C-Suite Perspectives On AI: Steven Kawasumi Of Intuit On Where to Use AI and Where to Rely Only on Humans
An Interview With Kieran Powell
Data is foundational to the effective use of AI. I always say, “If you don’t collect it, you can’t measure it.” Because without data, we’d have no basis for training models or any way to assess their impact. And because AI, particularly machine learning models, rely on vast amounts of data to identify complex relationships and patterns that are not easily discernible through simple rules. Simple relationships might be manageable with straightforward algorithms, but as complexity and scale increase, these rules rapidly become unmanageable.
As artificial intelligence (AI) continues to advance and integrate into various aspects of business, decision-makers at the highest levels face the complex task of determining where AI can be most effectively utilized and where the human touch remains irreplaceable. This series seeks to explore the nuanced decisions made by C-Suite executives regarding the implementation of AI in their operations. As part of this series, we had the pleasure of interviewing Steven Kawasumi.
Steven Kawasumi is an accomplished product management executive, consultant, and artificial intelligence thought leader whose technical, business, and leadership abilities have been critical to his success. Steven’s innate ability to empower and push executives and employees to achieve greatness and his propensity to transform organizations into AI-forward companies has given Steven the opportunity to develop and implement numerous moonshots and skunkworks projects and take the lead on core corporate initiatives and strategy. Steven currently serves as a Product Management Leader at Intuit, a global technology platform specializing in financial software for small businesses and consumers.
Thank you so much for your time! I know that you are a very busy person. Our readers would love to “get to know you” a bit better. Can you tell us a bit about your ‘backstory’ and how you got started?
Currently, I am a Product Leader at Intuit. My primary focus is transforming the organization and reshaping our customers’ experiences by leveraging artificial intelligence. A key part of this has been developing generative AI solutions that have accelerated the adoption of durable, scalable, and reusable AI into several of Intuit’s offerings, such as TurboTax, Credit Karma, QuickBooks, and Mailchimp. These initiatives have helped establish Intuit Customer Success as an AI/ML-driven organization.
Additionally, my executive experience extends throughout my prior roles. At Motorola, I led the Data Science and Machine Learning Group in Customer Success, where my organization built the AI/ML infrastructure and strategy from the ground up, applying AI to customer data to create frameworks that identify and solve complex consumer experience issues. I’ve also fostered organizational growth and employee development through being a dedicated mentor with a positive leadership style.
I hold an engineering degree from Stanford University and an MBA with High Distinction from the University of Michigan. My academic foundation has been pivotal in my journey, which also includes founding a private equity fund focused on AI adoption for small to medium-sized businesses, solidifying my reputation as an AI/ML and data science expert.
It has been said that our mistakes can be our greatest teachers. Can you share a story about the funniest mistake you made when you were first starting? Can you tell us what lesson you learned from that?
When I embarked on a Skunkworks project to drive transformational change, I assumed that securing funding, assembling a top-tier team, and developing an innovative solution would suffice to gain support. However, I soon realized that true transformation hinges on a cultural shift.
Despite our revolutionary introduction of chat and chatbots as alternatives to traditional phone and email interactions, we initially faced significant resistance. It wasn’t until years later that these innovations gained widespread acceptance. This experience taught me that successful innovation goes beyond advanced technology; it requires nurturing a culture that is open to change for the adoption and impact to truly take hold.
Are you working on any exciting new projects now? How do you think that will help people?
I’m currently working on several projects that harness generative artificial intelligence for interactive conversations. My expertise is in hyper-personalization, expert-assisted tax and accounting workflows by leveraging large language models.
Thank you for that. Let’s now shift to the central focus of our discussion. In your experience, what have been the most challenging aspects of integrating AI into your business operations, and how have you balanced these with the need to preserve human-centric roles?
One of the foremost challenges is building trust in AI, both from the perspective of customers and employees. Customers need to feel confident that AI can reliably and securely handle their data and provide accurate assistance, while employees must trust that AI will enhance rather than replace their roles. To address these concerns, the focus must be on transparency.
In customer-facing roles that require a personal touch, such as customer support or consultation, it is crucial to ensure that the transition from AI to human assistance is seamless. This involves minimizing friction and ensuring that the handoff includes full context. For employees, AI should augment their workflow, enriching their capabilities without distracting them from their primary goal of helping customers.
Operational roles also face unique challenges with AI integration. To build confidence in AI within these teams, leaders must prioritize transparency and control. This means providing clear insights into AI processes and allowing human operators to step in and override automated decisions when necessary.
Ultimately, successful AI integration hinges on fostering a culture that embraces technological advancements while valuing the irreplaceable contributions of human expertise.
