Lisa Woodley: I just realized I held up a book for a podcast. So... (Laughs)

(CATALYST INTRO MUSIC) 

Clinton Bonner: Welcome to Catalyst, the Launch by NTT Data podcast. Catalyst is an ongoing discussion for digital leaders dissatisfied with the status quo and yet optimistic about what's possible through smart technology and great people. Make sure you are subscribing in your audio feed, help spread the word on these great conversations that we think are truly worthy. Speaking of truly worthy, I have an amazing guest. A friend, a coworker, a colleague, someone I've even shared the stage with recently. We're going to be talking with Lisa Woodley, Head of Design at Launch by NTT Data. On this podcast every now and again we like to review, you know, current events, events that are actually physical events as well, things that we travel to, because we understand that, hey, we have the privilege of going to, you know, some of the biggest events throughout the entire world. So recently, I think it was late June, Lisa, you had the chance to go to the Forrester CX event. But I actually forget. What town was it in this year? 

Lisa: It was in Nashville this year. So, always a fun place to go for a conference. 

Clinton: Any special places you like to go? When... You're in Nashville, the event is winding down, or the day is winding down. Is there a particular place you found, you know, good grub, a good drink or two? A high recommend. Because you travel to quite a number of cities. 

Lisa: You know, I do. The funny thing for this one is, it was a quick in, quick out for me, so it broke my heart a little bit, i didn't get to spend as much time in Nashville. But, the event itself was at the Gaylord Opryland. 

Clinton: Yes. Yep. 

Lisa: So, I did get a chance to go over and just check out the Grand Ole Opry, which was kind of fun because it was literally right next door. So that was about the most Nashville I got. You know, I'm a big fan of speakeasies. 

Clinton: Yeah. 

Lisa: And Nashville has a couple of them, but I didn't get a chance to hit any of them this time around.  

Clinton: Got it. Got it. Well, you know, a place to have an achy breaky heart, of course, is Nashville. But next time, next time. Alright, so the Forester CX event. You were commenting on Slack as it was going on, and one of the major things you were flagging was like, whoa. This entire event is really, really talking about a single topic that was really on everybody's mind. Can you share what that was? 

Lisa: Yeah, sure. Let me back up a little bit, though... 

Clinton: Sure. 

Lisa: Just because I don't know if everybody knows what Forrester's CX summit is. So I just want to back up a little bit on that one. 

Clinton: Lay the groundwork. Let's go. 

Lisa: So, I mean, I think everybody knows who Forrester is, one of the big analyst firms. And Forrester really is heavily focused in my area, which is design, customer experience in particular for this. So their CX summit... 

Clinton: Mhm. 

Lisa: There's global. This one was for North America. It's really one of their biggest events. And it's bringing together digital leaders, customer experience leaders, design leaders, all coming together to sort of talk about what's impacting our industries. Really specific to around, like, people interacting with products and everything from marketing to actual product, like what we do here at Launch. So, you know, I've attended several of these in the past, because it's one of the, like, the big ones for the design industry. The thing that really struck me this year... So they always talk about, you know, there's multiple tracks every year. And the multiple tracks are different aspects of customer experience. And the thing that really struck me with Forrester this year is there was not a single track that was not focused on AI. And that is completely different from any other year. And don't get me wrong, there were different tracks for customer experience from a marketing perspective, or from a customer service perspective, or from building digital products perspective. So there were different tracks, but the theme of every single track was, engaging with your customers using AI. And, you know, leveraging AI for marketing. I've just... I've never been to a Forrester Summit where there was such a focus on one technology. And I paused when I say one technology, because it really made me realize that AI is not exactly a technology. 

Clinton: Right. 

Lisa: Like, it is. (Laughs) But it's a lot more than that, because you would never... You know, Figma is super important, but Forrester would never do, like, all tracks on how you use Figma. But to have every track on AI really showed me, first of all, how quickly, in even the span of a year, AI in the design and digital product space has completely taken over. But then, second of all, the implications of it, and what it means, and the uses of it and the considerations was more than enough to fill up the full four days of this summit. With, like, really interesting topics. Like, I mean, I know when I came back, I was like, we got to get this on a podcast. 

