Episode 44: Resilient Leadership and AI in Regulated Industries - Erika Kiely

Hello again and welcome to another episode of SphereCast. In this mini-series of CTXcelerator, we welcome Erika Kiely, powerhouse in tech leadership, whose journey from surviving political unrest in Peru and leading a large-scale technology initiatives in the insurance world is nothing short of remarkable. So, Erika, welcome with us, and in this one we have also Mario, Mario Schwarts, our Data and AI Director at Sphere. So, he will be also my host today in this podcast. Erika, let's start first with you. So, I love that people introduce themselves because they know themselves much better than I do. So, can you give us a little bit of insight who Erika Kiely is?

Absolutely. And, Adin, Mario, what a pleasure to be here today with you. I appreciate this opportunity to voice a little bit of the thoughts and the journey that has brought me right here.

Originally born in Peru, and from the professional side, I have been leading technologies efforts and program efforts for the past 20 years. Started a little bit with the process improvement side, and after that, exited a Fortune 100 company not long ago with 20 years of a great, great journey and leading all the international technology and all the international programs. And we were able to be pretty successful with the deployment in Latin America and the Caribbean, which are the markets that we are serving.

Right after, I just decided to find my own voice, and that is called ProTechtive Strategies, which is really the soul of my journey, and is a consulting company that focus on technology, focus on regulatory and compliance matters, very tied to AI because the thought process is technology has to go along of certain guardrails, and I think merging those and ensuring that there is a synergy is a great value that we can bring to the table.

So, quite a bit of the journey on the professional side, but at the end of the day, I'm a human being. I'm a mother, I'm a friend, I'm a wife. And I am someone who is deeply hopeful about improving the status quo, especially of the people in my country.

Well, thanks for the intro, Erika. To continue on that intro, your story is deeply inspiring. So could you share with our listeners what shaped your early resilience and how your upbringing in Peru influenced your leadership style today?

That opened the door of a very intense phase of my life. Growing up in Peru was very challenging. During the era of the '90s, Peru was a country that was heavily run by the corruption on the politics. We were hit by the terrorism with the Shining Path, and then the desperation of the people who doesn't have much caused really to not have any character or any values to follow, and that was the perfect combination for the tragedy that my family went through.

So back in the '90s, my father was running for Congress, and as the result of certain decisions, my dad and his team and another candidate, they were assassinated. Back in that time, I was only 14, and I realized that at times there is not a limit for when man looks for power so desperately, thinking that that will be the essence of realization.

So, recover from that experience, I don't think that will ever end. I think recovering is a continuous matter. But the experience that I went through didn't diminish me, nor did define me. Actually, it caused to shape a way to lead people in the way that I do today, and I think the keyword is resilience. And resilience has changed the meaning along the way. It's no longer survivorship. I think resilience is the posture that we have when we are going to any challenging situation. And the resilience being paired with compassion and empathy can move masses and can cause transformational effects. And not only for my recovery, but also for—it's a transformational effect for the people who is around me.

And that has been translated in a leadership model that has generated terrific results, because when we bring resilience to the table, we don't pause. We take some risks, realizing that not everything will go right, but we have the capacity to quickly pivot, get back up, and try one more time.

Well, thanks for sharing that. Yeah, definitely sorry to hear about your dad, yeah. So, Mario, please go ahead, yeah.

All right. So, evolving from that, why did you launch ProTechtive Strategies, and how did your experience shape the mission of this beautiful company that you have put out there?

I had a terrific career at a Fortune 100 company, and it happened that when I entered this business, I was able to join the area that serves Latin America and the Caribbean. So I was away from home, but not necessarily away from home. I was able to serve the people in Latin America and the Caribbean, especially coming from the experience of understanding what happened when someone passes and there is not life insurance. That made it very real. So I poured a lot of heart, and when we pour heart, we also pour a lot of energy. So the past 20 years was a lot about continuous improvement.

My initial degree is in journalism and communications, and midway, I realized that technology is critical to have technology knowledge. And I started getting very close to matters of continuous improvement. So let's talk a little bit of DMAIC, Lean Six Sigma, and it felt very natural, very organic to the point that I was offered the opportunity to run the technology and the programs. And for a company that is also emerging, because having a successful business, the next step was to diversify and expand.

And that was the past three years in that company, and the greatest accomplishment as a professional was to lead a full implementation of another insurance company that is in the international private medical insurance from zero, from ground zero all the way to launch. What that means? That we talk about regulatory matters, compliance, marketing, operations, systems, integration, security, vendors, and it was just like the greatest gift, but also the greatest challenge, because it forced me to learn very quickly about every single workstream that needed to be implemented in order to launch this company.

