_04: Right?
• 29:48
  SPEAKER_04: That's vintage.
• 29:50
  SPEAKER_02: It's like it's like um, you know, vinyl records.
• 29:53
  SPEAKER_02: They're they're gonna come, those newspapers are coming back any day now.
• 29:57
  SPEAKER_02: Well, K.
• 30:00
  SPEAKER_01: I I I have a funny story.
• 30:01
  SPEAKER_01: I was trying to pay using my Apple Watch the other day, and I couldn't get it to work.
• 30:05
  SPEAKER_01: And I said, Looks like I'll have to go back to the old-fashioned pay with my iPhone.
• 30:12
  SPEAKER_04: That's funny.
• 30:14
  SPEAKER_04: You know, we're we're also on these edge, too, of you know, so obviously a lot of this, you know, information, and we we we we do talk a lot on this or um on this podcast about just the fundamentals, right?
• 30:26
  SPEAKER_04: Get your data in a in a you know, in a sense so you can actually understand the information.
• 30:31
  SPEAKER_04: And understanding the information is going to really unlock a lot of the questions that you might have in terms of you know putting it against a model and trying to drive, let's say, the patterns of outcomes.
• 30:40
  SPEAKER_04: What you know, and then you know, and that's just one aspect, right?
• 30:43
  SPEAKER_04: We have AI, but then we're also diving into aspects too, where we're doing augmented reality, artificial reality.
• 30:51
  SPEAKER_04: We've got um, you know, just in terms of just general business, so you know, and we see, you know, we're starting to see a convergence of a lot of these things happening.
• 31:00
  SPEAKER_04: And we did some some media related um topics with another gentleman, and it was really interesting where yeah, where synthetic media, yeah, where the convergence too now of classic data, unstructured uh data, and now rich media data is now you know starting to really proliferate around, you know, not only the classic, you know, here's what's said within a document, but now here's the metadata about that information, and now it's rich media, and now here's what's what you know, the vision is being pulled out of it.
• 31:35
  SPEAKER_04: Human augmentation stuff, machine augmentation stuff, you know, all these key things.
• 31:39
  SPEAKER_04: You know, how much of you know this, Tom, where you know, are you seeing around organizations trying to figure out do they need to get ahead of the curve on some of this augmented reality piece?
• 31:49
  SPEAKER_04: Do they need to actually figure out how to ingest their rich media content at the same time with their non-structured and structured content?
• 31:57
  SPEAKER_04: What does that look like?
• 31:59
  SPEAKER_02: Um, yeah, I don't work all that much in that space.
• 32:03
  SPEAKER_02: A little, you know, I've done a little bit of research in places where they use AR and VR to um, you know, capture and disseminate knowledge, which is a great tool for.
• 32:16
  SPEAKER_02: Um, but yeah, I think we're still sort of figuring that out.
• 32:20
  SPEAKER_02: I some of my friends at Deloitte, where I'm a senior advisor, say, oh, yeah, they get a lot of requests from clients about the metaverse.
• 32:27
  SPEAKER_02: I don't get personally very many, but um, I thought it was very interesting yesterday that uh Mark Zuckerberg had a I think an investor conference or something, and he said, Well, you know, this stuff is gonna be really influential in the 2030s, and that's why we've been spending so much on it.
• 32:50
  SPEAKER_02: And the Wall Street analyst said, Are you crazy?
• 32:53
  SPEAKER_02: You know, you if you keep spending at the level you are until the 2030s, you're not gonna exist anymore.
• 33:00
  SPEAKER_02: So um uh, you know, just how forward looking should we be on some of this stuff is I think a little unclear at this point.
• 33:08
  SPEAKER_01: Yeah, yeah.
• 33:09
  SPEAKER_01: No, that's a really good point.
• 33:10
  SPEAKER_01: When you come back to the knowledge world and you know, think give give us your thoughts on uh how knowledge management is going to be uh kind of uh conceived of and and operate.
• 33:23
  SPEAKER_01: Yeah, what are your thoughts?
• 33:24
  SPEAKER_01: I mean, do executives think, oh, we're not gonna need that because we're gonna use AI and machine learning.
• 33:29
  SPEAKER_01: I hear that a lot.