Clinton: Yeah. 

Lisa: I think my head was just like, so filled with amazing things that people were doing with AI. Cautionary tales. So that, to me, it really struck me, was like, AI has really, in this short span of a year, completely taken over everything. 

Clinton: Yeah. That speed in which it engulfed a traditional event, like you said, that happens every single year, and that it just became... Not only the centerpiece, but permeated all facets of it as well, is truly remarkable. 

Lisa: Right. 

Clinton: And I guess the question from my perspective is, did it feel frothy and hypey, or, no, it just, it felt like it needs this level of entertaining and this level of focus? 

Lisa: None of it felt frothy or fluffy or whatever. 

Clinton: Right. 

Lisa: And I am... I am old enough that I've been through technological changes. And on this side of it. I went from print design, I was a print designer. And then the internet thing came out, and all of a sudden I was like, that's interesting. I loved the impermanence of it, meaning, like, you print a mistake and it's... (Laughs) It's out there... 

Clinton: For. Ever. 

Lisa: So, forever. Right? I loved the fact that things were changing and evolving. And then the same with mobile. But the speed with which this happened... Like, web, we were slowly building up to that. 

Clinton: Sure. 

Lisa: To that really taking off. And while it felt overnight, the mobile and the iPhone taking over felt like it was overnight, it was definitely the span of more than a year or two before it, like, actually got out there. And this is literally, like, super condensed in terms of the speed with which people are adopting it. And so, normally, if it was any other technology, I would be very skeptical, and like, well, you know, maybe you're just hyping, you're jumping on the bandwagon and hyping up the latest thing. But there's this weird intersection of more and more companies adopting the agile, fail-fast sort of mentality, intersecting with what AI can do. And I say that because there's, like, genuine, real successful use cases already. And that was the thing that Forrester really focused on on the summit, was getting people up on stage talking about, here is how we're doing it, here's what we're doing. And I think that's the big difference between a lot of other hype that happens, is this is... It's already been sort of demonstrated. And there's different aspects of AI as well. You know, there's AI as the tool. 

Clinton: Yep. 

Lisa: That we might use to create things. There's AI that is how we might have an AI interaction, right? So you may be interacting with a bot. And then, of course, there's the AI to use data. And all three of those affect so many different parts of the business. So that's the other thing that I think is what's propelling this and making it not hype right now is, it's, it's, one is there's real use cases. People are already out there using it. But two, it's not just one area or one kind of solution that it's impacting. It's, there's different ways to use it. And that's why I don't think of it as a technology. 

Clinton: Right. 

Lisa: It's a technological advancement, but many ways to use it. So no. And it... Forrester was incredibly thoughtful about how they talked about it, and the types of topics. So, the ethics of this is always very high on my priority list. 

Clinton: Yeah. 

Lisa: For a multitude of reasons. And that was really kind of one of the big topics. And the whole... The whole thing was called AI plus human. And even in the keynote, they started out with putting that out there, that initial keynote where the CEO got up and talked about, sort of the, you know, things that we see coming. But then it was very quickly followed by, I believe it was J.P. Gownder got on and talked about... I think there's some quote from, I don't know, Elon Musk or somebody that was like, you know, within the next year or two, AI is going to be smarter than any other human. A human with AI is going to be the smartest person on earth. So, right? It was that idea that the AI by itself isn't going to beat anything, but a chess master with AI is going to beat everything. So, the entire thing was really focused around, how do we use this to improve? So, combining human with AI. To improve, whether it's the way we do our job, or how we interact with customers. 

Clinton: It reminded me of the old story of Paul Bunyan. And of course, you know, Paul Bunyan's a logger and he's out there and he has his axe. And then, of course, Paul is against the machine, they introduce a logging machine to go to cut down the trees so they can produce what they're producing. And I always thought it was kind of funny, like, even as a kid, I was like, why doesn't he, the best human, use the machine? Like, he'll be unstoppable. 

Lisa: Right. Exactly. 