When I reached the final point, I had to take a little bit of time to pause and reflect what is next. And the "what is next" came back as a little whisper from the time when I was 14, because growing up, I had the best leadership and character model very close to me, and that's my dad. A man who sacrificed his life for his country, and that means there was no limit, but also there was purpose, mission, and value on that.

And I do have a young daughter. She's 20, and she is watching me. And I think that's the greatest gift that I can give to her, not what I say, but please, I hope you pay attention to the decision-making and hopefully to emulate.

So, 20 years in that company reached a point where I had to reflect, what is next? What brings purpose to life? And for some reason right now, the Maslow pyramid is starting to hit. We go through many phases, and transcendence is what the answer that I got. And I think that's what is ProTechtive Strategies. It's transcendence. It's no longer about what are my personal pursuits, what are my professional pursuits, but what I can give back: give back to my community, give back to my country, and hopefully to establish a legacy for my own child.

So, ProTechtive Strategies really is the vision that I have for what technology implementation should look like. The speed is incredible, but we definitely need to bring the compliance mind, right? We need to be ethical about that. But also we want to make it accessible. So ProTechtive really is about caring and protecting and guiding, because sometimes we make technology so complicated and we think that it's going to be so expensive, and we forget that that also is paired with innovation, and it doesn't have to be expensive. But when we sprinkle a little bit of creativity, we can do a lot with very little. So really that is the message, and that is something that I want to translate for home.

Nice. Nice. All right. So, let's talk maybe also about the National AI Program you're leading, Erika. And I would love to understand, and our listeners I think as well, how does it reflect your values on ethical innovation? And what is the vision, of course, behind the AI program?

So when I reached that pause that I just mentioned, leaving a great company, and I left at the highest possible level. Many people will stay. I wanted for more. That restlessness was there. And I do believe in the law of synchronicity, that when something is ready for us, if I am paying attention, it will be revealed, and that's what happened.

As I was building ProTechtive Strategies, heavily, heavily powered by AI, I had an old friend that contacted me. We both went to the same university back home, and she started sharing about the first Peruvian Chamber of Artificial Intelligence. And the more that I was listening, the more I was connecting the dots, and I kept thinking, "I have done that. I have done that. I have done that." And when she was sharing is, "We would like to do that. We would like to do that. We would like to do that." And it was just pretty organic, and I said, "Can I please have the opportunity to meet the group?"

And when I met them, I shared a little bit of my journey, what I was able to do with artificial intelligence, how—I think the "how" is sometimes the big mystery. And I was able to explain that in a way that was so easy to understand that became appealing. And in no time, I became a member of the Chamber, which is not just a subscription. You have to go through a screening process, through an alignment process. It's really an honor and as a privilege, and so grateful to the Chamber, and so grateful for the opportunity to be at service.

And as the conversations were taking place, we were trying to align the level of scale with the workstream, and also the vision, the availability, because we are volunteers, and we are professionals, some of us in different parts of the world, some of them—the majority in Peru. It's a multidisciplinary group. And one of the directors said, "Can you join us because at the enterprise level, we would like to have like a model, something that can be attractive for the companies." But it was kind of a high-level thought, and what I did is I grabbed it, I studied, and I said, "What about an AI maturity model?"

And from that, we needed to bring the enterprise piece. Is, okay, we can evaluate artificial intelligence, but what is the reason to tell you you are in a low, mid, or high level? That is not enough. It's not enough in a country that has the responsibility to generate opportunities for its people and create more work. So what I did is I grabbed that initial thought, and I coupled with a balanced scorecard, and I coupled with the KPI and ROI, and I built a house. It was kind of, okay, so if we're thinking to bring a model, what do we give back to the companies? It's just not the initial road map that says, "This is where you are, and good luck." No, we want to take full advantage of the multidisciplinary group that we have here, and build a process from beginning to end.

And it starts with phase zero, with a 101 with the CEO and said, "Okay, I have studied your financials. I know this is the toughest part. We can offer a model, let's start with a diagnostic. We can provide a road map. Let's find a POC, a proof of concept that can be implemented quickly and with very low expense. Demonstrate the power of AI, see how it moves the needle and the balanced scorecard, how moves the needle in the KPIs and the ROIs, and be able to just say, this is what we can do." What is key here is to demonstrate that with a very low budget. And I think because we're talking about a country that's still in development.

So the vision is the CEPIA, the Chamber of Artificial Intelligence, wants to be an ethical reference, but also wants to be a source that can propel opportunity through AI, removing the fear concept that makes people believe, especially in our countries, that it will take my job. Instead of that, to make it accessible, to make it appealing, and say, it's a matter of creativity, it's a matter of innovation, it's a matter of asking the right questions. That is how the model took place.

