Leveraging Artificial intelligence (AI) and machine learning (ML) in insurance
The insurance sector continues to evolve, some of it driven by the changing customer behavior and some of it driven by the availability of new technological solutions. Big data and analytics are helping organizations in understanding customer needs better. In recent years, artificial intelligence and machine learning have seen increased adoption driven by the increased digitization and digital means of interaction with customers. Even in an industry like insurance, which happened to be physical intensive, the covid-19 pandemic has brought about increased adoption of digital methods of fulfilment in the insurance life cycle. With this rapid acceleration in the adoption of digital technologies, AI and ML is becoming an important component of every organization’s technology arsenal.
At ICICI Lombard, we continue to build technological solutions that not only make it easier for insurance customers to purchase or renew their insurance but also simplify their servicing needs. Artificial intelligence and machine learning is playing a significant role in achieving this as we have started to use these technologies in simplifying the customer journey across the insurance lifecycle. Customer centricity continues to drive the adoption of most of our digital solutions and is the fulcrum for identifying these innovative technology-based solutions.
Leveraging AI based BOTs for quote generation and policy issuance
Today, our customers and partners can instantly get answers, quotes or can easily complete various transactions without any manual intervention through our AI based chatbot platform – MyRA. e.g. customers can buy TW insurance, renew their health and motor policies on this platform and even find out claims related information. Similarly, we are leveraging Natural language processing (NLP) and Robotic process automation (RPA) technology for automating the manual process of quote generation and policy booking for our corporate and SME customers. For specific SME products, over 90% of our policy issuance is done through this automated system. Through our NLP and cognitive services enabled chatbot, our retail customers can have the same chatbot experience through the website or through Whatsapp. We aim to open this same experience on other social media and networking platforms as well.
Leveraging AI and ML in motor inspection
One of our first AI solutions that was developed and deployed was the break-in AI solution. This solution helps customers whose policies have lapsed to simply click and upload photographs of their car, using our app. An AI solution on the cloud helps analyze these images to identify damages to the vehicle and automatically decides whether the policy needs to be issued or needs to be looked at by a manual inspector. Prior to this solution being implemented, it used to take between 24-72 hours to complete this process of physical survey. Currently, this process is now done within minutes and 75%+ of our break-in inspections today get processed straight-through first-time using this solution. During the peak of the pandemic, we were able to accommodate 3 times the volumes of such inspections driven by extensible architecture on the cloud.
Leveraging AI and ML in health claims authorization
In health claims, we are using AI and ML to process the cashless claim requests made by hospitals. The policy related information, doctor’s diagnosis and the course of action recommended by the doctor is ingested in the AI algorithm, which decides the admissibility of the case. This was decided by a doctor at our end, earlier. Based on the case admissibility, an ML program decides on the optimum claim amount to be sanctioned based on the overall policy sum insured, the diagnosis, the hospital, the proposed length of stay and other factors. The ML program sends a message with the authorized amount back to the hospital. Where the algorithm is not able to authorize the claim, a doctor takes a decision and that gets fed back to make the algorithm better over time. The final sanctioning amount is established during the discharge of the patient from the hospital. This entire initial sanctioning process takes about 90 seconds. The same process when done manually would take 3 to 4 hours. Over 60% of our corporate cashless authorizations are done straight-through using this solution.
Using AI and ML in fraud claims identification
While we strive to make things easier for the customer during claims settlement, it is equally important for us to identify customers who try to take undue advantage of the system. So, each one of our digital claims solutions is backed by powerful fraud identification solutions that help us in identifying new patterns of fraud that keep emerging. Earlier, we would flag off claims that needed scrutiny basis rules and triggers that were put in the system. This would be investigated thoroughly and marked as fraudulent or kosher basis the investigation. Today, with significant data availability and technology, we have started using AI and ML based fraud detection models to predict and highlight probable fraudulent claims in real-time. All this happens immediately, once the claim is intimated and logged within the system. These algorithms are helping us in identifying and prevent frauds faster giving us 30% higher efficiency.
Leveraging speech-based AI solutions in medical tele-underwriting
At ICICI Lombard, we have been leveraging BOT based technologies for a few years to help in processing and resolution of customer queries, for quote and policy issuance and even in claims. Recently, we have also started leveraging voice-based technology that helps in guiding customers through the policy purchase process, especially during renewals. With the help of an InsurTech partner, we have co-created a solution using cognitive services and natural language processing that helps us in understanding customer responses at renewal and guides the customer at every step of the process. We have also started using the same technology in the field of medical tele-underwriting where the customer is asked several medical questions by an interactive voice BOT and the customer’s responses are recorded immediately. Basis the voice responses given by the customer to each of the underwriting questions, the BOT then applies the underwriting lens to segment customers. It helps decide which customers are a safe risk and can be issued a policy, versus those that need to be evaluated further by a medical professional. This medical evaluation is then done by a qualified doctor over the phone in most cases. This process has, in many cases, eliminated the need for medical tests to be done by prospective customers.
Newer technologies on the digital front, including Artificial Intelligence and Machine Learning will continue to play a major role in providing innovative solutions from a customer experience standpoint. As consumer expectations and behaviors change, we believe that the insurance industry will continue to change by embracing newer technology, be it through an AI chatbot that acts as a Virtual Insurance Advisor, or an AI detection model that acts as a Digital Claims Adjuster or a ML model that acts a smart underwriter. There will be prolific use of AI, ML and cognitive services that will help address needs such as digital adoption, customer experience management, operational efficiency, underwriting profitability, claims optimization and much more.