• 33:30
  SPEAKER_01: Uh or do you really think that people still will need to get uh you know their act uh together with regard to knowledge and knowledge processes and structured uh you know, to converting tacit to explicit, you know, to power things like virtual assistants and bots and all of that?
• 33:47
  SPEAKER_01: I mean, to me, that's kind of where uh I see some some uh trending need.
• 33:52
  SPEAKER_01: And uh Gartner just did a uh uh projection saying that knowledge management for AI will be the fastest growing segment and the largest spend.
• 34:01
  SPEAKER_01: And I and uh what are your thoughts about that?
• 34:03
  SPEAKER_01: Does that uh resonate with you, or do you believe that's gonna be the case?
• 34:06
  SPEAKER_02: Yeah, I don't know exactly what that means, knowledge management for AI, but I certainly am a big supporter of it.
• 34:14
  SPEAKER_02: I'm not quite as loyal as you are, Seth, to knowledge management because I, you know, I I want to move to where the you know high levels of interest are in companies and um uh where they you know want to talk to me, the things they want to talk to me about.
• 34:32
  SPEAKER_02: And for years I did knowledge management stuff, and then people stopped wanting to talk to me about it.
• 34:37
  SPEAKER_02: So uh fortunately, I realized that there were kind of some aspects of knowledge management that um could be applied to the whole area of analytics and then to big data and even to AI, the the sort of human side of it.
• 34:53
  SPEAKER_02: It was quite well known in knowledge management, but not so well known in those other areas.
• 34:59
  SPEAKER_02: But I I'd love it to come back.
• 35:01
  SPEAKER_02: I think knowledge is the you know, it's the most important resource that we have.
• 35:05
  SPEAKER_02: I am occasionally um quite pleased to see that companies are combining AI and knowledge management.
• 35:16
  SPEAKER_02: I think to do a good job of AI, you have to have your knowledge in order.
• 35:20
  SPEAKER_02: I do agree that um with you that you know, if you want to have AI, you need IA.
• 35:27
  SPEAKER_02: Um, however, um I'm reluctant, you know, I before knowledge management, I did most of my work in business process reengineering, and I keep hoping that'll come back too.
• 35:38
  SPEAKER_02: And I have friends who work with me on business process reengineering who see it coming around the bend any day now for decades.
• 35:49
  SPEAKER_02: Right.
• 35:49
  SPEAKER_02: And so I just I try not to wish too hard for these things to come back, even though I think they should.
• 35:56
  SPEAKER_02: I just try to say, well, what are what are people interested in at this moment?
• 36:00
  SPEAKER_01: We'll be right back.
• 36:02
  SPEAKER_01: We'd like to thank Marketing Artificial Intelligence Institute for their support of this podcast.
• 36:07
  SPEAKER_01: Marketing AI Institute is a media, event, and online education business that makes AI approachable, actionable, and accessible to marketing leaders around the world.
• 36:16
  SPEAKER_01: They run the world's leading marketing AI conference, MACON, which brings together hundreds of marketing and business leaders to better understand, pilot, and adopt AI.
• 36:27
  SPEAKER_01: Today's podcast is brought to you by Simpler Media Group, owner of the Customer Experience Decoded Podcast, covering the latest in customer experience and marketing.
• 36:34
  SPEAKER_01: CX Decoded is a semi-monthly podcast designed to deliver insight from leaders driving the evolution of customer experience with an emphasis on digital touch points, customer journeys, respectful data management, and voice of the customer.
• 36:53
  SPEAKER_04: Early Information Science is a professional service firm with proven methodologies designed specifically to address product data, content assets, customer data, and corporate knowledge bases.
• 37:03
  SPEAKER_04: Their information architecture expertise has allowed clients from all over the world to efficiently organize, access, and utilize their data, ultimately resulting in better business outcomes.
• 37:12
  SPEAKER_04: They deliver customer-driven solutions that scale and adapt to the needs of a business as it evolves.
• 37:18
  SPEAKER_04: Founded in 1994, Early Information Science has proudly delivered result-driven solutions for more than 100 brands across the top industries.
• 37:26
  SPEAKER_04: You can get more information about Early Information Science at early.com, E-A-R-L-E-Y.com.
• 37:35
  SPEAKER_04: Working with customers in AI, the number one question that I get usually is where do I start?
