Accelerating Your AI Adoption Across The Business (Guest: Mary Purk)

Andreas Welsch: 0:21

Today, we'll talk about accelerating AI adoption in your business, and who better to talk to about it than someone who's seeing this from both actually industry and academia. Mary Purk. Thank you so much for joining.

Mary Purk: 0:35

Thank you, Andreas, for having me. Really appreciate it.

Andreas Welsch: 0:39

Awesome. Hey, why don't you tell us a little bit about yourself, who you are and what you do?

Mary Purk: 0:43

Sure. I am currently the executive director for AI and analytics for business at the Wharton School at the University of Pennsylvania. I've been there for four years and the way I've received that great honor to be there is I've both been in industry and academia. I have big consulting experience with Accenture and then data and analytics experience through Nielsen and information resources. But I did do a stint in between that at the University of Chicago. And ran a marketing research center there. So I do know how important it is to bring academic talent with industry to solve the current business problem. So I'm really excited to be in that intersection and be here today to talk to you about AI and analytics.

Andreas Welsch: 1:31

Fantastic. Thanks again. It's great to hear you have such a wealth of experience and I'm sure we'll have an interesting show and episode. So for those of you who are just joining the stream, drop a comment in the chat if you are already using tools like generative AI ChatGPT and so on, and what do you use'em for? But Mary, maybe should we play a little game to kick things off?

Mary Purk: 1:54

Okay, let's play a little game. What's it gonna be?

Andreas Welsch: 1:58

So this game is called In Your Own Words, and when I hit this buzzer, the wheel will start spinning. And when they stop, you'll see a sentence. I'd like you to answer with the first thing that comes to mind and why. In your own words. And so to make it a little more interesting, you'll only have 60 seconds for your answer. And for those of you watching this live, drop your answer and why in the chat as well. I'm really curious. Mary, are you ready for, What's the BUZZ?

Mary Purk: 2:27

I am.

Andreas Welsch: 2:28

Okay, then let's get started. If AI were a fruit, what would it be? 60 seconds.

Mary Purk: 2:39

If AI were a fruit, I think it would be an apple. Only because when I was at school, apples always remind me of education. And teachers are some of the best role models that I've had in my life, especially when I went to Montessori school. And so it would be an apple, because not only would it teach me wonderful things that I can learn, but also it is healthy and there is a wide variety of them.

Andreas Welsch: 3:08

Fantastic. And well within time. With the ice breaker out of the way, let's jump right into the first question. Maybe we start with the most obvious,one because we titled the episode around accelerating AI adoption. But I'm curious, what are you seeing? What are maybe, first of all some of the common challenges for AI adoption in business that you see?

Mary Purk: 3:30

There's a couple of different ones, but I'll highlight maybe four of them. And this first one, it's like beating a dead horse, but, I think it's the most important challenge for AI adoption. And that is, identifying a clearly defined business problem for AI. We have to do that all the time. At our center, when we're talking to companies, we spend at least three or four weeks just explain to understand what the problems that they're trying to solve and dissecting that. So many times, you might decide you're gonna look at a problem, but the second part of that is to also look at the data that you need to solve that problem. And in doing that, you'll have to look at and see if the data is biased or not. As most companies and individuals, they say, I wanna use AI. You can't use AI to solve your business problem unless you have data. So it's one and the same. To use AI, you need data. So the keys to having AI adoption would be the business problem to solve for a really clearly defined business problem, the data, and then two other things are your team, you need a cross cross-disciplinary team to do that. And you need IT and marketing and finance and operations. It's multidimensional and it's very complex. And then finally, what are your use cases? What is the use case you're going to solve this for within your organization? And then if you solve for, can your company implement it? Because why would you solve? Some people just wanna solve to solve, but why use all those resources to solve, if you know that your company couldn't necessarily take the solution and implement it? So the keys are making sure you know what that business problem is. That you have the data and you have enough data to solve it. A multidisciplinary team. And then what the use case is, and can you in fact use the solution and implement it in your organiz?