Clinton: And of course, it's a sad story, right? He ends up, him and his ox end up walking into the sunset, they kind of have to retire. That was the bet. A bet's a bet. But I'm like, man, Paul, just pick up the machine. You'd be absolutely unstoppable. And I think that is the, kind of the grand lesson that you're saying, is like, look, it is that fusion, of course. And I like what you said earlier too, about... Well, a couple of things. Like, first of all, the speed in which it moved through, it's pretty evident that obviously, without internet and without mobile and the consumerization of the applications that could use AI, well, they all helped AI get absorbed that much quicker. 

Lisa: Oh yeah. Yeah. 

Clinton: And so, they just kind of build on each other and they, speed, too, in which they could permeate, just seems to continue to accelerate until there's some, you know, breaking point or whatever you call it, like, speed limit at which the information can be picked up at, that quickly. But the piece that you said, too, about, look, we've got the consumer side and how, like, a designer or marketing might use it. And then the other wave that's quickly coming in, like you mentioned, the data side, and the advancements in things like private 5G in the capital equipment and edge AI, the things that can be computed in a secure box outside on an edge that's owned by a company, and that could just, hey, it doesn't even have to go to the cloud. It could just... The AI could do its work on the edge, on prem, and do routing and make decisions for you right there with fresh data. I mean, for the enterprise, whole new world there. I know the CX event tilted more towards back designers. 

Lisa: Yeah. 

Clinton: But, any thoughts on, kind of, that bleeding edge that we're now beginning to see as well? 

Lisa: Well, I don't... It didn't really come up there. It does impact designers. And I think it's interesting that you talked about whatever the speed limit is. 

Clinton: Right. 

Lisa: And then, I was about to say, what is the speed limit now? And then you brought up, exactly. Right? It's, the "speed limit," quote-unquote, is changing with edge compute. 

Clinton: Right. 

Lisa: I was going to make a quantum computing joke, but it's probably not even a joke anymore, right? Like, between all of those edge computing edge services on prem, private 5G, all of that enables us to do this faster. Like, incredibly faster. And it is, for me, it is from an experience. 

Clinton: Sure. 

Lisa: A customer experience perspective. Because, the more of all of that stuff we have, the more the speed that we have with all of that, the faster we can engage. And when I say faster, like, there's no lag time. And so what that means is, should we decide to really go full on with, like, an AI interface? I hate to call it a chatbot, because nobody... Like, chatbots are like, you know, glorified FAQs sometimes. 

Clinton: Mhm. 

Lisa: But I'm talking about a chatbot that has an actual, like, AI behind it. So, more of a virtual assistant. 

Clinton: Yep. 

Lisa: That stuff is going to happen a lot faster. We're going to enable it to happen a lot faster, and so we're going to be able to use it in more cases. Right? So, I mean, I think those are incredibly important, because the inability to actually have the calculations that your AI has to do in order to produce something... Like, speed is going to be of the essence. And so, that opens the door for us to create all kinds of different experiences. Because we don't have to worry about compute time and lag time and all of that. 

Clinton: And even the data sets, the data sets can be markedly smaller. 

Lisa: Right. 

Clinton: Because, like, as it becomes less and less of a barrier to deploy AI out on the edge, you could do it for more discrete use cases, meaning it doesn't need to ingest entire, the entirety of Wikipedia to spit out an answer. 

Lisa: Right. 

Clinton: It might just need to know the data coming off some capital equipment, and maybe who's credentialized in a certain area to go fix something if something was broken. And it could triage and make decisions, blink of an eye, it's not... the speed of light, is how it can do it. 

Lisa: Right. 

Clinton: Or just below the speed of light, which is insanity. And really, really powerful. And then, back to the conference with Forrester. It did have a large focus on design. So, is there a fear in the room? Is there a fear in general around the perception of, hey, that's going to take our jobs, that's going to take our being and our purpose out there as designers, as AI becomes more and more prevalent? 

Lisa: Yeah, I think that was the whole reason that Forrester wanted to focus around AI and human. To sort of alleviate that fear. It's less and less, as designers start to work with these tools and see them and start to see their limitations. And I've been playing around with it, you know, when GPT came out with Dall-E. I've been playing around with it. Just because I really want to get an understanding of what can and can't, it can't do. And as we start to work with it, it starts to become really, really clear. Back to, you know, Forrester's point, a designer with AI is going to be the most powerful thing for our industry. Versus a designer by themselves. And what I'm realizing is... Like, it's great for production work. You come up with an idea and it sort of cranks out 10 or 11 different things, and then you can dig into it and say, okay, this works, this doesn't work. It's kind of like... To a greater extent, but I am old enough to remember when Photoshop first came out. So I was a production artist right out of college when Photoshop first came out. 

