In a recent CDO magazine interview, Sue Pittacora, Executive Advisor at Wavicle Data Solutions, and Abhi Seth, Chief Enterprise AI and Data Officer and Vice President at The Boeing Company, sat down for an in-depth discussion, highlighting the critical role of data governance in developing effective AI models.
In continuation from the first part of the interview, this conversation with Sue and Abhi explores the importance of robust data management and compliance in innovating with AI and building an AI-ready company culture.
Watch the full interview here or scroll down for a detailed transcript of Abhi’s insights.
Sue: Hello and welcome to the CDO Magazine interview series. I am Sue Pittacora with Wavicle Data Solutions, and I’m delighted to be here today with Abhi Seth, the Chief Enterprise AI and Data Officer and Vice President at The Boeing Company. Abhi, thank you for joining me today.
Abhi: My pleasure. Thank you for having me.
Abhi: I would say data governance is a very important aspect of our overall AI strategy. You can’t go too far without robust and trusted data. Data governance is not only part of the AI strategy because we want to do AI, but a lot of the focus on data management and data governance is focused on risk management and establishing some guardrails on how to drive compliance around what data should be visible to whom. We play in the defense sector, so we have a lot of LiDAR data, and it’s restricted to US personnel only. How do we mark the data, how do we make sure that only the right people can see it, and how do we understand personas? I think those aspects become important.
Plus, I think, to unlock value of the data as we create these single-source-of-truth data stores, we are creating these lakehouses for our important functions like a quality lakehouse, and a supply chain lakehouse. We are focused on creating governance around defining what are your critical master data assets, how we drive data definition around those master assets, how we define data quality around those, and then identifying stewards in the business who can take ownership of those master data assets and drive improvement in traceability and quality of those master data assets. So that’s how we are driving a data governance strategy.
Abhi: I think some techniques are evolving, and if you think about it, the AI techniques are evolving much faster than people’s data maturity. Not so many years ago, we were talking about big data and machine learning. Now we are talking about deep learning, and suddenly, we have entered into the age of generative AI, and it’s not like it will take five years. I think everybody’s driving generative AI applications at a very active phase.
So, I think the breakthroughs will be in applying and harnessing the AI technologies available to the industry. How you create differentiation is about how you adopt and leverage these capabilities with unique application areas that give you a sustainable competitive advantage. Applying these techniques to individual industry sectors or individual companies is kind of the secret sauce that creates and unlocks value.
I think the companies that are taking the approach of value-driven or outcome-driven analytics and can then prioritize and pick some of the most important and strategic problems will create a lasting competitive advantage and will be the companies that are going to be ahead in the game.
The second thing from a pure play technology perspective is that we’ve been playing very actively with generative AI in the last year and have found many interesting use cases. Boeing was one of the first companies to roll out a generative AI platform to 170,000+ users in 2023. At the end of 2023, we built a platform and rolled it out to everybody in the company. We rolled out a generative AI academy to upskill everybody. We now have 22,000 active users, and the coolest thing we have found is it’s more of a secure and governed way to use gen AI, but people are leveraging the platform more as a test bed to test and push the art of the possible for what generative AI can do for their specific problems without writing a single line of code. That’s kind of the most exciting part that I see.
A lot of subject matter experts who are domain experts can bring in their domain knowledge, train the gen AI model, and then ask it questions to see how far they can push it, which is pretty interesting to watch in this domain. So, one of the things that gen AI brings is the ability to search and retrieve.
My background is working with large, 100-year-old industrial companies like Boeing. A lot of the challenge in older companies is the domain knowledge and all the stuff we have learned over so many decades. How do you really get that knowledge to a person who’s just joined a company like Boeing? As we are trying to make decisions, how do we get that domain accessible? Generative AI can actually ingest and have decades of domain knowledge available at people’s fingertips, and it’s truly powerful.
I’ve been involved in knowledge management, knowledge curation, and accessibility. I remember we used to have a technical information library, and this, and that, in different companies I worked for, where you’re like, hey, this is our domain knowledge for our company. Now imagine digitizing all that and making it available to anybody as a small popup on your computer window. You can ask it any question, and there’s 100 years’ worth of knowledge available to access to a new employee at an instant.
I think that is going to be a very powerful capability. I just gave some examples, and you know, just the proposal writing, like there is a way to write a proposal. Can we train a generative AI agent to write the proposal the way a specific company writes it? And then it can start writing. Then it becomes an assistant to a human, so we don’t take weeks writing a proposal. We can drive that productivity and get away from that writer’s block. But imagine that the same assistant is not just sighting but able to actually leverage decades of knowledge and bring it to your fingertips. I think your decision-making becomes a lot more robust, and I think it’s a truly powerful capability. It just remains to be uncapped in the near-term.
Abhi: A lot of our focus is building a very strong community of talent. It’s really about upskilling our team and driving digital awareness. We have a generative AI academy and a cloud academy where we are upskilling a lot of people on next-generation and up-and-coming cloud technologies. We also have programs for AI and machine learning. How do we augment our very technical workforce? How do we augment them and their skills in AI and ML so they can push the art of the possible out there?
A lot of the focus has been on enterprise talent and upskilling. The second is really driving a culture where we can pick the right problem and execute a solution within 12 weeks. A lot of the change in Boeing is that we were working on annual cycles in a lot of our digital programs, and we are now shifting from 12-month cycles to 12-week cycles, which is a huge shift. So, we are saying, “hey we want to build something in 12 weeks, and we want to deploy it, and then we want to start to observe how it’s working.” And we are now doing it over, and over, and over again.
The innovation cycle becomes much faster once you get that machine going. You can build a new capability, involve the users early, deploy it, see how you realize the value, and then build again. I think that discipline becomes unstoppable once you get that going because people feel like they are on a high once they see not just the analytics built in their lab but those being deployed in production and somebody using it. The true success of any analytic or AI capability is getting it sustainably embedded in a business process and creating value on an ongoing basis. Then, you are successful.
I think a lot of times, as data scientists and AI people, we get very excited when we can predict an outcome with a high level of accuracy, and it’s like, hey, we solved the problem. Yes, we solved the analytical side of the problem, but I think the business problem-solving is about how you package that prediction in a business process that is trusted, used, and adopted by the business community on a day-in-day-out basis. So that’s the shift that we are looking to drive.
Sue: Well, Abhi, I think that’s a great way to wrap up by talking about how your culture embraces this. You did a great job today. Thank you so much for all your insights and for joining me today. To all, please visit CDOmagazine.tech for additional interviews.
Abhi: Thank you.
This concludes our two-part interview with Abhi. For further insights, read more of our data governance and AI thought leadership here.