• 37:40
  SPEAKER_04: What data do I need?
• 37:41
  SPEAKER_04: How do I gather it?
• 37:42
  SPEAKER_04: And that's great questions.
• 37:44
  SPEAKER_04: Well, if you want to catch the wave of AI, not be crushed by it, you really do need to read a book written by Seth early called The AI Powered Enterprise.
• 37:52
  SPEAKER_04: It gives you the ability to look and understand how to harness the power of ontologies to make your business smarter, faster, and more profitable.
• 38:00
  SPEAKER_04: Seth goes into the intricate details of how to build an ontology, develop and employ an ontology as a secret weapon for the basis for how to actually utilize AI as a powerful competitive advantage.
• 38:12
  SPEAKER_04: This book is really the first one to combine sophisticated explanation of how AI works with a practical approach of applying AI to the problems of the business.
• 38:21
  SPEAKER_04: And it's applied to, you know, various use cases from customer experiences to business operations to product development.
• 38:28
  SPEAKER_04: It's a powerful, powerful tool in the arsenal of understanding how to deploy AI to make your business better.
• 38:33
  SPEAKER_04: The AI Powered Enterprise by Seth Early.
• 38:36
  SPEAKER_04: Harness the power of ontologies to make your business smarter, fashion, more profitable.
• 38:39
  SPEAKER_04: Out on Amazon.com today.
• 38:41
  SPEAKER_04: And now back to the show.
• 38:44
  SPEAKER_01: Just to refresh people, uh, we are speaking with Tom Davenport uh today.
• 38:48
  SPEAKER_01: And uh Tom has written a number of books on written a couple of books on on AI and lots on analytics and and so on.
• 38:56
  SPEAKER_01: Tom, what are you writing these days?
• 38:58
  SPEAKER_02: Well, I only I only, as you say, I only had a couple of books on AI, so I'm trying to add to the total.
• 39:03
  SPEAKER_02: So I have three more on AI coming out later this year.
• 39:07
  SPEAKER_01: Um and and in your free time, do you wonder why you have no free time?
• 39:11
  SPEAKER_02: Why, yeah, why do I have no friends?
• 39:14
  SPEAKER_02: Why do I uh why is my tennis game not ever getting any better?
• 39:18
  SPEAKER_02: Um uh so yeah, I like to write a lot, and um, I thought AI was the most important thing around for a while.
• 39:25
  SPEAKER_02: I'm still I'm trying to figure out what to do next, but I'm still sort of fascinated by AI.
• 39:31
  SPEAKER_02: But anyway, um, one is on companies that use AI really aggressively in their business.
• 39:38
  SPEAKER_02: I wrote a book um uh 15 years ago, I think, 12 years ago, called Competing on Analytics.
• 39:44
  SPEAKER_02: And this is sort of competing on AI, effectively.
• 39:48
  SPEAKER_02: Um, and then one on people who work with AI on a day-to-day basis, that's called working with AI, um, and um one on um healthcare AI.
• 40:01
  SPEAKER_02: Um so those are the three things that I've been working on.
• 40:08
  SPEAKER_01: And when you uh when you start looking at what people are trying to do around customer experience and personalization and recommendation uh engines, are you seeing uh organizations doing that effectively?
• 40:21
  SPEAKER_01: Are you you have some experience in seeing uh what they're doing in those spaces?
• 40:25
  SPEAKER_01: Because that seems to be an area where people are saying, well, I'm gonna use these uh magical uh machine learning tools uh to personalize my customers' experience.
• 40:34
  SPEAKER_01: I have a philosophy on that.
• 40:36
  SPEAKER_01: But what are your what are your thoughts?
• 40:37
  SPEAKER_01: What are you observing?
• 40:38
  SPEAKER_02: Well, you know, that's one of the areas that we've been waiting a very long time to come.
• 40:43
  SPEAKER_02: That uh I don't know if you remember this book, um, one to one on one-to-one marketing.
• 40:48
  SPEAKER_02: Yes, yes, yes.
• 40:49
  SPEAKER_02: And I I don't know about you, but I have yet to receive any good one-to-one marketing.
• 40:57
  SPEAKER_02: Um I sometimes say once I got an uh um offer for a discount on Groupon to a restaurant that I really wanted to go to, and it almost brought a tear to my eye.
