RGA’s pioneering study demonstrates how GenAI-powered tools are transforming insurance underwriting. By analyzing diverse evidence types to detect hidden health impairments, GenAI promises to revolutionize risk assessment and pricing in the industry.
Generative artificial intelligence (GenAI) has officially arrived in insurance underwriting. A groundbreaking RGA study sheds light on the transformative potential of GenAI and explores how GenAI-powered tools can significantly enhance the processing efficiency of underwriting evidence.
What makes this case study so pioneering is its large-scale approach, using an unprecedented number of records. This led to more meaningful, substantial results and analysis.
Protective value in insurance underwriting refers to the ability of various types of evidence to reveal critical health impairments that might otherwise go undetected. Traditionally, establishing protective value has been a manual and time-consuming endeavor, with underwriters reviewing individual cases to determine how each piece of evidence contributes to identifying mortality risk factors.
However, the rise of structured data and advanced GenAI technologies have opened new doors for streamlining this process, introducing cutting-edge tools that are game changers in medical record analysis.
RGA’s study tested the capabilities of DigitalOwl, an advanced GenAI tool that employs sophisticated algorithms to summarize medical records. Its ability to extract key impairment data from underwriting documents and organize it into a structured format allows for large-scale analysis across multiple impairments and evidence types.
This marks a significant leap forward in the field of protective value analysis. By enabling a more systematic and efficient approach, GenAI tools help insurers make faster, more informed decisions while potentially reducing costs.
The RGA study used approximately 2,000 underwriting applications, submitting them to DigitalOwl with various permutations of evidence. Evidence sources analyzed included:
The output identified 40 key impairments, assigning each a mortality severity rating. This granular analysis allowed researchers to conduct multiple pairwise comparisons, evaluating the effectiveness of different types of underwriting evidence in identifying key impairments.
While digital evidence sources showed significant promise, the study also compared their effectiveness to traditional underwriting methods. The results revealed that APS and insurance lab data still captured more impairments than digital sources alone.
This finding underscores the continued importance of comprehensive evidence gathering in underwriting and suggests that an ideal approach may be a hybrid model, combining the speed and efficiency of digital sources with the depth and detail of traditional evidence.
The RGA study represents a significant step forward in understanding how GenAI can enhance the underwriting process. By enabling large-scale, systematic analysis of protective value, these tools are paving the way for more efficient, accurate, and cost-effective underwriting decisions.
As GenAI models continue to advance, their ability to interpret complex medical data will only increase.
This progression could lead to even more nuanced underwriting processes, potentially allowing insurers to distinguish the relative importance of various evidence types at the impairment level.
The RGA case study demonstrates the immense potential of GenAI-assisted underwriting to transform how risk is assessed and managed. By harnessing the power of these tools, insurers can streamline processes and gain deeper insights into applicants' health profiles.
However, while AI shows great promise, it is not a complete replacement for human expertise. The most effective approach will likely be a synergy between AI capabilities and human judgment, combining the efficiency and pattern recognition of machines with the nuanced understanding and decision-making skills of experienced underwriters.
Moving forward, continued research and development in this field will be crucial. The insurance industry must remain adaptable, embracing new technologies while maintaining the fundamental principles of accurate risk assessment and fair pricing. Those who can effectively leverage GenAI-assisted underwriting may find themselves at a significant competitive advantage, better equipped to serve their clients and manage risk in an increasingly complex world.
RGA is unlocking the future of underwriting through data analytics and technology. Learn more about digital underwriting optimization.
