← The Corporate Director Podcast
The Corporate Director Podcast
Incorporating gen AI into business strategy
00:34:35
In this episode of The Corporate Director Podcast, we talk with Sophia Velastegui, a renowned advisor on AI business strategy with experiences at Microsoft, Google, and Apple. Sophia shares her insights on the evolution of AI in the business...
Loading summary
Transcript59 lines
- [00:00]
Sophia Velastige
Foreign.
- [00:09]
Dottie Schindlinger
Welcome to the Corporate Director Podcast where we discuss the experiences and ideas behind what's working in corporate board governance in our digital tech fueled world. Here you'll discover new insights from corporate leaders and governance researchers with compelling stories about corporate governance strategy, board culture, risk management, digital transformation and more.
- [00:35]
Megan Day
Hi everybody and welcome back to the Corporate Director Podcast, the voice of modern governance. My name is Dottie Schindlinger, Executive director of the Diligent Institute, and I'm joined once again by my co host extraordinaire Megan Day, principal solution designer at Diligent. Megan, how are you doing today?
- [00:52]
Hey, Dottie. And happy Halloween. Oh yeah.
- [00:56]
So. So what are you dressing up as this year? I still remember your costume when you went as Elizabeth Holmes.
- [01:01]
I, I mean, I will never forget that. That's. I can't top that. I will say I was just asking Chat GPT what I could dress a toddler up for Halloween with items found around the house and the, the responses are, are pretty good. A robot, a flower, a baby chef. All of those things are definitely doable. We have a Halloween parade on Friday afternoon and I've got to come up with something with, for somebody who, who doesn't know the concept of Halloween yet, but is is making her way through, she knows what a ghost is. A ghost says boo. And Frankenstein looks like Daddy.
- [01:42]
Poor Nick. So, so what have we landed on? Or should I say, what is daddy gonna tolerate?
- [01:51]
I mean, daddy is Frankenstein. Let's make it happen. I love it.
- [01:56]
I love it. That's great, Megan. Well, you know, you brought up using AI to find your Halloween costume. And I was so excited about today's interview because we had the chance to sit down with Sophia Velastige, who has just worked in AI now for a very long time. She's worked with all the big companies. Microsoft, Google, Apple. Now she's the AI advisor to the National Science Foundation. So she works with the Biden administration on AI policy and really thinking about sort of ethical AI and AI things, all AI things related. And so we had a chance to sit down with her because she's also on several corporate boards. And I thought, boy, those corporate boards.
- [02:34]
Are lucky to have the AI expert.
- [02:37]
Actually on the board and in the room. And I was eager to talk with her about, you know, how she, how's she doing? How are things going in the world of AI?
- [02:46]
Well, let's give it a listen, Dottie. I mean, this is certainly, I don't want to call it like the topic du jour. I mean, it very much is it's but it's than that it is fundamentally changing how companies operate, how boards need to think about things. And so really timely topic.
- [03:00]
Great. Let's give it a listen. Joining us on the Corporate Director Podcast today is Sophia VELASTIGE. Sophia advises Fortune 200 companies on AI business strategy, leveraging her experiences at Microsoft, Google and Apple. She was a Council Chair and the Microsoft AI Representative for the World Economic Forum. Sophia is a member of the National AI Advisory Committee at the National Science foundation and she serves as an Independent Director and Chair of the Technology and CyberSecurity Committee at Blackline. Sophia, welcome to the show.
- [03:39]
Sophia Velastige
Welcome and thank you for having me here. Dottie on Diligence Podcast I'm so delighted.
- [03:44]
Megan Day
To have you here today on our special Scary Halloween episode.
- [03:48]
Sophia Velastige
Woo.
- [03:49]
Megan Day
We're going to talk about the scary world of Generative A. Hi. Well, listen, I mean beyond me, just rattling off all the amazing things that you've done in your career. I wondered if you could start by sharing with our listeners a little bit more about your current work and sort of how you arrived at what you're doing now.