Can you share a specific instance where AI initially seemed like the optimal solution but ultimately proved less effective than human intervention? What did this experience teach you about the limitations of AI in your field?
One instance I’ve come across in my recent role was addressing simple yet emotionally charged customer challenges with our product or service.
Initially, we implemented AI-driven systems to handle these issues, assuming that standard policies and automated responses would efficiently resolve them. However, we soon discovered that while AI excelled at processing data and applying predefined rules, it struggled to empathize with customers’ emotions and provide the level of understanding and support they needed.
This experience taught us a valuable lesson about AI’s limitations in our field. While AI can excel at certain tasks, such as data processing and automation, it falls short in situations that require empathy, emotional intelligence, and a nuanced understanding of human emotions. Simple issues may seem easy to “solve” with AI, but the lack of a sympathetic ear on the other side can lead to the worst outcomes.
This further solidifies the importance of striking a balance between AI-driven automation and human intervention to deliver a more holistic and effective customer experience.
How do you navigate the ethical implications of implementing AI in your company, especially concerning potential job displacement and ensuring ethical AI usage?
The key is to emphasize not the displacement of jobs but rather the upskilling of the workforce. There are always areas where humans can leverage AI to enhance their capabilities and provide a uniquely human contribution. By investing in training programs and fostering a culture of continuous learning, for example, leaders can empower employees to adapt to new technologies and take on roles that leverage AI to their advantage.
As far as ethical AI usage is concerned, I’ll repeat a similar sentiment I made earlier. It’s important to prioritize transparency and data governance to ensure that our AI systems operate ethically and responsibly. This includes having strong principles in place to guide our AI initiatives, as well as mechanisms for accountability and oversight. Companies must also offer customers the choice to opt out of AI-driven interactions where appropriate, respecting their privacy and preferences.
Could you describe a successful instance in your company where AI and human skills were synergistically combined to achieve a result that neither could have accomplished alone?
Our live services combine AI automation and augmentation to benefit our customers and experts. Our platform allows customers to be matched with an expert most suited to them and automates and assists the customer and expert throughout the interaction. This synergy between AI and human expertise aims to deliver advice at scale by streamlining both customer and expert experiences.
Based on your experience and success, what are the “5 Things To Keep in Mind When Deciding Where to Use AI and Where to Rely Only on Humans, and Why?” How have these 5 things impacted your work or your career?
1 . Data Availability and Complexity
Data is foundational to the effective use of AI. I always say, “If you don’t collect it, you can’t measure it.” Because without data, we’d have no basis for training models or any way to assess their impact. And because AI, particularly machine learning models, rely on vast amounts of data to identify complex relationships and patterns that are not easily discernible through simple rules. Simple relationships might be manageable with straightforward algorithms, but as complexity and scale increase, these rules rapidly become unmanageable.
In my experience, data availability has unlocked increasingly sophisticated AI models and transformative experiences. For instance, in a project to optimize customer service operations, we initially implemented basic rule-based systems. However, as we began collecting more detailed data about customer interactions, we were able to train advanced machine learning models to predict customer needs and automate responses, significantly enhancing efficiency and satisfaction. The evolution from simple rule-based approaches to sophisticated AI-driven solutions was made possible entirely by the availability and utilization of extensive data.
2 . Customer Expectations
Customers who prefer to self-serve or are forgiving of occasional AI errors are more likely to benefit from AI-driven solutions. Demographics also play a significant role — early adopters of new technology and those open to change are typically more receptive to AI. Implementing AI in these contexts can streamline processes and enhance user satisfaction. In contrast, customers in more conservative fields require a delicate approach to AI integration. In my career, I worked on deploying an AI automation tool in a highly traditional industry. Initially, there was notable resistance due to concerns about reliability and change. By investing time in educating customers about the AI’s capabilities and consistently demonstrating its accuracy and benefits, we managed to build their confidence. This careful, customer-focused approach ultimately led to the successful adoption of AI, transforming the way they interacted with our services.
3 . Human Discretion/Exceptions
In areas where tasks frequently run into exceptions or require a high degree of human discretion and empathy, it is generally wiser to rely on human intervention. Certain situations demand nuanced understanding, emotional intelligence, and the ability to make judgment calls that AI cannot replicate. For example, legal matters and customer escalations often involve complex scenarios where the stakes are high and the consequences of errors are significant. In these contexts, the potential for negative outcomes from AI mistakes is too great, necessitating a human touch to manage liability and ensure satisfactory resolutions.