So following that, because this is an intense story, right, and I think our listeners would like to really get from you, how do you use it? Because that's what people say, "Yeah, yeah, I see it, but how do you get to use the AI in a real enterprise setting?" And not only that, is AI only GenAI, or is it also machine learning, is it also the other pieces of, you know, really artificial intelligence that people just now only relate to AI? So if you can share some examples of how that added value, I think that they would appreciate very much that insight.

Absolutely. I think the use of AI now is "generative" is the keyword. That is what people interpret as an AI. We started using AI, goodness, 12, 13 years ago with the group that I was leading, and it was a group of three innovators, two people in process improvement, and one of my key engineer. And we started this applying to an enrollment, an electronic insurance enrollment model, machine learning.

So comes from very long time ago. How did we do it? We were planning to—were developing a new product, designing a new insurance product, needed to be offered to Latin America and the Caribbean. At that time, we remained a very small group, and let's go ahead and bring resilience, necessity. We get very creative, and I said, "We have a tiny group. We need to be mindful about not burning people out with excessive hours. How can we help each other?" And that's when we learned about machine learning.

So one of the uses, very practical use, was, number one, what are all the rules for this new insurance model? Document everything and start writing down what will be the questions, and how do we grade the questions, and how impact the final process? That was our first touch with machine learning. Developed these huge engine rules with questions, possible answers, and what will be the rating model. So the moment that I will complete my insurance application, in no time it will tell me, "Ooh, you got approved, you got approved Preferred, you got approved Standard, you got postponed because we have to wait six months until you recover from this condition." All of that was powered back in the day with machine learning.

As time keep advancing in order to sustain that business, because we continue growing and always with the challenge to have only a few of us working on this endeavor, the next step was how do we improve all our workflows? That was another effort. So we transitioned from manual, from paper, from scanning to have different type of technology that instead of us reading a medical examination or the underwriter reading a medical examination and taking three hours, having scanning of the full data, transfer the data, knowing that a bare minimum of quality control was needed, that is when the human is so key because we need to quality control the output.

And what we did is we pretty much retired legacy systems, and we also automated a full workflow full of machine learning, some of artificial intelligence with the bots, which we scripted, because being an insurance is highly regulated, the use of the LLM was pretty limited. And we were able to reduce cost in that way.

But I think maybe one of the key examples in that journey was also the remediation that we did for this insurance company. When we're dealing with insurance, so many of these products accumulate cash value. And then we were subject to an audit, and when an audit comes, the questions is, what is your KYC model? How does your AML program work? And we were so early in our maturity level on those matters that we had to work on a full remediation. So the first thing is was to think about resources, and I had around 17 college students joining my team, and we pretty much wiped out the entire data. Over 7,000 accounts, we were able to remediate.

But as we were remediating also with those brilliant minds, we start creating, how can we make this sustainable in the long run? So we created automated flows, how to capture the data, how to standardize the data, and how the data can propel me to make the right decisions. That was phase one. Finished that, completed that remediation, the program, the AML, anti-money laundering, Know Your Customer program kicked.

And then even with all that automation that we did with the few tools we had was not enough to sustain the new business, the new incoming business, and also the new company that we were trying to implement. So we decide to partner up with a well company in the United States that does that automation and the monitoring, the financial monitoring of the premiums, transactions, patterns. And that was another opportunity just to bring the power of AI, heavily powered by business rules.

I think that is when we cannot discard the human being from technology and artificial intelligence. That human being needs to be able to connect the dots, ask the right questions, provide the right mapping, and then improve all those artifacts with artificial intelligence. So those are a few examples.

So I would say in short, before I pass the baton to Adin, in your examples, what you're saying is don't fear for your job. It's really use AI to assist to make you better at your job, right?

Absolutely, absolutely. I think the key for us has been AI is a capacity increase tool. If you are right, you have the right skill in the right place and you want to multiply the capacity of that resource, really, I have no way to measure what AI has done for us. And if we just bring ProTechtive Strategies back to the table, that full company has been built with AI tools, but with a human heart and with human mind.

Can I go one level deeper onto that? I think this is very interesting for our listeners, Erika. So, and Mario, you can also add your comment maybe as well onto that, but what do you guys think is the skills and the foundational knowledge that needs to—every leader should have before they want to integrate effectively AI into their teams? I think many people ask themselves because we see that from many events that we do, that people have the AI on the road map, but they don't know what, where, how, and what to start with, basically. So, Erika, can you share something about that, and Mario, then please share also your opinion as well.