- [04:06]
Sophia Velastige
So currently I advise various Fortune 200 companies on AI business strategy, but technology strategy as a whole and been also Chief Product Officer and Chief AI Technology Officer at companies like Microsoft and Activ. But how did I get here? In truth, my education was in mechanical engineering. I had nothing to do with AI. I was an engineer that designed computers and cell phone in the early 2000 time frame or so. My background started as semiconductor and I noticed that semiconductor. With the boom of that happening in the early 2000, consumer devices will explode in the market. And so I transitioned myself to Apple and when I got to Apple and iPhone just came out to state the timeframe of things is that I noticed that for every iPhone and iPad that there were unlimited apps that you can have. So I transitioned to software and I saw that the software and apps that did the best were embedding AI and I noticed the benefit of how it's been able to provide new functionalities and delight the customers. And that's how my journey in AI started. Then what I did is I continued to increase my work and projects around AI products starting from Apple, then at Nest with emerging technology infused in all of these IoT devices, the unloved thermostat, the smoke detector and your doorbell cameras. Then 100% of my work became AI. Once I joined Microsoft as the General Manager of AI Products and Search and to give a little bit of time frame about Microsoft, it was 2017. Genai first was published and came in the market around 2016. The partnership happened with OpenAI around 2017. So I was there pretty early in that journey and I was brought in as part of the executive cohorts to help with the transition from Microsoft being a software company to become an AI company. Continued doing that with the AI products and then was promoted to become the Chief AI Technology Officer at Business Application Groups Dynamics Power Platform, Power Apps. It's been an amazing journey to see Genai being infused in the Microsoft products and assets and it's led to significant growth like 45% year over year growth just on intelligent cloud alone. And then I worked at Aptiv in self driving car technology. Aptiv is $20 billion in revenue working on self driving car. And the reason being is that AI will come to the physical world and understanding that is amazing. Then like you stated, I advise National Science foundation. But in addition I just recently got announced as joining the Georgia Tech President's board and focusing on some of the AI strategy for Georgia Tech.
- [07:18]
Megan Day
Well, congratulations on that new appointment and hats off to them for being so smart to hire you. That's really fantastic. I think as I was listening to your description of your journey, one of the reflections I'm having is AI is not new. I mean it's been under development for well over 50 years at this point. And yet I feel still when you have conversations with directors and business leaders, as if AI just emerged on the scene two years ago with ChatGPT and it's so clearly not the case. And so I think you're probably the best person to ask this question. You know, I know you don't have a crystal ball, but what are some of the things that we should all be expecting about where it's going to go over the next, let's say three years? I mean there's been this explosion of development in the last two. Where are we going in the next three?
- [08:02]
Sophia Velastige
For digital native companies, they'll really focus on AI powered innovation in product development and decision making. And we're seeing this because first of all they've been working on it 50 years, but really in earnest in consumer products and enterprise for 30 years. One of the things they've realized for significant value creation, it's not an operation efficiency, it's actually in the innovation and new product and new category development creation, you'll see that and there'll be optimized products and prototyping and simulation and come to market that much faster. I want to give a great example of Airbnb as a prime example of a digital native company. They're leveraging it in their dynamic pricing, their smart suggestions, and it's significantly improved the host as well as Airbnb revenue by automatically adjusting rates based on predicted demands and competitor pricing for the customer. AI is driving that personalized search ranking and tailored to your preference and your past behavior. So they're finding the relevant properties faster and it's smoother and it aligns with their expectation. Third, they're using AI, especially gen AI, to optimize listing, enhance their photo selection to resonate with the target customers. So this is like using traditional AI as well as gen AI. Now the second cohort is the non digital native companies. They are first focusing on operational efficiency and reducing costs. As they become more comfortable with the technology, then they'll realize the true value will be in innovation. I want to give a great example of Procter and Gamble. They actually leveraged it for their supply chain efficiency by predicting demand variation and optimization in real time. What actually led them to invest in AI was actually the pandemic you can imagine supply chain scarcity and so forth occurred. Well, this has led to their success and agility in their supply chain in production and distribution across all market demands. And I thought that was a very astute way of leveraging AI.
- [10:22]
Megan Day
Those are some really, really helpful examples. And it also, I think underscores for me why it is so incredibly important that the AI strategy isn't this side hustle. So do you want to talk a little bit more about, you know, really why companies need to make sure that AI strategy is directly tied into the overall business strategy and sort of maybe some pointers on how to do that?
