The initiative focusing on the integration of artificial intelligence into business strategies delves into the impact of AI on organizational strategy development and implementation.
Prem Natarajan, the chief scientist and head of enterprise AI at Capital One, has a background shaped by growing up in a multilingual environment, sparking his interest in language and leading him to a career in machine learning and technical development.
In this episode of the Me, Myself, and AI podcast, Prem discusses how Capital One leverages AI to provide value to customers, particularly in areas like fraud detection. He emphasizes the transformative potential of generative AI in enhancing the developer experience and underscores the importance of blending product, science, and engineering expertise for successful AI and machine learning projects.
Prem Natarajan, Capital One
As the chief scientist and head of enterprise AI at Capital One, Prem Natarajan is responsible for shaping technology strategy, architecture, and development across the company’s data, analytics, and AI and machine learning initiatives. With over two decades of experience in leading science, technology, and commercialization efforts in various applications of machine learning, including natural language processing, speech recognition, and computer vision.
Transcript
Sam Ransbotham: The adoption of generative AI necessitates a delicate balance between product innovation, scientific advancements, and engineering capabilities. In this episode, a prominent figure in the financial services sector shares insights into navigating these challenges.
Prem Natarajan: I’m Prem Natarajan from Capital One, and you’re tuned in to Me, Myself, and AI.
Sam Ransbotham: Welcome to Me, Myself, and AI, a podcast dedicated to exploring the intersection of artificial intelligence and business. Each episode introduces innovators leveraging AI in various domains. I’m Sam Ransbotham, a professor of analytics at Boston College and the AI and business strategy guest editor at MIT Sloan Management Review.
Shervin Khodabandeh: And I’m Shervin Khodabandeh, a senior partner at BCG specializing in AI business applications. Our collaboration with MIT SMR has enabled us to delve into AI research and insights, engaging with practitioners and surveying companies to understand the dynamics of building and scaling AI capabilities.
Hello, everyone. Today, Sam and I are in conversation with Prem Natarajan, the chief scientist and head of enterprise AI at Capital One. Prem, thank you for joining us on the show. Let’s dive right in.
Prem Natarajan: It’s a pleasure to be here, Sam and Shervin.
Shervin Khodabandeh: Could you elaborate on your role at Capital One and provide insights into your career journey leading up to this position?
Prem Natarajan: My role at Capital One, particularly in the realm of AI, revolves around advancing the company’s tech-forward approach. Capital One has been a pioneer in embracing technology, being one of the first major enterprises to fully adopt a single public cloud infrastructure. This foundation sets the stage for harnessing the potential of machine learning and AI to drive value and innovation across the organization.
Drawing from my experience and expertise spanning various domains such as natural language processing, speech recognition, and computer vision, I aim to bolster Capital One’s AI capabilities, tools, and research endeavors. My career trajectory has been shaped by a deep-rooted interest in language and technology, starting from my formative years in a multilingual community to my tenure at leading institutions like BBN Technologies and Amazon, where I honed my skills in AI and machine learning applications.
Shervin Khodabandeh: It’s evident that your diverse background has equipped you well for your current role. How do you perceive the ongoing inflection point in the AI landscape?
Prem Natarajan: The current phase in AI evolution marks a significant inflection point characterized by the convergence of cutting-edge technologies like transformers and generative AI. This juncture signifies a shift towards AI systems demonstrating adaptability and contextual learning capabilities, a phenomenon often referred to as “in-context learning.”
Reflecting on the historical progression of AI, from expert systems to probabilistic models and now to transformative capabilities of generative AI, each transition has heralded advancements that eventually become integral to everyday applications. The democratization of AI tools and the integration of novel technologies like large language models hold immense promise in reshaping the developer experience and unlocking new possibilities for innovation.
Shervin Khodabandeh: Given your extensive background in AI and technology, how do you envision the evolving composition of AI teams and the requisite skill sets for driving AI initiatives?