Clinton: Nice. 

Lisa: And I was still doing paste-up and mechanicals and all that stuff, where you're, like, literally cutting things out and pasting them up on a board. Did Photoshop get rid of the job of production artist? Yes. Did I love the job of production artist? No. I took that job because I was working my way up to being a designer. 

Clinton: Right. 

Lisa: So, when Photoshop came along, and then I was like, wow, you're going to do all of the production? Like, I'm going to just do that so much more quickly. I don't have to physically cut things out. I literally have scars. 

Clinton: (Laughs) 

Lisa: Like, I'm not going to have to have stitches anymore because I cut too hard through a mat board. All of that stuff enabled designers to express themselves better, without worrying about the tool. 

Clinton: Right. 

Lisa: And so the beautiful thing here is that designers are going to be able to execute their concepts faster, without that production work getting in the way. Now, one thing that was brought up in several of the talks, Forrester is also taking this... we're on this precipice. And it's a precipice right now with AI between, they're calling it magic and mayhem. And right now we're seeing both. So the magic is obviously all the things we hope AI can do, right? It can help you do your production work and it can do all of this stuff. The mayhem is going to be for people who think it can replace a designer or whatever. So, like, the mayhem is going to be more... You know, I caution my designers now when we're talking about it. We're going to see a lot of bad design over the next 4 or 5 years. Hopefully shorter. And I say that because there's going to be a lot of companies out there that are like, great, we don't need a designer. We've got AI. So I can tell, and Figma's got all kinds of new AI tools that they just launched at Config this year. 

Clinton: Yep. 

Lisa: I can just say, like, make me a website for a jeweler that sells this kind of jewelry, and bam, there's a website. Good. It's good to go. But you look at it and it's like, it's not great. And it reminds me a little bit when PowerPoint first came out. And even now. But like, there's some pretty bad PowerPoint design out there. Because you got people that have an easy-to-use design tool, that don't understand how to curate the design itself. And so AI is kind of that same thing. And that's what, when Forrester was talking about the magic and the mayhem, the mayhem side is going to be people that are just using it. And they pulled up a great example of, you know, asking an AI image generator to generate a coffee mug brownie. 

Clinton: Mhm. 

Lisa: So most of us, you know what a coffee mug brownie is, right? 

Clinton: I do.  

Lisa: You... coffee cup and you put it... Well, it generated this, like, bizarre, like chimera of... (Laughs) Of a coffee cup and a brownie, like, merged together. Like, it was a cup made out of brownie. 

Clinton: Which also sounds delicious, by the way. So... (Laughs) 

Lisa: That does sound good too, but you get the idea of, like, there's something in the translation there. And so... I also teach design at Rutgers University. We have the next class coming up in late September. I'm having to completely revamp that because of this. Because all the things I was talking about even last semester, I've got to start adding more stuff around AI, and that stuff has to do with... The designer getting the most out of AI is going to depend on a designer's ability to write a prompt. Because they're going to have the design vision, but that AI is not going to be able to help you execute your design vision if you don't know how to ask it for what it wants. Now, it's a lot easier than, you know, the old days of Google, when you needed the operators and you needed to know how all that stuff worked. 

Clinton: Sure. 

Lisa: But... It is still pretty easy. You, it's just a sentence. But what you ask for, you don't want to ask for a coffee cup brownie. You sort of have to think, okay, you're probably going to misinterpret that. 

Clinton: Right. 

Lisa: How many ways might you misinterpret it? So what I'm going to ask you for is a picture of a coffee cup in which I have cooked a brownie, you know what I mean? It's just, you gotta get a little more detailed. And so, I think designers should embrace this tool, because it's going to help them be great designers, but they also need to up their skills around how to embrace this tool. How to write a good prompt to get what you need.  