- [10:44]
Sophia Velastige
Well, tying AI strategy directly to the overall business strategy is essential because it ensures that the investment aligns and supports the company's key objective full stop. And when it's not integrated into the broader business framework, they risk being isolated tech experiment you may hear of like a prototype, death by prototyping and so forth with limited impact on business outcome. So the key thing to think about with AI strategy and business strategies one, strategic alignment and focusing on value creation. Thinking about how can it be not AI for the sake of AI, but it's something very specific and a use case that will give you that return on investment. Like for instance, a retailer. Tying AI to their business strategy means using it not just to analyze the customer data, but to drive decisions around product development, inventory management and personalized marketing. That's very cohesive and what that does is it enhance the customer experience and improve revenue. Things that are key to a retailer being successful. The second area where it's super important for AI strategy to align with business strategy is that enhanced competitive advantage because you're able to be more competitive at a shorter period of time with less effect and you can have operation efficiency, customer satisfaction and market expansion. And that really differentiates you. Amazon, we're all familiar with them. Their AI driven logistic and recommendation system is so tightly aligned to their strategy of customer centric innovation. They can get a product to you the same day. Right. And that's a competitive edge that you can't that other E commerce have not been able to meet.
- [12:37]
Megan Day
So I'd love to have you talk a little bit more about some of the other benefits that businesses achieve when they integrate AI into their strategic framework. And then I'd love to also have you reflect on are companies well positioned to do that though? Because I think about who leads companies and I know precious few of them have your resume. Sophia and so I'm just wondering if you've got some guidance for as they're integrating this into their strategic framework, who should they call for help? They're clearly going to need it.
- [13:06]
Sophia Velastige
So first of all is what are some of the benefits companies can see? I see three that come very quickly to my enhanced decision making based on vast data. Number two, innovation and number three is operation, efficiency and flexibility. What do I mean by that for the decision making? There's so much data that's out there that can impact to make the most timely real time decision. That is like a prime example of leveraging AI to do financial forecasting, modeling a varying situation. I talk about Procter and Gamble with supply chain when there was so much disruption and you can pivot that so quickly and something that used to take weeks or I mean excuse me, months or years now you can do in days and weeks. And think about the proctor gamble was able to get their products out to their customer faster than their competitor. Significantly number two is innovation. I talk about how new product lines can be develop faster than your customers. Taking in real time feedback from your customer and market trend like Tesla announced in their 2024 investor day it plans to develop electric motor that requires no rare earth element. And this brainstorming with the engineers, they did it with their internal gen AI and what would have taken years, multiple years they were able to achieve within about three months or so. And then on the third one, operation efficiency and flexibility. This is a lot of people are leaning. The reason why people are leaning into Cloud based AI services is you can scale up and down based on your business need in a dynamic market. And we're definitely in a dynamic market.
- [14:51]
Megan Day
Why this is so important.
- [14:52]
Sophia Velastige
Right.
- [14:53]
Megan Day
And it clearly it is so critically important for businesses to integrate AI into the overall strategic framework. But the leaders of those organizations may or may not have a lot of expertise in this area. And I think, you know, a good example, Sophia, is not many companies are being led by someone who has your resume. How can they make these decisions? Where do they need to go to get some help to do this?
- [15:14]
Sophia Velastige
Well, so one of the things in the hype cycle of AI is there are definitely AI experts coming everywhere. And I would say that AI is not something that can be learned in a fraction of time or reading a bunch of articles from Harvard Business School and so forth. But who are the people that's actually executed on AI products or AI infused products and businesses having those leaders infused in the executive rank. But also the board will provide a lot of insight. I myself, one of the reason is after my operating role I decided to take some time off. But many people came to me and asked us about helping them with their AI business strategy. And so that's why I started consulting some of these Fortune 200 companies. There are many advisors that leveraged AI in the business and leveraging their advice.
- [16:08]
Megan Day
Above all, I think that's really good advice. Well, let's talk a little bit about that governance role because from a governance perspective, I think boards need to really understand how to get their arms around the risks associated with AI. And there's a lot of risk, there's a lot of opportunity, but there's also a lot of risk. And so what are some of the ways that companies should address the risks that are associated with implementation of AI? And I'm thinking specifically around, you know, the fact that this is constantly evolving. There's so many rapid changes happening in AI. So how do you do a good job with the risk oversight there?