Prem Natarajan: With each technological wave, the demand for diverse skill sets within AI teams undergoes a reconfiguration to align with emerging requirements. The synergy between product vision, scientific innovation, and engineering prowess forms the cornerstone of successful AI endeavors. While established use cases like fraud detection may emphasize the fusion of science and engineering expertise, nascent applications demand a holistic approach encompassing product ideation, domain-specific insights, and technical execution.
The future of AI teams hinges on cultivating a product-centric mindset across the organization, fostering collaboration among specialists from varied disciplines to navigate the complexities of AI implementation effectively. As AI technologies evolve to offer scalable and adaptable solutions, the strategic alignment of product, science, and engineering capabilities becomes paramount in maximizing the value delivered to customers and stakeholders.
Sam Ransbotham: The interplay between product, science, and engineering facets underscores the multifaceted nature of AI initiatives. Could you shed light on a specific instance where this triad of expertise has driven transformative outcomes at Capital One?
Prem Natarajan: A compelling example of this collaborative synergy can be witnessed in our revamped fraud platform at Capital One, where machine learning serves as the linchpin of real-time decision-making processes. By integrating advanced ML models into our fraud detection system, we have achieved unparalleled efficiency in processing complex data streams and delivering swift responses to potential fraudulent activities.
The success of our fraud platform epitomizes the harmonious convergence of product vision, scientific innovation, and engineering excellence. From envisioning impactful use cases to translating them into inventive solutions and ensuring seamless operational deployment, the alignment of diverse skill sets has been instrumental in fortifying our fraud detection capabilities and enhancing customer experiences.
Sam Ransbotham: The collaborative efforts across product, science, and engineering domains are evidently pivotal in driving AI innovations. In your view, where do you perceive the greatest potential for AI to create value in the current landscape?
Prem Natarajan: The transformative potential of AI lies in democratizing access to a myriad of services and resources across diverse societal strata. By leveraging AI to deliver enhanced customer experiences, optimize business operations, and foster inclusivity, organizations can harness the power of AI to drive positive societal impact and foster innovation on a broader scale.
Shervin Khodabandeh: Reflecting on the current AI landscape, where do you believe organizations might be over-relying on AI, potentially overlooking the nuances of human-centric interactions and experiences?
Prem Natarajan: The allure of AI lies in its capacity to streamline processes, enhance efficiency, and reduce friction in various domains. However, organizations must exercise caution in deploying AI solutions where human-centric nuances and personalized interactions are paramount. Striking a balance between AI-driven automation and preserving the human touch is crucial to ensure that AI complements, rather than supplants, human capabilities in delivering seamless and empathetic experiences.
Sam Ransbotham: Transitioning to a more speculative realm, what capability do you wish AI possessed currently that remains beyond its current scope?
Prem Natarajan: If I could envision a fantastical AI capability, I would wish for AI to elevate my singing prowess to virtuoso levels. As a music enthusiast with a passion for singing, the ability for AI to enhance my vocal talents and create awe-inspiring musical performances would be a dream come true.
Shervin Khodabandeh: The convergence of AI with creative pursuits like music underscores the boundless potential of AI to transcend conventional boundaries and unlock new realms of human expression. Thank you, Prem, for sharing your insights and experiences with us. It has been a pleasure engaging in this dialogue.
Prem Natarajan: Thank you for the enriching conversation. I appreciate the opportunity to delve into the multifaceted landscape of AI and its transformative impacts.
Sam Ransbotham: Thank you for tuning in. In our upcoming final episode of Season 8, Shervin and I will be in conversation with Mark Surman, the president and executive director of the Mozilla Foundation. We look forward to exploring the dynamic intersection of AI and societal progress. Stay tuned.
Allison Ryder: Thank you for listening to Me, Myself, and AI. We believe that the dialogue around AI implementation extends beyond our podcast. To foster continued discussions and collaborations, we have created a dedicated LinkedIn group called AI for Leaders, tailored for listeners like you. Join us to engage with the show’s creators, share your insights, pose questions, and access valuable AI implementation resources from MIT SMR and BCG. Visit mitsmr.com/AIforLeaders to become part of this vibrant community. The link will be included in the episode notes, and we look forward to connecting with you there.