• 41:09
  SPEAKER_01: It's such a rare experience.
• 41:11
  SPEAKER_01: They actually got sent me something I want.
• 41:13
  SPEAKER_02: Yeah.
• 41:14
  SPEAKER_02: Um, so now I think in terms of the personalization space, people are calling it hyper-personalization because you know, we never really got personalization right in the first place, and now we have to have a whole new name to describe it.
• 41:28
  SPEAKER_02: I do think that there are some fantastic opportunities to use machine learning to um hyper-personalize interactions with customers and their their customer experience or digital experience and so on, but still I I have yet to see great examples of it.
• 41:48
  SPEAKER_02: Um uh I've been writing some interesting um or doing some uh uh discussions of interesting work at say Morgan Stanley in that area with regard to um kind of personalizing um their uh investment recommendations.
• 42:08
  SPEAKER_02: Um but it go ahead.
• 42:11
  SPEAKER_01: No, tell me more.
• 42:12
  SPEAKER_01: I was just gonna say, what what is what are the preconditions for them to be able to do that successfully and and how are they doing?
• 42:19
  SPEAKER_02: Yeah, well, you know, the the person who really wanted this to happen was their vice chairman.
• 42:24
  SPEAKER_02: He had this vision that, you know, just as we make recommendations about Netflix movies, we should make recommendations about um investments.
• 42:32
  SPEAKER_02: Of course, he's retired now.
• 42:34
  SPEAKER_02: This is a decade ago or so.
• 42:36
  SPEAKER_02: But um, so they developed this um next best action system, as they call it, um, to identify um investment recommendations that you might be interested in that fit with your preferences and so on.
• 42:52
  SPEAKER_02: Um, but over time, I've you know, I've been writing about this for four or five years now.
• 42:58
  SPEAKER_02: They've sort of evolved and they started out as a separate system and said, well, separate systems don't work all that well.
• 43:04
  SPEAKER_02: Let's integrate it with the way we communicate with our customers, which happens to be um Salesforce.
• 43:11
  SPEAKER_02: And what they've realized is that the recommendations are important, but what's really important is making the client think that you're, you know, engaged with them and paying attention to them and have their welfare at heart and yeah, that you care.
• 43:27
  SPEAKER_02: So um they um now it's sort of a 50-50 recommendation and communications system, and um, it's working really well.
• 43:40
  SPEAKER_02: It's fantastic during the pandemic when you couldn't actually go sit at um in the office of your Morgan Stanley financial advisor.
• 43:48
  SPEAKER_02: But the people who they're they're being very um uh hands off in terms of making people use it, but the financial advisors who use it tend to be much more successful.
• 44:00
  SPEAKER_02: Surprising, they have about 200 clients each on average, and um, to come up with a good idea used to take 45 minutes, now it takes a second.
• 44:10
  SPEAKER_02: Um, for for that client.
• 44:12
  SPEAKER_04: How do you see um, or do they have an appetite for doing self-service related items as well?
• 44:19
  SPEAKER_04: You know, not not always interacting with human to try to you know get some of that done.
• 44:24
  SPEAKER_02: They have, you know, their average um client, um, I think is in their 60s.
• 44:30
  SPEAKER_02: Um, so a lot of them um are and and it's also you know, they they'd rather charge more money and not be self-service, but they did establish a separate offering for um younger, I guess more digitally oriented people who could get those recommendations directly and not go through a financial advisor.
• 44:54
  SPEAKER_02: I don't know how successful it's been, but they realize, you know, um they've got to do something like that.
• 45:01
  SPEAKER_04: I think the the it goes back to kind of that uh chicken and egg scenario, too, where um for me anyway, where you know my kids want that self-service instantaneous, you know, look at it, but they don't have the education to know what the hell is being sent to them either.
• 45:18
  SPEAKER_04: So it's it's like, hey, invest here, do this, do this.
• 45:22
  SPEAKER_04: And I see them like jumping back and forth, like, well, do I do that?
• 45:25
  SPEAKER_04: Do I do that?
• 45:25
  SPEAKER_04: And then they come to me and like, Dad, what should we actually do?
• 45:28
  SPEAKER_04: And then we take a minute to take a step back and think about it from the perspective of what they're missing is they don't even know what their whole goals are that they want to work towards.