Clinton: And I think a lot of it comes down to, call it more traditional education around communications. And in this case, you know, let's just keep it in English because we're prompting an English in this example. But, language. 

Lisa: Yeah. 

Clinton: And understanding how to be very, very specific. You could be verbose or not. That may not matter. But I would imagine the more verbose you are, as you intend to prompt with clarity, you actually could cause confusion. 

Lisa: Yeah. 

Clinton: As opposed to being like hey, sharper, less, more direct, more specific language can be really, really healthy as prompts. And that's an interesting, just, thing that I'm sure, I'm sure, as you're out at Rutgers - and go Scarlet Knights for the Rutgers fans out there, right? 

Lisa: (Laughs) 

Clinton: So with that, though, I'm sure that is baking its way into the college curriculum too. Is like, this is the best ways to to prompt and to get the most value out of it. And you brought up a couple of examples, too. You were talking about, you know, just people making decisions to throw things over to AI and saying, okay, we can get rid of the human. We don't need Paul Bunyan, we've got this thing now. Meanwhile, again, the idea being that the combination is what's going to give you the most lift. And actually, I want to share a story about my mom. My mom was a... I grew up on Long Island, and she was a lab tech supervisor. She ran the labs. So, blood bank, chemistry, phlebotomy. But back in her day, when she was training, and she went to University of Stony Brook, when she was in college, to create the slide samples to go analyze the blood, they had to draw the blood with little pipettes through their mouths. And these were disease-ridden things. 

Lisa: (Laughs) 

Clinton: They were... Still, at that point, before it was like, oh crap, we probably shouldn't use pipettes. Now, of course, there's a whole litany of different tests that are all automated, that are all computerized. But the highest-value thing that the lab person could do is really understand, get under the microscope and help diagnose what's wrong with that human. Like, what can we do to solve... That was the highest-value thing. Because that's what they tell the doctor. They're the ones who bring the diagnosis and say, hey, this is what we found based on whatever we're looking at under the microscope. It did not take away the need for lab, lab techs and people to understand what they're looking at. It just kept leveling up how much time they could spend on the most valuable thing. And I wonder if there's going to be some parallels here with, design and AI. 

Lisa: I mean, absolutely, and that is what it's all about. And even when I talk about, you know, Photoshop came along and made it so I didn't have to spend all this time cutting stuff out, now I could spend time on creating the design. And it's the same thing. And it also... It's not just for the production work, it's that you can, you have this idea and then you can spin off 5 or 6 different iterations of that idea really quickly, that would enable you to then step back and look at these 5 or 6 different designs and think about them. But yeah, the key really is, the human needs to be looking at it. And you use the Paul Bunyan example, which is not the positive. Like, I use Iron Man. 

Clinton: Sure. 

Lisa: As the example. Because Tony Stark with J.A.R.V.I.S is Iron Man. Otherwise, J.A.R.V.I.S is, like, a smart AI, and Tony Stark's, like, a rich whatever. But, like, you don't get Iron Man unless you've got the two of them together. And that is sort of that same thing, of, think about how he is using J.A.R.V.I.S for everything that he does. 

Yeah. 

All the superhero stuff. And it's kind of the same thing, like, use your AI like that. Have it, you know, run the calculations. If you know there's a particular... You know, you're doing an illustration, you know there's a particular curve that you want, or something that you want. Like, yeah, I can go into illustrator and, like, click on all the different paths and, like, extend them, and start to... Or I could give the specifications of, like, exactly what kind of curve I want and boom, it's there. And now I can move on to the rest of the design. 

Clinton: Right. 

Lisa: It's about that sort of thing. Where it's like, I don't need to take the time to draw that out. You draw that out for me. I'm going to put it all together into the right thing. 

Clinton: Enabling vision to come to fruition more seamlessly and more often. Right? 

Lisa: Right. 

Clinton: Which is super nice. So when we have guests on that are at different events... We've had our friend Clemens, who went to Tokyo Auto World, went to CES. We had a guest on recently who was at Dell Tech World, and now you were at Forrester CX. We always, I always like to ask about, like, hey, what about some of the cool use cases? Because you walk around the room, you see different interpretations. You see enterprise-level use cases of, hey, this is what's happening in market. Or if they had them on the main stage. Are there maybe, you know, one, two or three use cases that really stood out to you in that room? 