Andreas Welsch: 5:35

Thanks for sharing. And to your point, it feels like we constantly need to read. But it's very good to hear also from you, these are the key things that leaders need to focus on if they want to have their AI initiatives succeed. And it also mirrors what I'm seeing others share as well. So very good to see how well aligned that is in keeping an eye on the chat here. People are still answering with their favorite fruit. If AI wear a fruit, it could be watermelon. Yes. And, Caryn's response about the durian. So why don't we move on to, our next question. I know when we had our preparation call for this session, we also talked about generative AI and that there's basically no way around this at the moment. And it really feels like there's so much talk in the industry about ChatGPT, generative AI. I see that a lot of business leaders are asking their AI teams to find that holy grail, that use case that really makes money or save a lot of money. But I'm wondering, what are you seeing? What can leaders actually learn and apply to that explosive interest around generative ai?

Mary Purk: 6:49

Just prior to what you were saying too, is, what do leaders really have to zero in on? And in terms of the use case, I think I heard you say that ChatGPT what is a use case and with that is remembering that you don't have to solve it to a hundred percent perfection. It would be best to solve almost to 80% and pilot it and see what that adoption is. And then, do a test and learn, test and improve. That is also really important. The only caveat I have to not having it be a hundred percent applicable or correct, is that you have to make sure your data is not biased. And, that you have to almost over correct and make sure that's why you also have to have a multidisciplinary team to make sure that data is not biased. These things sound really I knew that I knew that. Well, why but why are they being said? And then know the motivation for that. The motivation for the multidisciplinary team isn't cuz it's so popular to do right now. It's a fact. You need it. That's your insurance to not put something in the marketplace that all of a sudden blows up and you find out that it's so biased and then you've lost half your customer base. Your people are your insurance to making sure that you can have successful AI adoption and revenues that would come from that. Okay, so now, leaders are. We have so much dinner with AI ChatGPT, what do we do? We're behind the eight ball and the bottom, the most simplest expression I can use is: get on board. Get on board. It's like you cannot be on the sidelines. You have to embrace it yourself. I would encourage you to open a ChatGPT account and experiment with it. If you haven't, just set aside some time on your calendar to do it. And then have some of your significant others or other people in your lives encourage them to use it. The more people are using it, it will help shape it for the good in our society. I think I heard like over it's the fastest growing app we've ever had. There's over what, a hundred million users on it, and it's gonna change. I would say get on board. Invest in it. As a leader, you can put out into your team use ChatGPT to solve a pain point that you have in your process flow for the week or a task that you have for the week. And then share that with your supervisor. It's do it like, this is an exercise and then as a leader you're showing how you're current. This is important to us. We're going to experiment, we're gonna discover, these are all the things you can do with ChatGPT. Now's the time to do that. So that's what I would say as a leader. If you haven't opened an account, open an account, use it, and then use it personally and use it in the in business.

Andreas Welsch: 10:04

I think that's a fantastic call to action. And the reason why I think that is because it's also so much more accessible now. Compare that to six, seven years ago when we were just climbing that hype cycle and everybody was getting excited about machine learning, AI, self-driving, cars, flying drones, packages delivered by drones and all that stuff. Now you can actually touch it and feel it. And you can feel it in so many different ways. Again, not just image, video audio transcription, summarization of text, generation of text, and so on. So I think that's an excellent call to action to get onboard.

Mary Purk: 10:40

And it's fun. There's discovery in it. But know that ChatGPT is a function of what everything that we as humans have put out on the internet, all of it. Most of it's true, some of it's not true. So we still have to use our own instincts and other knowledge points to say we might need to change that. It's not all fact. We just have to remember that. And then as we're reading things or hearing things, we are going to have to use our own filtering to know what is good and what maybe we have to take out that we don't necessarily believe. And you might have to use other resources or other sources to validate

Andreas Welsch: 11:23

I'm looking at the chat. I see one message from Ken who says, you mentioned the notion of a multidisciplinary team of people. So do you think that we might soon see multiple AI used to cross-check each other's recommendations?

