The rise of artificial intelligence (AI) has been nothing short of revolutionary. Within underwriting,
Since a significant portion of their time is tied up to administrative tasks and sales support, underwriters are seen as not doing much underwriting work. However, when they do, 40% of the decisions made are based on either outdated or incomplete data. Translated into financial loss, the impact is staggering. These inefficiencies are estimated to cost the insurance industry to the tune of $85 billion to $160 billion over the next 3 years.
Underwriting challenges
The biggest challenge to underwriters is how to utilize and access structured and unstructured data collected through various sources such as financial reports, social media, telematics, IoT and third-party databases. Despite the vast amount of data available, underwriters are often than not overwhelmed by the sheer volume and complexity of information, which could lead to analysis paralysis.
AI technologies such as machine learning and natural language processing (NLP) empower underwriters by streamlining data access and enhancing data capabilities. This ultimately equips them with smarter insights to make more informed decisions and at the same time enhance the accuracy of risk assessments.
Machine learning algorithms, for instance, analyze large volumes of historical and real-time data to identify patterns, trends and correlations that may not be apparent to the naked eye. This provides underwriters with smarter actionable insights on risk scoring and pricing.
Capturing unstructured data
In today's digital age, about 90% of data generated daily is unstructured. This includes emails, social media comments, images, audio recordings and chatbot conversations. However, the explosion of digital content creation on platforms such as TikTok and Instagram adds to the growing volume of unstructured data generated, as more people share videos, live streams and stories as part of their need for social validation, self-expression and instant gratification.
This unstructured data offers rich insights that traditional structured data may overlook. The ability of NLP to extract meaningful, smarter insights from unstructured data offers underwriters a deeper understanding of customer preferences and market dynamics, which effectively enriches risk assessment and aids in identifying emerging risks.
As AI technologies evolve, the future of underwriting centers around the harmonious union of AI and human judgment, which promises a more nuanced approach to underwriting. While all the routine and repetitive tasks in underwriting are automated by AI, human underwriters complement AI underwriting in terms of handling complex decision-making that extends beyond data-driven decision-making. This includes the contextual understanding of unique cultural differences and social norms that AI will likely never be able to fully replicate or recreate but is crucial when making informed underwriting decisions. Despite AI's advanced capabilities, applying professional judgment and intuition in risk assessment are human attributes that are not easily computed or simulated by technology.