Lisa: It's hard for me to narrow it down to one or two. Like, my brain is absolutely swimming in terms of the different use cases that they had. I mean, selfishly, the ones that I really focused on were the ones that I felt like I could use right now, today. So one of them that... One of the things that opened my mind. I'm the head of design. I have not had my hands on keyboard actually physically doing the design for a few years. You know, it happens. I miss it some, unless I'm doing, like, my own thing, but... 

Clinton: Sure. I get it. 

Lisa: It was the tools that actually struck me. Because the use cases, you could almost go to any conference and see these use cases, of, here's how we leveraged data to better understand the end user so we could bring those two things together. Here's how we leveraged data in order to get to insights faster. But it was the, sort of, the tools that just felt tailor-made for the annoying things in my world. We've been doing a lot of design thinking workshops, and there's some tools there where... We could use all the tools like Miro and everything that we're using currently, and then AI can come in and pull out, here's your journey map and here's your personas. 

Clinton: Yeah. 

Lisa: You know, you came up with them, but we're going to pull it out. So, things like that were actually the things that excited me, because... And I know it sounds really stupid, tactical, and maybe even a little pedantic, but when we're with a client doing co-creation with the client, I want the focus to be on creating. 

Clinton: Yeah. 

Lisa: Not on, you know, artifacts and outputs. And the faster we can pull things together while we're in there with the client, and then put them up and say, is this what we're talking about? And that's one of the things these tools, from a design perspective, have enabled us to do much faster, is, some of the co-creation kind of stuff that we're doing. You know, and then, in terms of use cases, a significant number of them were the, sort of... get a better understanding of your customer. And give them what they need. Now, the beauty that I liked that is very different from previous marketing, especially when you go on to marketing, everybody's talking about ethics now, and I think that was the thing... So there were people talking about use cases, getting deep, deep customer knowledge. But then there was always that deep customer knowledge, and then pulling that back, and identifying what of that data are we going to use to deliver what the customer is asking for? And how are we going to think about the ethics of using the data so that we're not manipulating people? I mean, I think that's the other thing that is, not just onstage, but the people that I talked to while I was there, is that very real concern about the whole ethical dilemma of things. But not a... Not fear, not fear like, oh, we shouldn't use that, which I think a lot of companies' fear of everything from cyberattacks to exploiting somebody's data is there. But all of the conversations were around, how can we make sure we're using this stuff, quote-unquote, "safely." Because we want to use it. We want to move ahead and we know it's going to be really valuable. So there were a lot of use cases that came up around the ethical use of this for customer data. And some really good pointers on, like, how should you actually be thinking about this when you're using it? What are the stopgap measures? Right? Because when I first came into design, it was... Web design, make it sticky. I want to make it sticky. Well, at some point sticky became addictive. Right? 

Clinton: Right. Sure. 

Lisa: And then you put AI in there, and AI makes it, like, we could really make it addictive if we wanted to. And so, I found a lot of the interesting thing about the use cases were all... This is real. We've got really real ways that we're using it. But we don't have the sort of blind optimism that we have had about technologies in the past that have maybe gotten us into trouble. 

Clinton: Yeah. 

Lisa: So I think a lot of those... Like I said, most of the use cases were around know your customer. Or use cases around, you know, how this enabled me as a designer to do the important work faster. 

Clinton: Yeah. Clearly, talking about ethics there. Where do you join the, where do you draw the line of what is ethical? And do ethics and usefulness... Are they partners? Are they the same? Are they... If they're different, how are they different? I wonder if you have any thoughts on that. 

Lisa: Oh yeah. And that was some of the topics that came up with... I had the chance to talk to a couple of analysts as well, and I chose that topic because of how important it is to me. Yeah. So I mean, first of all, just the fact that you... Just asking the question is a huge step. But there's two ethical questions here. And for both of them, I will tell you that here at Launch, we have a design ethics checklist that we use. And we talk about it, and we've got an escalation path. So if any designer anywhere has anything that, like, feels funny. And it's literally like, if it doesn't feel right, talk to somebody. Like, the buck stops with me. And, like, we'll have the conversation. And there's some, you know, some checklists that you can look through of... So when we're data collecting, are we collecting all the data we could possibly find out about this person? Or the minimum amount of data that we need in order to fulfill whatever we're trying to fulfill? So there's that piece of it. But it is never going to be black and white. You're never going to have a checklist that's like, yep, that's ethical, no that's not. 