I recall a project involving customer service for a technology company. We explored using AI to handle customer complaints but quickly realized that many issues required nuanced responses and a deep understanding of individual circumstances. When customers escalated complaints, they needed to feel heard and understood — something that AI struggled with. The consequences of mishandling these situations were too severe, potentially leading to legal issues and loss of customer trust. Therefore, we chose to maintain human agents for these interactions, ensuring that empathy and discretion guided our responses, thereby managing liability concerns effectively and preserving customer relationships.
4 . Integration Opportunity
The potential to integrate AI into specific steps of a process and combine it with other steps is a critical consideration. If AI can be seamlessly embedded into various parts of a workflow and multiple AI capabilities can work in harmony, the overall efficiency and effectiveness of the process can be significantly enhanced. A powerful advantage of AI integration is the ability to create an end-to-end experience that feels seamless and cohesive. This often involves orchestrating data and context handovers between different AI systems to maintain continuity and coherence throughout the user journey.
In my experience, creating a seamless end-to-end customer experience using AI can lead to highly positive outcomes. For instance, we developed a comprehensive customer support system that integrated various AI capabilities, from initial query handling to advanced problem-solving. By ensuring smooth transitions between different AI-driven components, we created an interaction flow that felt natural and intuitive to customers. This not only improved response times and accuracy but also delighted customers with a consistently high level of service.
5 . Scale and Automation
One of the most compelling reasons to implement AI is its ability to scale efficiently with increased volume and across diverse use cases. AI can handle vast amounts of data and numerous simultaneous processes, making it ideal for streamlining complex workflows and reducing the burden on human resources. AI allows human workers to focus on more strategic and creative aspects of their jobs when it automates tasks, enhancing overall productivity and operational efficiency.
A key example from my career involved scaling AI solutions to address not just the most pressing issues but also the long-tail problems that typically require significant manual intervention. Initially, we focused on automating solutions for high-frequency customer queries, which immediately reduced response times and improved satisfaction. However, as we expanded AI’s reach to tackle less common but still impactful issues, we unlocked even greater potential. The overhead associated with managing these long-tail issues manually was overwhelming, but AI handled them with ease, leading to a significant reduction in operational costs and a more robust, comprehensive service offering. This demonstrated how effectively scaling AI can transform an organization’s ability to manage both common and complex challenges.
Looking towards the future, in which areas of your business do you foresee AI making the most significant impact, and conversely, in which areas do you believe a human touch will remain indispensable?
Personalized and proactive product experiences, coupled with customer-assistive technologies, will revolutionize how we engage with consumers. AI-driven support systems will enhance efficiency and effectiveness, providing timely assistance and solutions to customer queries. Additionally, AI will play a crucial role in augmenting human capabilities, particularly in support and consultative roles, empowering employees to deliver enhanced services through data-driven insights and recommendations.
Despite the advancements in AI technology, there are areas within our business where a human touch will remain indispensable. Expertise and certification, particularly in specialized fields, require human judgment, intuition, and experience that AI algorithms cannot fully replicate. Likewise, empathy and handling of exceptions in customer interactions necessitate human intervention, as AI may struggle to comprehend and appropriately respond to nuanced emotional cues. Human executives also bring strategic vision, creativity, and critical thinking to the table where AI cannot.
In emerging areas where AI has yet to mature or where ethical considerations are paramount, human oversight and intervention are essential to ensure responsible and ethical outcomes.
You are a person of great influence. If you could start a movement that would bring the most amount of good to the most amount of people, what would that be? You never know what your idea can trigger. 🙂
In addressing the prevalence of toxic business cultures, I firmly believe in the power of positive leadership. Positive leadership fosters an environment of trust, collaboration, and growth, where individuals feel valued and empowered to contribute their best work. Particularly during times of change and transformation, such as those marked by fear, uncertainty, and doubt within organizations, positive leadership becomes even more crucial. Positive leaders like myself can navigate challenges effectively, inspire confidence in their teams, and cultivate a culture of positivity and progress.
How can our readers further follow your work online?
They can check out my website or follow me on LinkedIn and X.
This was very inspiring. Thank you so much for joining us!
About The Interviewer: Kieran Powell is the EVP of Channel V Media a New York City Public Relations agency with a global network of agency partners in over 30 countries. Kieran has advised more than 150 companies in the Technology, B2B, Retail and Financial sectors. Prior to taking over business operations at Channel V Media, Kieran held roles at Merrill Lynch, PwC and Ernst & Young. Get in touch with Kieran to discuss how marketing and public relations can be leveraged to achieve concrete business goals.
C-Suite Perspectives On AI: Steven Kawasumi Of Intuit On Where to Use AI and Where to Rely Only on… was originally published in Authority Magazine on Medium, where people are continuing the conversation by highlighting and responding to this story.