Sure. I think as a leader, the first thing is I have to be curious. I cannot just delegate. For me, leadership is synonymous of character, and it's not just me reading the financials and the ROIs, but I have to be very curious.

So I am spread very thin, and I consider that a problem of abundance. And so with that in mind, okay, I am leading my company, I have pretty limited time, how can I make my time worth, and curiosity. So I think the first thing is just try it yourself. What does that give you? Okay, we spend three hours reading a financial statement to make a couple decisions. Why can I not just grab that financial statement, put it in a secure AI bucket, and ask the right question? And I said, "This is where my company is. This is what I think the company is. This is what the financials is. Can you please help me to understand, give me a summary, give me the five highlights, and how will that impact the current status, and what decision-making do you suggest?"

If a leader doesn't try, it's still very intangible, and it's still something that maybe an engineer will do it. And I think that is the other thing. Engineers are wonderful building the models, but really my purpose is to bring the model to the user and say, "This is absolutely easy to do, but let's go ahead, and you and I, let's work together and asking the right questions, bringing the right context so we get great output that will help us to make great decisions and then game on."

Well, thank you for that. Mario, what do you see talking to many, many clients and prospects?

What I see, I see with Erika. I mean, a lot of what starts and on what makes the success, you know, being on that 20%, because Erika, as you know, 80% of these initiatives with GenAI fail, right? So how do you get into that 20%? And the challenge with that 20% and the factor you see with them is that they are curious. I agree with Erika on that end.

The other piece is a lot of the success comes on understanding that AI is not going to substitute people. They're going to make them more efficient.

And third, but not last, I would say, is they're looking at doing not just a POC, but more of a—take the proof of concept to another level of a real MVP. So a product that can be implemented quickly, because if you just do a proof of concept, depending on how it needs to be, you know, Erika touched on security and other aspects of it, usually you don't care about all those things when you're just trying to do a proof of concept, and then incorporating that to a POC is a big task. While if you're designing as a product, as an MVP, you take into account all those things. So helping understand how that journey is, that maturity, those leaders that understand that they need to embark into understanding the level of maturity will be more successful than the others.

Interesting. Thanks for sharing that, Mario.

And Mario, let me go ahead and add a little bit, and this is experiential. So while I was still with the prior company, the whole generative AI boom start happening, and start happening very quickly. Part of the success of this company has been early adoption, quick adoption, but responsible adoption. So the fact we couldn't deny the fact that AI was starting to get everywhere.

So we had to put one more time a pause and determine, okay, where would we put the guardrails? And what we did is do a full study of all our policies, procedures, and processes, and start to adding just simple lines: "Okay, if a vendor comes, decides to be part of our ecosystem, we need to add additional questions. Do you use AI? Do you use our AI to train your models?" But it had to be a very quick adjustment, change management for sure, because we were just—it goes so fast that if we miss a beat, we could put the company in a huge risk.

But it starts with the CEO buy-in. Hopefully the CEO is curious enough, and then it has to cascade first to the security piece, compliance piece, legal piece. Now that is in a country like the United States where you have the luxury to have all those departments. When you translate the experience to countries in development, the conversations are a little bit more easier, but also has to be more gentle, because there is not so much exposure and knowledge of all the items that I just brought. So the thought process is, let's start with curiosity and let's find, let's determine how this impact today your ROI. Let's see how we can build a simple proof of concept that will not cost you much, but we can see the results.

And I would say, Erika, following to that, to be honest with you, I think you and I should probably do another podcast and focus on how important change management is. I think it's one of the pieces that people don't understand the value on a project, how important it is. And with this technology, if you don't really put it as a foundation, as well as you do with data quality and with security and compliance, there's no way that you'll be successful. So I will invite you to have another discussion and focus on that. Now that people know you a little bit, maybe have another discussion and go a little more in depth on the importance of that and how that makes your projects be more successful. And I know that Adin has one last question about if you would recommend somebody for us to have another chat.

I have something before that, Mario, actually. So I have something on the personal note, I think. Erika, when we talked to each other initially, you definitely—and I saw it also in the PR pieces that I read, your daughter is definitely one of your main center points in your life, I think, right? So I want to go there and just see like you and your daughter travel with a mission to give back and grow. That's kind of the message I got. So I just want to understand, how do you see AI playing a role in global development and community building, and what legacy do you hope to leave through your work?

Thank you for that question, Adin, and it is a very special area of my life. And just going back to Mario very quick: yes, I would love to do a podcast on change management.