Clinton: Right. 

Lisa: It has to be a conversation. And it has to be a conversation with a designer in the room. And I say that because we represent the humans. And so, I think we need technology, everybody needs to be there to talk about it. The cases that are interesting, that come up with the ethical pieces, some of them are a little more clear cut, and people are still not sure what to do about it. You know, I mean, I had an instance during Covid, I had a client that wanted social media data, because they wanted to identify people who were the most afraid of dying of Covid so that they could sell them life insurance. 

Clinton: Mm. 

Lisa: And at the time, the designer on deck was like... (Laughs) "I, um..." You know, and she came to me, and I was like nope, not okay, let's have a conversation with the client. And to be honest, the client wasn't at all trying to be nefarious. They just didn't really think about the implications. 

Clinton: Right. 

Lisa: And that really... A lot of the conversations at the Forrester CX summit were about that. It wasn't about, how do we know when it's ethical or not? It's, how do I have the conversation so we can come to a shared understanding of whether or not we are going to proceed with whatever it is that we're doing, and that we all agree. And look, stuff's going to slide through, right? 

Clinton: Sure. 

Lisa: But it's just having the conversation. But the other piece of it is, of course, the bias that's inherent in the data. That really, you know, revolves around, you know, are you targeting the right or the wrong people for something, right? So is there bias in the data where you're not actually... You're thinking you're doing one thing and it turns out you're doing something else. And, you know, there's all kinds of famous examples with, you know, mortgage applications. And, you know, you rely on historical data. Not the person's historical data, but just historical data. And if you have traditionally underrepresented groups who, because of injustices, were not able to get mortgages in the '50s and '60s and whatever... 

Clinton: Yep. 

Lisa: ...Then you might look back and be like, well, what the AI was doing was looking back and saying, well, like, you're part of a group that has been shown to not get approval for mortgages, so you shouldn't get approved. 

Clinton: Right. 

Lisa: And it's like, well, no, but that's based on biased data. You know, the same thing with AIs that are screening resumes. You know, you ask AI... And they're getting so much better at this. You know, ask AI, you know, what does a doctor look like? And if you're not really specific, you're going to end up with, like, a white male in his 40s or, you know, whatever. AI is getting better with that. But it also, then, it applies that same thing to what kinds of backgrounds, what kinds of schools, what kind of names do people have that in the past have been very successful at getting hired, because then we think that will be the right criteria. So, those are the things where, let the AI do the editing, but then the human needs to look at it to say, what did we miss? 

Clinton: Right. 

Lisa: Did it edit something out that it shouldn't have? Did it add something it shouldn't have? And all of... it's that combo. And this whole summit, to get back to that, was all about that. It was about those two things coming together. When, where, how, when do we talk about... You know, when do we raise the question of ethical, and it's like, it's always. We should always be talking about it. 

Clinton: And you had a podcast just on that topic as well, on the ethics around AI on Catalyst. 

Lisa: Yep. 

Clinton: And I would encourage listeners to go dive into that as well, because really, really great stuff. But I think the big takeaway is, hey, like you said, I love that idea of like, look, have the designers in the room. They are the representative of the humans that are going to be using the technology. Doesn't mean the product engineers are not thinking about the human at all. And with that, though, they are not usually tethered with, or like, it's not their job to understand what the human really desires, and design for it. 

Lisa: Right. 

Clinton: That's just not what their role is when you're building technology. It's not good, bad or ugly. That's just the, you know, how it's usually set up. However, we believe that to get that mix right, hey, have product strategy. Have designers, have engineers together from day one. And if it involves AI, like you said, have AI there day one also, so you can start looking at some of the outcomes it's producing. And like you were hinting at, or saying, if you're seeing things that feel a bit off, then, well, call it out early. You know, like, talk about it. 

Lisa: Right. 

