Zurich leverages AI to transform life insurance underwriting

Zurich Financial Services Australia (Zurich) has completed the first phase of a collaboration with the University of Technology Sydney (UTS) to create artificial intelligence (AI) models for streamlining the underwriting of life insurance applications.

After working for over six months with UTS Rapido, a research and development centre at the university, the AI models now enable Zurich to make immediate underwriting decisions on applications with mental health disclosures, eliminating the need for a doctor’s report that previously took up to 22 days.

“By using AI models to assist the underwriting process, Zurich can reduce the need to access doctor’s reports, eliminating significant friction within the application process and reducing applicant concerns around privacy,” said Jacqui Lennon, head of retail at Zurich Australia & New Zealand.

Zurich’s AI models for life insurance underwriting

The AI models use anonymised data from the past seven years of applications to identify factors associated with mental health exclusions, including:

• non-participation in sports or recreational activities;
• presence of chronic health issues beyond mental health, such as pain, cancer, or heart disease;
• hospitalisation or medical treatments in the past five years; and
• family history of conditions like diabetes, cancer, or heart disease.

Mental health-related claims at Zurich

Currently, around 25% of Zurich’s life insurance applications disclose a mental health condition.

In 2023, mental health was the third most frequent cause of claims at Zurich, accounting for 20% of all claims and $255 million in payouts.

For some retail life insurance products, mental health is the leading or second most common claim reason, comprising up to 35% of claims.

“Using AI allows Zurich to analyse thousands of data points, which would not be possible for a human to do. This project is a key step in Zurich’s journey to being Australia’s leading digital insurer, and importantly, it allows us to provide protection to more Australians at a time when they need it most,” said John Kim, chief data officer at Zurich Australia & New Zealand.

AI models expected to enhance Zurich’s underwriting process for other conditions

In addition to underwriting, the insights from the AI models will inform Zurich’s ongoing development of preventative health and wellness services.

Moving forward, these models could potentially enhance the underwriting process for other conditions such as cancer, cardiovascular diseases, or musculoskeletal disorders, provided sufficient data is available, according to Lennon.

Zurich’s AI models will complement decisions made by its underwriting team and the existing rules engines, adhering to the company’s policy of maintaining human oversight in AI processes.

In separate news, Zurich Insurance recently decided to halt new underwriting for oil and gas projects.