Women Leaders in AI: IT Leader Seema Gaur on putting AI to work in insurance
Seema Gaur is the head of IT at one of the largest general insurance companies in India. She was honored in 2019 by IBM as a pioneer in putting AI to work in her business. Seema and her team are working to bring in the best technology to make the user journey more efficient, improve the agent experience, and process claims much faster. We spoke with Seema about how AI is transforming her industry now, and how it can continue to transform it in the future.
Tell us a little about IFFCO Tokio General Insurance Company.
We are a non-life insurance company operating within India. We are a joint venture between IFFCO, the largest fertilizer company in the country, and Tokio Marine Holdings of Japan, one of the top 10 insurance companies in the world, and the largest private insurer in Japan. We offer a variety of insurance policies including health, motor, home, travel and casualty, etc.
We are a $1.1B company and are growing at a rate of 14 – 15% year over year.
What is your role at IFFCO?
I’m the Executive Director and Head of IT at IFFCO Tokio General Insurance Company Ltd.
What is your mission and your objectives as head of IT?
What drives me is the need to serve customers, to enhance the customer experience as well as the experience of our intermediaries—our agents. I want to be sure we’re serving customers and agents in a manner which provides an exciting experience in buying our products and services. These experiences are omni-channel, because there are all sorts of devices being used by people: tablets, smartphones, laptops, etc. Our processes need to be as simple as possible, and we need to make it easy for people to understand our products and buy them online. That goes for our other services as well. Our goal is to have claims settled in the minimum time possible to the satisfaction of customers.
My mission is to identify the best technology on the market and adopt it into the systems we’re using, with the goal of providing seamless availability to end users.
Can you describe some of the IT operations challenges you see in your organization today?
We face a real challenge with scale, as we continue to expand our IT environment to meet the evolving needs of our customers. As we adopt new technologies, our ability to deploy these solutions quickly is important. There are many niche technology solutions out there that don’t have a good support system in place, so when you’re implementing and need help deploying it, the human skillsets and knowledge you need are scarce. So, when it comes to scaling, it all falls apart.
We also face some challenges with skills and retention within our team. Technologies that improve the productivity of the team and that can be easily integrated, deployed and managed are particularly attractive to us.
How do you see AI helping to address some of these challenges? How does AI provide new opportunities for your organization?
AI helps us automate processes that would otherwise need to be done manually. For example, we’re using AI for visual recognition and image processing to assess damages in motor vehicle accidents and process claims immediately. Without AI, support would need to come to the scene and process the claim manually.
Another challenge AI can address is correlating across data to help us detect issues earlier, such as for fraud detection and mitigation. Due to our large population, India is prone to fraud—health claims fraud and motor claims fraud. These types of fraud are typically only detected after payment. What’s needed is a solution that detects fraud at the first stage of approval. When a large claim is initiated—for example, someone is pretending that they’re sick, claiming an amount of money as recompense for a disease or ailment they don’t have—these cases become difficult to monitor and investigate as they progress. We need AI to help with fraud analysis, so we can catch these cases earlier without too much human intervention.
We are also exploring the use of AI for contact center automation, using natural language processing and understanding for speech recognition, to account for the many dialects in the country.
What should AI technology vendors keep in mind when designing AI solutions for IT? Are there specific requirements or needs?
Solutions should be easy to use. Models should be developed to be easy to update and manage. Many times, when working with a model you’ll discover that the self-learning that was claimed is not up to the mark. Such cases need to be easy to retrain and update, with the support in place to assist in case in-house skills are not available.
If you could instantly invent the perfect AI for IT solution, what would it do?
It would have the ability to mine the vast amounts of data that we have residing in different places, to get as close to accurate predictions as possible. Self-learning should be automated to reduce the dependency on human support staff. Models would rebuild themselves frequently, taking into account all the new data they learned along the way, retraining with greater accuracy, and all of this would be done automatically. As I mentioned before, we face challenges with productivity and skills, where AI help would be valuable.