• 45:37
  SPEAKER_04: Yet they have all of this barrage of information hitting them, and that's I think the big difference between, you know, like we talked about having somebody there to help lead guide and help them create the knowledge and understanding, as opposed to just getting blasted with all this, you know, this digital stuff.
• 45:55
  SPEAKER_04: So it works and it doesn't work in a lot of cases, right?
• 45:57
  SPEAKER_04: Hey, I want to, you know, don't you don't need to authenticate me.
• 46:00
  SPEAKER_04: I don't need to, you know, do all of that kind of stuff, but I do need to be educated, and I don't know that I need to be educated, so I'm gonna get frustrated and then finally just kind of take a step back and say, uh somebody help me.
• 46:12
  SPEAKER_04: Dad, help me.
• 46:13
  SPEAKER_04: What do I do?
• 46:14
  SPEAKER_02: You know, that's that sounds like very good advice, Chris.
• 46:16
  SPEAKER_02: You charge them one percent of their assets for that advice.
• 46:19
  SPEAKER_02: Well, oddly, oddly enough, they asked me, like, Dad, what are you leaving to us?
• 46:25
  SPEAKER_04: I'm like, This is weird.
• 46:26
  SPEAKER_04: So I get to pay for you, I get to invest my money for you, and then I get to also help you invest where you're gonna go at the same time.
• 46:36
  SPEAKER_04: Like, how the hell is this working out?
• 46:38
  SPEAKER_04: I feel like it's just an out that you know, just I'm just hemorrhaging all the time.
• 46:42
  SPEAKER_04: Being your parent is a heck of a deal, isn't it?
• 46:45
  SPEAKER_04: Yeah, exactly.
• 46:46
  SPEAKER_04: So it's a yeah, very little you know, investment coming my way, but it's all good.
• 46:51
  SPEAKER_01: So people talk about democratization of AI.
• 46:55
  SPEAKER_01: What does that mean, uh, in your opinion?
• 46:58
  SPEAKER_02: Uh you know, I think we're moving toward a situation where um through various sorts of automated tools, automated machine learning, automated data discovery, et cetera, where we don't have to rely on data scientists or even you know professional analysts to um help us find out what's going on in our data and you know find models that that fit our data really well.
• 47:25
  SPEAKER_02: I think we're still in the early stages of that, and not too many organizations have you know made a substantial commitment to that approach.
• 47:35
  SPEAKER_02: But I do think you know, there just aren't enough data scientists in the world, and there never will be to explore all the data that we're generating.
• 47:45
  SPEAKER_02: And you if you have that domain knowledge, um, and you can combine it with um the ability to create um a good model and choose the right features or variables or whatever, um, I think that's gonna be a superior approach.
• 48:00
  SPEAKER_02: So I think in the future we'll see a lot more companies embracing that.
• 48:04
  SPEAKER_01: And it's going to be more about understanding your business problem, understanding that outcome.
• 48:09
  SPEAKER_01: And I think the process engineering is going to play a big role.
• 48:13
  SPEAKER_01: You can't automate what you don't understand, and you can't automate a mess.
• 48:17
  unknown: Right.
• 48:17
  SPEAKER_02: So well, yeah, as I say, I don't um spend too much time hoping that that will come back, but I would I would like to see it come back.
• 48:25
  SPEAKER_02: And I I am working on a little paper saying AI could bring re-engineering back, even despite my best um instincts.
• 48:35
  SPEAKER_01: Well, it's interesting, and you know, the other thing we're seeing interest in, of course, I wrote the book about uh ontologies and uh and there's interest in knowledge graphs.
• 48:45
  SPEAKER_01: Do you think that and I see knowledge graphs as kind of uh you know, another name for a lot of stuff we've been doing for a long time?
• 48:52
  SPEAKER_01: Do you have any uh insights of how organizations are looking at uh those types of things?
• 48:58
  SPEAKER_01: You know, looking at that enterprise view of uh uh you know the conceptual relationships, you know, the knowledge scaffolding, the information, the enterprise architecture.
• 49:06
  SPEAKER_01: I know you don't believe in the big picture, big bang MDM types of initiatives, but you know, but how organizations still need to, what is your opinion about the need for that sort sort of thing or the perspective on that, or or even the appetite?