- [16:42]
Sophia Velastige
First of all, leaning and leveraging current governance framework around opportunity and risk and having a cross functional team of AI council as an example where their responsibility or consideration is. Number one, there's seven things I would state. One is implementing a robust AI governance structure. Number two, establishing clear AI ethics guideline. Number three, having a continuous risk assessment and management, especially with the speed at which AI is advancing. Number four, data privacy and security compliance, especially with the regulatory environment changing. Number five, establishing transparency and explainable protocols. Six, incorporating human in the loop oversight and decision making. Number seven, scenario testing and stress Analysis. So the reason, especially the scenario testing and stress analysis, is sometimes you think that something will go a certain way. It's not until you actually stress test set that you will see the pros and cons of it. Just like OpenAI, you see the hallucination, it's actually based on design because it's based on everything on the Internet. But that was a risk that OpenAI decided to move forward with.
- [18:00]
Megan Day
I feel like you've just given us a great sort of checklist for directors to ask at a meeting. Right. So what are we doing around AI ethics? What are our guidelines? How are we continuously assessing and managing risk? How do we ensure data privacy? How do we establish transparency? Those are great things for directors to be asking at board meetings. So I think that was a really helpful list. I also want to talk about sort of the other side of AI risk, which is, you know, most AI models, I think it's fair to say, have a very Western centric view. Or if I'm being really honest, they have a US centric view. Let's be candid about that. And you know, companies are global. We live in a very globally connected economy and society. So how can companies really ensure that their AI strategies are inclusive and are considering diverse perspectives, international lenses, you know, how do you do that?
- [18:48]
Sophia Velastige
Well, that's a very astute observation and it is true. And so one of the things to think about AI system and these large language model, it starts from the beginning, when you're creating that AI strategy for your corporation, that the people around the table are diverse in thought and perspective because they're representing the global customers, employees that you have. And I want to leverage this idea of digital representation. What do I mean by that? Is having digital representation so we can develop AI that serves all people. What does that mean? Today, AI is being trained and developed based on data from developers and the people who use it. The second part is where we can get diversity and perspective incorporated for the people who use it. Their input and feedback help train the model and deliver new information that aligns with the global customers and employees. But if we want to represent, we want all the people to be represented in the AI, both now and the future. We all have to be involved in building, building that. So what do I mean by that? Is that when you're creating the AI strategy, you're developing it, you're testing it, that there's a diverse group of people that are doing that. So there are six things that I consider. What is the data source? Is it diverse enough? If not Gather that data and input that. Number two, who's designing and developing these products? Are they also inclusive in their thinking? Three, how do you have local context or consideration around the world? So it's not just like you said, US centric, but you're taking into account Asian perspective, European, African, South American and so forth. Number four, bias audit and fairness evaluation. This is the oversight aspect of it. Five, when you are ready, have a product that you're ready to dog food or kind of like test in the market before it goes launch. Having that community and stakeholder engagement. That's diverse, they're testing it. And number six is making sure that you're implementing the ethical AI framework across all of these activities.
- [21:10]
Megan Day
So I think you've already given us a number of really great examples of companies integrating AI. But I wonder if you have some specific case studies or other examples where aligning the AI strategy with the business strategy had a really significant impact on the business outcomes, whether those were positive or negative.
- [21:28]
Sophia Velastige
One that comes to mind is Pfizer. As recently as 2023 they announced they used AI to accelerate drug discovery. And what they did is they looked at its pipeline of vaccination and therapeutics and that was directly aligned to the business strategy about bringing innovative products to the market at a speed that is faster than their competitor. And what they use and leverage is machine learning models to identify potential compounds and the drug efficiency and effectiveness that resulted in a drug discovery phase reduced by 40%. Then also their R and D efficiency was greatly increased because now they are able to evaluate thousands of compounds simultaneously in this simulated AI environment. And that saved millions of dollars in R and D costs. Hopefully the drugs will cost less even to the end consumers like you and I who are leveraging these life saving capabilities. Another one that I want to talk about is Goldman Sachs. I've always admired the work and their thoughts on innovation. They're using AI for a risk management and fraud detection platform to make sure that they are looking and detecting like anomaly trading patterns and potential fraud. This directly aligns with their business strategy of maintaining market integrity and reducing operation risk. They're enhancing the real life decision making and compliance monitoring. These are real numbers fraud detection. They were able to reduce false positive by 60%, detect high risk activity 70% faster than traditional methods. The escalation of these risky behavior was caught when it had the minimum impact for operational efficiency. They automated compliance checks which saved thousands of people's hours, reducing cost significantly. And so this is really enhanced your clients trust and satisfaction as well as the trading performance. And of course there's Microsoft, but I thought these were two other ones that were very much of note.