Mary Purk: 11:39

Oh, sure. I, would think that. We talked about this, remember? Everyone's so excited about using the application. We're so busy discovering it that some people have looked forward, but this is a very practical use you just brought up. There will be those who have invested early and understand some of the capabilities of ChatGPT who actually develop applications to further verify specific things that come out of it. It could be how there's different applications for a dictionary and Grammarly and things like that. There might be FactCheckerGPT or something to that effect. There will be different splinters within ChatGPT. They'll be even more specific and more narrow in what they do.

Andreas Welsch: 12:28

Fantastic. I think that's a very good outlook and a good summary. Can you share some examples of what that collaboration looks like that's successful between business leaders and business teams and technology teams? Just to get more adoption. Maybe an additional question to that is generative AI all of a sudden so much different compared to what we were doing on November 29th, before ChatGPT was released.

Mary Purk: 12:58

I'll take more of a simple approach to this in terms of the collaboration between business and technology that foster AI and machine learning. There might be some people that remember, but we needed like CDs and we needed albums and stuff, but then came along a company called Pandora and they decided they were going to offer many different songs to people. A whole library of songs that were gonna be available. And then it was going to be customized to your taste. But there was no data around that. So they then scraped all these songs and provided all these different qualifiers for songs so they could create a library. Very pure data set that then provided very good recommendations for individuals. So that just was an explosion in the music industry, because they saw that need and that personalization. But a lot of that was dependent on business needs and technology to create that data that was needed. Then there was a competitor, Spotify, that saw that and they said, we're not gonna do that. That was way too time consuming. But it's a really good idea. And they then slowly grew their own dataset through uses, through people using the data. And then that dataset grew and that's what's gonna happen with ChatGPT's. All this it's gonna kept getting bigger and bigger as we all are contributing to the data. And the bottom line is for both of those organizations, they had people that understood what the end user needed, but technologists that understood data. We can't talk about generative. AI without talking about data. People really have to realize data fuels the AI algorithm. And so as we're talking about that, after we get done with all the novelty, we're gonna start spending more time on understanding how that data's gonna be collected. Who's gonna be continuing to contribute to the data? How are we gonna filter out truth or untruthful data? And so that would be one application of AI and ML that brought a whole new industry to us. We also know what's happened with Amazon, good things and bad things. Amazon's able to feed us even better things that we might want from our previous buying history or previous browsing history. But then they also got in trouble because of some hiring recommendations when they they only used their current data. And they didn't even think that maybe it was biased towards one gender or one type of skillset. And they're like, oops, guess we have to expand it. So that's why you need both business and technologists to look at those algorithms. But like I said, both of those examples I gave were related to understanding what data you needed to get the best output. Bad things into a process, possibly bad recommendations. So I think just keeping that simple analogy at top of mind is also very important for both sides of the teams of the business and technologists.

Andreas Welsch: 16:04

I see some questions here, and I think Ramnath is asking did Microsoft open the floodgates too quickly without guidelines on GPT-3 and these kinds of technologies? But building on your point around data, right? You see this concern of ChatGPT and this kind of technology answering the user in ways that might be creepy, or that might be perceived as being creepy or, just inacceptable in so many different ways. And then when you look at the type of data, the corpus that these large language models have been trained on, hey, if it's scraped off the internet in publicly available forums in and sources. Then it's also a mirror and a reflection of how we as humans and as people communicate and the biases that we hold. Whether they're conscious or subconscious or explicit or implicit. But I see Mike has an interesting one here. He says, what are your thoughts on build versus buy AI in the enterprise? What are the commercial factors that will govern the decision in an organization? And I can see this applied to both before generative AI really popped in now after.

Mary Purk: 17:17

I'm, from the Midwest. And I'm a mom of four, and I'm the oldest in my family. So I am super practical. I am super practical. Time is of the essence. I'm a busy person and so people always ask me to do something cuz I guess a busy person always knows how can get more things done. And I would say, as you can see where I might be leaning, if there is an application out there that meets your. needs. I would choose that AI application before building it in-house. You obviously have to vet the individuals or the company that's building that AI application, but they are that much more ahead in their technology and knowledge. And then the key that the enterprise brings is making sure that you have the right people evaluating that. And they have not only the focus of what the current situation is in the marketplace for them, but looking forward and that they can clearly define what the business problem or pain points are that they're using AI to solve for. So that's where I would go. I think there was someone like, what about in the regulatory industry? There'll be certain companies that cannot buy anything off the shelf and will have to build in-house for security reasons. And I would say in other instances, it will have to go that route.