• 49:21
  SPEAKER_01: You could share any of your thoughts.
• 49:22
  SPEAKER_02: Um, I I think certainly for any sort of um you know uh content-rich AI system, you're gonna need that.
• 49:34
  SPEAKER_02: And um it's a labor-intensive process, as you know better than I.
• 49:40
  SPEAKER_02: Um, there are some organizations that are starting to play around with using AI to help identify, you know, what are the relationships between different um concepts and terms and so on in their organization.
• 49:53
  SPEAKER_02: And, you know, I think we'll see AI helping with a lot of different areas of data and um knowledge, content management.
• 50:02
  SPEAKER_02: Um, but we're in the very early stages of it thus far.
• 50:05
  SPEAKER_02: But I do think it's an important activity for a certain type of um AI system, you know, um anything related to chatbots or intelligent agents or even um recommendation engines, you have to know sort of what are the relationships between your products if you're gonna know what you're gonna going to recommend.
• 50:26
  SPEAKER_01: You need good product data and you need good foundational uh stuff.
• 50:31
  SPEAKER_04: Okay, Thomas, besides uh besides three, you know, 40 books that you're writing for this next year, which is impressive, by the way.
• 50:39
  SPEAKER_04: Uh what else does 2022 have in store for you?
• 50:42
  SPEAKER_04: Either personal fun or or uh professional?
• 50:47
  SPEAKER_02: Well, I hope to get back over to Oxford.
• 50:49
  SPEAKER_02: I was there last fall, but there was really no no there were no other faculty there.
• 50:55
  SPEAKER_02: And my um teaching ended up to be virtual teaching to um Dubai.
• 51:02
  SPEAKER_02: So I was wondering exactly what I was except for some time zone benefits.
• 51:06
  SPEAKER_02: I was wondering what I was doing there.
• 51:07
  SPEAKER_02: So I hope to go back there.
• 51:09
  SPEAKER_02: Um, first grandchild is supposedly arriving this year, so that'll be Seth tells me that's a big thing.
• 51:17
  SPEAKER_03: Um well you get to rent them at all.
• 51:20
  SPEAKER_04: I've got I've got my first, he's actually here walking around, and I gotta tell you, kids is fun and it's life-changing, but grandbabies is a whole new level of great.
• 51:30
  SPEAKER_04: It is finally the actually that's that's the payback, like we talked about before.
• 51:35
  SPEAKER_04: It feels like like we talked about that the wallets are emptied and the cut and the coffers are cleaned out.
• 51:40
  SPEAKER_04: As soon as that grandbaby comes back, and you're like, Oh, good lord, man, this is what it's saying for.
• 51:47
  SPEAKER_02: Congratulations.
• 51:48
  SPEAKER_02: Something to do with why you moved to Montana or it does partly, yeah.
• 51:52
  SPEAKER_02: Yeah, yeah, exactly.
• 51:53
  SPEAKER_02: Yeah, we bought a house in Santa Barbara for the same reasons.
• 51:56
  SPEAKER_02: Yep, yeah, yep.
• 51:57
  SPEAKER_02: You'll get to rent, not own.
• 51:59
  SPEAKER_04: Get to rent well, yeah, but it's at the same time, see you do like crazy things like that, like picking up and relocating and moving or buying, you know, these things just to do you know for vicinity.
• 52:10
  SPEAKER_04: So it will change your life in the best ways.
• 52:13
  SPEAKER_04: So congrats.
• 52:14
  SPEAKER_02: Yeah, we don't we don't need to be near our co-workers, but we do need to be near our family members.
• 52:20
  SPEAKER_04: Yeah, exactly.
• 52:22
  SPEAKER_01: So you're uh you're gonna be back up in Falmouth.
• 52:25
  SPEAKER_01: Um, I know you uh you one of the things you uh uh inserted into the uh intro of the book was your experience on uh on the Thundercat.
• 52:33
  SPEAKER_01: Is is uh is that going to be uh you're gonna try that again someday in uh in your future?
• 52:38
  SPEAKER_02: You know, I don't know.
• 52:40
  SPEAKER_02: I uh I must have come close to a heart attack the last time I was on that um vehicle, though.
• 52:48
  SPEAKER_02: No, it was a lot of fun.