- [23:50]
Megan Day
I think that's really helpful too. And I like that you're kind of giving us examples from different industries. Well, Sophia, this has been such a great conversation, but before we let you go, there are three questions that we like to ask all the guests that join us on the podcast. And so the first one is, what do you think is the biggest difference between boardrooms today and 10 years from now?
- [24:08]
Sophia Velastige
I think one of the biggest difference will be this real time digital governance platform. Like right now, meetings are really centered around these preferences prepared materials like presentation and static financial report of the past and some modeling of the future. But I think the boardroom will now operate in a more dynamic digital governance platform where we will see real time data visualization and interpretation of the data and that AI will be this virtual co pilot so directors can see real time analytics, ask them questions about all of the other board decks and information that was sent to get trend analysis and also start simulating like what if we have a economic recession and it's a soft landing, it's a hard landing, or there's no recession at all? What are the implication of that to our business in our market? The second thing I see is this shift towards continuous governance model. So if you are able to have access to this real time data platform, governance right now is still very based on periodic meeting, quarterly, biannual and so forth. But in the future you'll have this continuous governance model that will be supported by AI and automated and so that board members will have live information, they'll be more agile to make fast decision in the moving market.
- [25:34]
Megan Day
Well, I have to say those are all very welcome suggestions Sophia, because those are all things that Diligent is working on. So I'm being very self serving, very happy to hear that you agree. This is where we're going.
- [25:45]
Sophia Velastige
I love it.
- [25:46]
Megan Day
The other thing we always like to ask our guests is what's the last thing that you read or watched or listened to that made you think about governance in a new light?
- [25:54]
Sophia Velastige
So one of the most thought provoking piece that I read on governance was actually the 2023 World Economic Forum report on governing AI and the implication of Gen AI for business and society? And why is that? Because it's really shifted my understanding of governance from being primarily a structured focus on oversight and compliance to one that underscores the future. Governance will be about foresight, ethical leadership and rapid adaptation. It's not just ensuring adherence to established rules. And this new lens suggested that governance will play an increasing prominent role in defining corporate purpose and public trust, making it integral to the organizational success in this AI age. That just fundamentally changed my framework.
- [26:50]
Megan Day
I love that. And we'll have to make sure to put a link to that report because it's a really good one so people can take a look at it. And finally, Sophia, what is your current passion project?
- [27:00]
Sophia Velastige
As I said recently, I was asked by Georgia Tech to join their Presidents Advisory Board and it's work and I've been working with Georgia Tech as a board member for College of Engineering Mechanical Engineering, but it's now been elevated to see how further Georgia Tech can lean on their expertise in AI and technology. I think that's really exciting because the students are our future in many ways. They're future in our many ways, but they also collaborate with industry and government in order to make sure these technologies are provided to a wider audience. And I think that's amazing.
- [27:45]
Megan Day
Well, Sophia, thank you so much for joining us on the show. It's been such a pleasure to speak with you today.
- [27:49]
Sophia Velastige
Thank you Dottie and Diligent for having me here.
- [27:52]
Megan Day
We've been joined today by Sophia Velastogi, President and CEO of Velastige Ventures and a member of the National AI Advisory Committee at the National Science foundation and a board member for blackline. Sofia, thank, thank you so much for joining the show.
- [28:05]
Sophia Velastige
Thank you.
- [28:14]
Megan Day
Great interview, Dottie. And I love all of her recommendations on how, how organizations, how boards need to be thinking about this at really just a tactical risk management level. And honestly it makes it a little bit more manageable, a little less scary. Keeping with the Halloween theme to wrap your arms around AI, I also thought.
