How Australian insurers are using AI and your data to price policies and process claims

Why Insurers Are Feeding Members’ Data to AI Models

Australian insurers are increasingly disclosing in their privacy policies that customer data is being used to train AI models. Notable companies like NobleOak, Australian Unity, Rest, HESTA, Pet Circle, Resolution Life Australasia, and AXA XL have updated their policies to reflect this practice. However, most of these disclosures lack specifics about the types of personal information processed or the exact AI use-cases, except for QBE Australia, which mentions using machine learning tools to assess the risk of claim denial.

Even insurers without explicit AI-related privacy disclosures have piloted or implemented AI across backend and customer-facing operations. These deployments aim to reduce manual workloads and improve personalization. For instance, Suncorp has rolled out hundreds of AI models across sales, service, claims, and supply chain management. Insurance Australia Group (IAG), QBE Australia, MLC Life Insurance, and Resolution Life Australasia have developed AI tools that partially automate claims assessment, a high-stakes application.

AI-Enabled Pricing

AI is widely used for actuarial tasks such as risk, cost, and demand modeling. The Australian Securities and Investment Commission noted these as common AI applications in the insurance sector. However, AI's use in pricing has attracted regulatory scrutiny. IAG is currently under investigation for allegedly using an algorithm to predict customer renewal likelihood and adjust premiums accordingly, potentially leading to unfair pricing.

QBE Australia referenced deploying underwriting AI co-pilots but declined to provide details. Meanwhile, Suncorp’s AI transformation executive, Priyanka Paranagama, shared how geospatial imaging combined with AI has halved the number of property questions needed during home insurance quoting, streamlining the customer experience.

Leveraging Non-Traditional Data Sources

Suncorp’s AI quoting system uses aerial imagery to automatically populate property details, such as roof material and the presence of swimming pools or solar panels. Other insurers are tapping into unconventional data sources to enhance pricing accuracy and automation.

• NRMA (owned by IAG) offers an opt-in feature for car insurance members that collects location data and driving behaviors, including acceleration, braking, turning, speeding, and mobile device usage while driving.
• MetLife Australia is exploring health risk modeling by integrating productivity and absenteeism data, alongside lifestyle factors like exercise, travel, and driving habits, to personalize pricing.
• Woolworths’ Everyday Rewards card users who buy car insurance through Hollard Insurance may receive premium discounts based on predicted safe driving, inferred from purchase patterns and shopping times.

Claims Processing and Assessment Using AI

MLC Life Insurance employs a dynamic triage model that analyzes customer data such as file notes and phone calls to alert claim handlers when a claim’s status changes or if critical information is missing.

In 2017, AIA Group reported a 40% reduction in claims turnaround times and over 99% accuracy in claims eligibility processing using IBM’s Watson AI platform. They later trialed a machine learning system comparing new claims against a database of 60,000 historical claims to accelerate assessment, though current usage status is unclear.

These AI applications enhance efficiency but raise questions about data privacy, transparency, and fairness. Insurers must balance operational gains with clear communication to policyholders on how their data is used.

For insurance professionals interested in understanding AI applications and their implications, exploring targeted AI training can provide practical insights and skills. Resources like Complete AI Training’s courses for insurance professionals offer relevant learning paths.