Andreas Welsch: 18:35

That makes sense. I think that also brings up the next question and the last one for us to end on today. So if we think about AI adoption in the enterprise, and there's AI, now there's generative AI. There's still this buzz and this interest around AI and how can we get value out of it. How can we get more adoption in business and IT leaders should work together? What do you think should they-business leaders and IT leaders- focus their teams on in the next six to 12 months in setting the expectations when it is about AI adoption in business?

Mary Purk: 19:14

The first thing, and I think we started out with this, is experimentation. I think that you have to start out with experimentation with your team and you as a leader tell them and be forthright. If you aren't feeling this, then don't say it. But you know that you wanna experiment with ChatGPT in in your processes and challenge your team to come up with the best ways to use generative AI or if it has to be built in-house or purchase or such. But you can use ChatGPT to simulate that. I think that would be the first thing. So experimentation, second data. We've had data engineers, data scientists, but every single person on your team in the organization needs to know you are a data steward. There's certain things we're compelled to do. If someone falls down, we go and pick them up. We need to teach people. This is also about data. If there is something bad going on data, they have to be able to speak up about it. It's in kin to be a whistleblower, but you just have to bring these things up. If you're at that table and you see something that's not right or something that's really good, it's up to you to speak up. So I think make giving them the permission to do that. And then I think for me, the glass is always half full. You can make lemonade out of lemons. It's a great. Time to be in with technology. There's so much that you can do and it's a level playing field. I learned FORTRAN and COBOL. That's not relevant. Guess there's some programs still in COBOL there's R and Python now. That's gonna be obsolete soon. But this, technology, I don't need to know how to code. I need to know how to write really good questions. Maybe really good logic. If you could say anything people need to know how to logically put together a story or a business problem to solve. That's what you need and you need to be a good editor. That's what we've been hearing, right? We don't have to be great writers, although I love writing. But you got to be a really good editor and know what you wanna communicate. So giving people really good, positive attitude to go and explore and discover, but they're responsible for contributing and their job will be there if they learn to use AI and use these tools.

Andreas Welsch: 21:25

I think that's a very encouraging call to action also to not only people who are in a formal leadership role, but actually everybody. Go try it out. Experiment, learn and become part of the discussion so that you can have an informed discussion as well. Maybe as we're getting close to the end of the show, can you summarize the three key takeaways for our audience today?

Mary Purk: 21:53

The three key takeaways. Remember what fruit you thought AI would be. I think that's important. And you can tell that at dinner tonight or this weekend if you're going out. I think that'd be a great conversation starter. Thank you for starting us off that way. Two, if you don't have an a ChatGPT account. Open it. And then, if you're a leader or you can influence your team, challenge people to use the technology. And thirdly really I think to me, data has always been very close to my heart in my career. And I think it just never goes away. So I think making sure that your company is doing better job of creating data stewards throughout the company. And if I could just add one last thing, is you don't have to get it a hundred percent correct to move the the idea or technology out of pilot, maybe into where it could be tested further. So if you do wait till it's a hundred percent correct, the opportunity might pass you by. So risk this, risk reward. I'm not saying things are gonna be oh easy now that we have ChatGPT, but it's gonna be a lot more interesting and you're gonna be able to have more free time. I'm looking forward to creating my own personal secretary now with ChatGPT. So those will be my summary points.

Andreas Welsch: 23:16

Fantastic. Thank you so much for sharing. So folks, we're getting close to the end of the show today. Thank you so much for joining us, and Mary, thanks for sharing your expertise with us.

Mary Purk: 23:27

Oh, thank you for inviting me. It was really a pleasure.

Andreas Welsch: 23:30

Fantastic.