- [28:36]
I agree with you, Megan. I think some of her tips were really practical. You know, one of the questions we asked her was about, you know, where do you see AI evolving across the business landscape over the next few years? Which of course is really the only question anybody wants the answer to. Like where is this going? And I thought her answer was quite interesting, that she really sees sort of two different paths, one for digital natives and one for non digital natives. And that digital natives will be really focusing on innovating product design and development and making decisions that are all using AI. And then for the non digital natives it'll be really focusing first on operational efficiency and reducing costs. That, that feels right to me. And what, what I love about that answer is no matter who you are, there is something for you with AI. It's not like anyone's being left out. There are some ways that you can leverage AI now that will be really helpful to you, especially in just in terms of operational efficiency. But then over time, as people get more up to speed, as they know more, as they bring in better experts, they can really upend their entire business model. And she gave a great example there about the way that AI was leveraged by Pfizer to develop the COVID vaccine. And we all kind of heard about that in the news. And I'm sure most of us, maybe I'm just speaking for myself, but most of us didn't fully understand what they were saying. Oh, totally right. But it is really interesting. In fact, we even had a chance to interview someone from Pfizer back when that happened that, you know, they accomplished in a matter of nine months what in years past had taken ten years to accomplish. And that is pretty breathtaking. I mean, that pace of change that can completely upend your business model. I mean, think about all the downstream impacts of that. Things that don't have to exist anymore or can exist differently, or be more nimble, more agile. I think I'm probably the most excited about what AI is going to do to the field of medicine and health care. Of all the ways that AI is being leveraged, I think that's where I'm most excited. I don't know your thoughts, Megan?
- [30:39]
Yeah. I loved personally her point about the World Economic Forum paper and how that's her take on governance. I wrote down and underlined repeatedly when she said boards in the future will and governance will be about rapid adaptation. And that is, if you just think about it, we have been heading in that direction. But like, that is not the, the phrase that people associate with corporate governance.
- [31:07]
It really isn't. I mean, you know, again, I always make this joke and it's not really a very funny joke, but I make it anyway, which is that one of the things you say that boards do in the boardroom is deliberate on issues and deliberate has a second meeting and it's slow. And so. So you know, that that is something that I think is ripe for change. But I love that idea of rapid adaptation because we have for so long now, for well over a decade now, been living in this time of volatility, uncertainty, complexity and ambiguity. And the only way to have resilience in that type of environment is to adapt and keep adapting. Because you can't sort of have a three to five year plan that you just create and forget and, you know, run the play. You've got to constantly adapt. It has to be sort of on a rolling basis. How are we still getting to that goal? Yeah, it's, it's really interesting. I think these are some exciting times. I do, I do often, I think we do on this show often feel bad for our corporate directors because it's just so much, it's just so much to take in and so much to try to wrap your arms around.
- [32:15]
Generative AI is going to make it easier. Come on.
- [32:19]
Yeah. And you know, it's interesting too. I'd gone to a conference in September, maybe I mentioned this in a previous episode where Florin Rotar had this really great line that he was like, you know, I think we've overestimated in the short term what Generative AI is going to do and we've grossly underestimated in the long term the change that it's going to have. And that line stuck with me. And I don't know if it was his, but it was something that he said and I really agree with it. I think that that is probably right.
- [32:46]
Well, Dottie, I think I sent you a link a couple of weeks ago of a automated AI based podcast that's now now available. And I will tell you we'll include the link in the show notes because out there the robots are creating some pretty solid podcast content.
- [33:06]
So do you, do you mean we could just stop doing this, Megan?
- [33:08]
I think so.
- [33:09]
Keep going.
- [33:10]
Amazing.
- [33:11]
All right, well, we'll see if we can. We'll see if the AI bot can nail this wacky laugh that I have. And if it can, then we're good. We can sign out and go do something else. Well, Megan, that wraps up another episode of the Corporate Director Podcast, the voice of modern governance. Like to say a few special thank yous. First and foremost to our AI expert Sofia Velastege, podcast producer Kira Ciccarelli, the team at Oxenfree for making this podcast a reality, and most especially Diligent Corporation for sponsoring this show. If you like our show, please be sure to give us a rating on your podcast player of choice. Five stars only, please. And you can listen to our episodes and check out inside today's boardrooms presented by Diligent Instit, a weekly web show covering best practices for today's corporate boards and committees by going to diligent.com resources. Thank you so much for listening.
- [34:00]
Dottie Schindlinger
You've been listening to the Corporate Director Podcast to ensure that you never miss an episode. Subscribe to the show in your favorite podcast player if you'd like to learn more about corporate governance and tools to help directors do their job better, visit www.dotdigent.com. thank you so much for listening. Until next time.