Clinton: It can go both ways. Because recently, while there were biases in historical data, because they just are there, then you have a bit of, like, an overrotation where people were asking for, like, imagery of George Washington, and it doesn't come out anything looking like... 

Lisa: Right. 

Clinton: He's a historical figure, but we know exactly what the gentleman looked like because he's been painted, you know, on horses and everything else. And you get that overrotation. But if a human's sitting there going like, hmm, it's not quite right, AI. 

Lisa: Right. 

Clinton: That combination will still correct it. 

Lisa: Yeah. And look, and to be clear, the human plus AI isn't just because of the ethical thing. 

Clinton: Right. Of course. Yeah. 

Lisa: That just happens to be, like, my personal crusade. It is more about the right thing. And speaking about, like, what is the right thing to actually build? And how can AI help? 

Clinton: Right. 

Lisa: Part of the reason it is so important to have, in particular design and technology together in the room in the beginning, talking about it, is because each of us is going to challenge the other one as to what is possible. 

Clinton: Mhm. 

Lisa: You know, and designers may even self-edit and have some idea that they think is impossible. But if there's a really good technologist in the room that really understands AI, they'll be like, hey, that's possible. 

Clinton: Right. 

Lisa: You know, it isn't just the designers. Everybody's representing the right solution. 

Clinton: Yeah. 

Lisa: And they're bringing slightly different perspectives, getting them all in together. So, that is where, in particular with AI. Because there's so many pieces of what the technologist does, what the designer does, what the product manager was, does, with the project. There's so many pieces of those, each of which could be replaced by AI. And is being replaced by AI. And so it is making sure that for, not just for my discipline, but for engineering's discipline and for any of the other disciplines, that there is a human there thinking about how we're using it. And that's the other reason it's important to get them all in. 

Clinton: Yeah, very cool place to end it for today. So, we've been chatting with Lisa Woodley, the Head of Design at Launch by NTT Data. And again, just to recap, Lisa was recently down in Nashville for the Forrester CX event. This entire event was infused and permeated with AI and the ways in which, you know, designers and those that really focus on the human should be thinking about how to use it going forward. Not to... Not to be scared of it, not to, you know, bury your head and ignore it, because that ain't going to work. But really understanding how to work with it in that future state, symbiotic. When I say future state, it's actually happening, as Lisa said, right? So Lisa, anything else you'd point people to? You know, great articles or things that you're reading up on, books or anything on this topic that you really, really gravitated towards? 

Lisa: Yeah. So right now I'm in the middle of Unmasking AI, which is Joy Buolamwini's book. So, she was an MIT, or she may still, I think she still is an MIT researcher. She was one of the founders of the Algorithmic Justice League. In fact, the subtitle of the book is "My mission to protect what is human in a world of machines." So that is... There's lots of stuff out there. This isn't a how-to so much, this is more of a hell yeah, as I read this, than a how-to. (Laughs) But yeah, you know, and the other thing is, I have... There were so many tracks at Forrester, I couldn't attend all of them, right, because they had multiple tracks. 

Clinton: Right. 

Lisa: So I'm honestly in the process of going through the attendee site and, like, looking at all the stuff that I missed. 

Clinton: Yeah. 

Lisa: And also just, you know, getting my team out there. But yeah, I mean, it's... There's a lot of different things out there with the tool. Anybody who attended Config, which is Figma's conference, that was all about AI too. So it's not, I'm not so much reading about the tools. I'm more reading about, like, the implications and the trends and how we need to be thinking about it. 

Clinton: Awesome. Yeah, well, we'll definitely have you back on the Catalyst podcast to keep talking about it. Because, you know, like we said, this is not froth, it's not hype. 

Lisa: No. 

Clinton: And it continues to accelerate. It's just going to be this blended future of people and AI and machines. So we can work more effectively. So, harnessing that is incredibly important. So Lisa, thank you so much for joining us on this, or leading this great conversation today on Catalyst, where we like to say that we believe that fast will follow smooth, and aiming to create digital experiences that move millions is a very worthy pursuit. Join us next time as the pursuit continues on Catalyst the Launch by NTT Data podcast. 

(CATALYST OUTRO MUSIC) 

‍