• 52:49
  SPEAKER_02: I did ruin my shoes.
• 52:51
  SPEAKER_02: I didn't realize that they were gonna get totally soaking wet um uh in the first minute of our venture, but um it was it was certainly a lot of fun.
• 53:00
  SPEAKER_02: I would I would consider another uh attempt at yeah hanging on to the boat.
• 53:06
  SPEAKER_01: So yeah, I remember the thing you said to me afterwards.
• 53:09
  SPEAKER_01: You've you've been on boats your entire life and you had never experienced anything quite like that.
• 53:13
  SPEAKER_02: So no, I've never been moved so fast, so close to the water as as that experience.
• 53:21
  SPEAKER_01: Well, I just uh just started the engine the other day to to you know get it going and uh inflate it.
• 53:27
  SPEAKER_01: It's a for people who don't know, it's a probo thundercat, it's an inflatable um uh tunnel hull.
• 53:33
  SPEAKER_01: So it's a little like uh power cat in a ran, which uh which is a lot of fun.
• 53:36
  SPEAKER_02: How fast does it go?
• 53:38
  SPEAKER_01: It's about 50, about 50 knots.
• 53:41
  SPEAKER_01: And you can actually yank the tiller and do a 90 degree turn if your cobalt is hanging out on the other side to keep it from flipping.
• 53:48
  SPEAKER_02: So yeah, and you're about two inches from the from the surface of the water.
• 53:55
  SPEAKER_01: It's a lot of fun.
• 53:56
  SPEAKER_01: Hey Tom, this has been a tremendous pleasure.
• 53:59
  SPEAKER_01: I've really enjoyed uh you know catching up with you again.
• 54:03
  SPEAKER_01: Uh anything else?
• 54:04
  SPEAKER_01: Uh any final words for uh organizations that are kind of still figuring things out and still trying to make their way?
• 54:13
  SPEAKER_02: Uh think big but start small is what I usually say about all these things.
• 54:17
  SPEAKER_02: You know, have a big plan for how you're gonna really change your business dramatically, but probably smarter to do it in small pieces.
• 54:25
  SPEAKER_04: I thought you know what, I read something else that you wrote too, where it was also along the lines of you know, that think big really needs to be like more the long-term plan, not just do something that's short term.
• 54:36
  SPEAKER_04: And I think that's where a lot of organizations get off track too, is again, they think it's just a quick hit and that's it, but it's so much longer than that, and they don't know how to set their expectations correctly.
• 54:47
  SPEAKER_04: Is that fair?
• 54:48
  SPEAKER_02: Yeah, you know, I think it's kind of funny that companies talk about these digital transformation projects.
• 54:54
  SPEAKER_02: It should be a way of life, and it and for the best companies it is a way of life.
• 54:58
  SPEAKER_01: Yep.
• 54:59
  SPEAKER_01: Yep.
• 55:00
  SPEAKER_01: And uh going from pilot to production is that big, big leap, right?
• 55:04
  SPEAKER_04: You don't have the you know what though, it's always just version one, right?
• 55:09
  SPEAKER_04: That's version one, yeah, and you can always increment from there.
• 55:12
  SPEAKER_04: So Tom is a pleasure to meet you and talk to you.
• 55:15
  SPEAKER_04: Thanks for the time.
• 55:16
  SPEAKER_01: Great to catch up with you.
• 55:17
  SPEAKER_01: Look forward to uh taking you out on the Thundercat again this summer.
• 55:20
  SPEAKER_02: So thanks a lot to both of you.
• 55:22
  SPEAKER_02: See you later.
• 55:23
  SPEAKER_01: This has been great.
• 55:24
  SPEAKER_01: All right, take care.
• 55:29
  SPEAKER_00: Thanks for listening to the Early AI podcast with your host, Seth Early at Chris Beatherstone.
• 55:36
  SPEAKER_00: We hope you enjoyed the program and took away some ideas that you can start to implement in your organization right now.
• 55:42
  SPEAKER_00: If you're stuck and need help, check out www.early.com for information about early information science and software services.
• 55:50
  SPEAKER_00: And of course, check the show notes for links to any resources we mentioned today.
• 55:54
  SPEAKER_00: That's all for now.
• 55:55
  SPEAKER_00: See you next time.