And so my daughter's name is Sophia. And as I have gone through certain hardships in my early years, she had her own. And I think human beings connect through vulnerability and also through sharing those hard experiences. When humans can do that, it built a bond that is incredibly strong. So Sophia and I have that.

I do believe that there is such a much grace in her because it baffles me at time that someone who has gone through the experiences that she has can has the capacity to exercise the level of grace and compassion that she has. So at times she becomes my role model. So it has been a beautiful, symbiotic experience of leadership.

So her traits were pretty clear and shiny since a very early start. And what I do believe is that we cannot value what we don't know. And the US is a wonderful place. My goodness, we have the safety, we have the organization, we have the structure, we have the protection, we have the right to exercise our voice. And with all of that, I wanted to show her that the rest of the world might not work on that way.

So since early in her life, we have visited Peru a few times, we have been in several areas in Europe, and not necessarily in the fancy and the glamour and what you see in so many social apps, but actually in the sides of the world, including Africa, where life does not work as fine and as precise as it work here. Why? Why do I do that? Because for me, it's absolutely critical that she does not remove the lenses of compassion and care to humanity.

One of her masterpieces was one of her essays when she decided to apply for college. And the name, the title of her essay was "A Citizen of the World." And that is how she sees her.

And why do I focus so much on getting her out? Because my hope is that she will carry the legacy of service, the legacy of selflessness, the legacy of resilience. And I think it has been an intentional partnership that every time that we have felt that life has hit in a way that might be painful, we have had each other to quickly brainstorm and to say, "Okay, these are the conditions. How we're going to get back up?" Now my hope is that someday she will have her own children, right, and she will be able to pass that character trait.

But along those lines, something that has kind of propelled the resilience is AI. We have a challenging situation, and it's kind of automatic: "Okay, our AI model has a name." I said, "Okay, let's ask Arlo. What are we going to do?" And we kind of just dump the situation, and we ask the questions, right? What are best practices? What are the most common traits? What has been the result of people making these decisions? And it can go to anything. It can go from caring for our dogs, it can be from being stranded somewhere in Bulgaria, it can be not knowing how to get out in a little street from Marrakech, it can be, "Okay, how do I negotiate if I need this amount of spices in the little souks and still be gentle and polite?"

So we have used AI for so many things, and if any, it has enhanced the way how we choose to live our life, how we choose to connect with community. Because another good question is cultural awareness. "I am here, I am in your land. You don't have to cater to me. I am the visitor. Now it is my responsibility to cater to you to the best of my ability and make this relationship work."

So she has been—I mean, I have had the joy to see powerful deliveries of her own journey in front of women in leadership here in Oklahoma, powerful deliveries about what service mean at high school. Today, she works with House Representative here in the state of Oklahoma, and also for an agricultural technology company that is injecting AI. So legacy, I think the legacy is there. It's not so much what she knows, but it's what she brings to the table with her grace and humility.

Well, thanks for sharing that personal note, Erika. So, yes, and now my final question, and I don't know if Mario has one, but we ask that to every guest on the podcast: so who can you recommend to do the next podcast with us? So the name, company, and we will hunt them through you, of course.

Absolutely. I can think about my friend Ben Rybicki, and both of us serve in the Project Management Institute Oklahoma City Chapter board. He's incredibly innovative, creative, has developed a great framework for project, successful project management that can deliver high value. I will be more than glad to connect with him and just establish those relationships, but I think he will be a terrific addition to the podcast.

Well, thank you for that, Erika. Thank you very much. All right, so that's it on this episode, I would say. So, Erika, I think this is maybe not the last one, as I can hear it from Mario. So let's see if we can make that happen, the change management one. Thank you so much for being on this one, taking your time from the day, sharing so much professional and private with us, Erika. So that's from me. Mario, I leave it with you and then with Erika, and then we can stop the podcast.

Thank you, Erika. It's been a real pleasure meeting you and sharing some battle scars. We'll talk a little bit more on the other podcast on change management. We definitely have a lot of war stories that we can share. And look forward to meeting again.

Thank you so much. And Adin and Mario, I leave with the final that AI is a mirror and a lever. It's a mirror because AI is going to mirror me, my thought process, my character, my skill, and it's going to enhance. And it can go either way. But also is a lever, and as a lever, can help us to propel so much progress, especially in those countries that need progress the most. I am very grateful to both of you, grateful for the space, grateful for allowing me to dust off some of my memories and helped me to relive the excitement of what this journey has been. So thank you so much.

You are—we need to thank you, Erika. So, yeah, definitely speak another time, I would say. So have a nice afternoon, Erika.

You too. Bye-bye.

Bye.