With a solid understanding of AI's capabilities — both in use and potential challenges — insurers can go into 2019 with a more realistic view of its potential. (Shutterstock)

Digital transformation is now a board-level agenda at many top insurance companies. All the digital tenets that allow insurers to become more competitive — analytics, automation, artificial intelligence (AI), and customer experience — are already here and available today. As they adopt these, insurers will become smarter, quicker, and more proactive in making underwriting decisions, managing risk, detecting fraud, and delivering positive customer experiences. Moving into 2019, some insurers may wonder if they are making the best use of AI. Much of the uncertainty stems from questions around the best use case for applying AI. To determine AI's impact in the next year — and for years to come — companies need to understand first what is happening today. |

AI in insurance today

It is possible to separate AI applications in insurance into four segments: |

1. Language: Using natural language processing, insurers can extract information from the unstructured data, such as text within first notice of loss documents, and turn it into structured data. This makes it easy to classify claim submissions for quick resolutions. With the right system of engagement, insurers can automatically check into numerous legacy applications to see if the customers submitting the claim has a policy in good standing and whether they have coverage for this type of loss. For underwriters and brokers, AI can extract and classify information, thus automating the laborious process of data intake, and allowing underwriters to make a go-no go analysis more quickly. Add machine learning to the mix and claims processing technology can learn and guide future decisions.
2. Conversation: One of the fastest growing applications of AI is the emergence of chatbots. Just as people can string together threads of conversation, AI can perform intent analysis on why people contact the call center. For example, when customers call to verify the coverage in their policy, a system can pull the right policy, pick up all documentation for eligibility, and communicate back to the client through voice, email or chat. Such machine intelligence makes subrogation easier, too. Some insurers overlook subrogation because it requires reviewing extensive notes from assessors. With AI doing the heavy lifting, subrogation is not only fast, but also predictive.
3. Sight: In automotive claims, a picture is worth a thousand dollars (sometimes more). With computer vision, AI can review a photo of an accident and determine the level of loss. By analyzing millions of photographs of previous accidents and their adjustments, which serve as labeled data for machine learning, AI can predict the payout amount for auto claims, significantly reducing the time between claim submission and payout, and at the same time improving the customer experience.
4. Knowledge: Companies can use AI to develop knowledge graphs specific to each type of incident (hurricane, earthquake, etc. Insurers can develop these knowledge graphs across the enterprise and use it to augment AI's results. By combining the collective intelligence of experienced team members who have deep understanding of the insurance industry with machines that have sophisticated machine learning algorithms, they can make the most out of the structured and unstructured data they are sitting on, turning it into value. Most of these applications are still early in development, but will be critical for insurers.

|

Preparing for the challenges ahead

Looking to 2019, AI will continue to shape how insurers approach the four areas above. But companies should consider possible challenges in their implementations to ensure a smooth year ahead, including: With a solid understanding of AI's capabilities — both in use and potential challenges — insurers can go into 2019 with a more realistic view of the potential AI can have in their digital journeys. Sanjay Srivastava ([email protected]) is the chief digital officer at Genpact. See also: Top insurance technology issues nagging at industry leaders