Indico Data elevates insurance AI with out-of-the-box capabilities for global carriers. Read more.
  Everest Group IDP
             PEAK Matrix® 2022  
Indico Named as Major Contender and Star Performer in Everest Group's PEAK Matrix® for Intelligent Document Processing (IDP)
Access the Report

Unstructured Unlocked season 2 episode 17 Sam Worthington, Managing Director, Aon

Watch Tom Wilde, CEO at Indico Data, alongside Michelle Gouveia, VP at Sandbox Insurtech Ventures, in season 2 episode 17 of Unstructured Unlocked with Sam Worthington, Managing Director P&C Technology, Aon.

Listen to the full podcast here: Predictive Analytics and Climate Risk: Aon’s Sam Worthington on Data-Driven Insights

 

Michelle Gouveia: Hey, everybody. Welcome to a new episode of Unstructured Unlocked. I’m co-host Michelle Gouveia.

Tom Wilde: And I’m co-host Tom Wilde.

Michelle Gouveia: We are thrilled to be joined today by Sam Worthington, Managing Director of P&C Technology at Aon. He’s joining us from London today. Sam, thanks so much for taking the time to talk with us.

Sam Worthington: Thank you, Michelle, and thank you, Tom. It’s great to see you both again.

Michelle Gouveia: You too. Before we jump into the long list of questions we have for you, that was a very brief intro on my part. Could you share a bit more about your background, your experience, and your role at Aon?

Sam Worthington: Absolutely. Thanks for inviting me to your Christmas special—I’ll try to do it justice! I’m the Managing Director of P&C Technology within Aon. I oversee the technology we license to the P&C market for financial modeling. This includes the Aon Pricing Platform for pricing, and the Tyche and Rama platforms for capital modeling and reinsurance optimization.

We also work closely with other parts of the business, like impact forecasting, which focuses on event response and catastrophe modeling. A lot of my role involves supporting risk transfer analytics for our clients.

Tom Wilde: That’s a perfect lead-in to today’s discussion. Given your role, we’d like to dive into the intersection of data, analytics, technology, and climate risk. Specifically, how risk is modeled and anticipated, and how insurers adjust their approach to pricing risk amid extreme weather events. What trends are you seeing in claims exposure related to these events?

Sam Worthington: That’s a great question, Tom, and very topical. We’ve seen significant losses from catastrophes like hurricanes, floods, wildfires, and severe storms in the U.S. While these events are often attributed to climate change, a substantial factor is increased exposure due to urbanization. For example, hail damage has risen significantly with the growth of solar farms, as solar panels are particularly vulnerable to hail.

It’s challenging to separate the impact of climate change from other contributing factors, but there’s clear evidence that climate change exacerbates certain events like wildfires and floods. Urbanization and reduced natural defenses also increase exposure, which insurers must account for when pricing risk. The challenge lies in validating models to ensure they reflect both changing risk and exposure levels accurately.

Michelle Gouveia: That point about validation is fascinating, especially with the rise of generative AI and data analytics in climate risk assessment. What’s changed in the last five to ten years regarding climate data and predictive capabilities? Are there any solutions demonstrating a significant improvement in accuracy?

Sam Worthington: The key challenge is that risk is evolving, and historical experience is not always predictive. Data sets for volatile, infrequent events are sparse. To address this, catastrophe models break the problem into components: hazard, vulnerability, exposure, and financial impact.

Technology has improved geospatial mapping and LiDAR capabilities, which help model exposure more accurately. For example, flood modeling has become far more precise in predicting water flow and affected buildings. Advances in machine learning have also enabled startups to forecast industry losses with surprising accuracy, sometimes outperforming traditional methods.

Tom Wilde: You touched on something interesting. Are we at greater risk, or are we engaging in riskier behavior? For instance, building solar farms in hail-prone regions or towns in floodplains seems inherently risky. How do these decisions factor into insurers’ calculations?

Sam Worthington: It’s a mix of both. Risky decisions, like building in floodplains, are influenced by various factors, including economic incentives. Ideally, insurance pricing should discourage risky behavior by making it expensive to insure, but other priorities often override this. Over time, we learn from events and adapt, such as building hurricane-proof homes or redesigning infrastructure after disasters.

Tom Wilde: That brings up another point—how reliable is historical data in predicting future risks, given the increasing uncertainty caused by climate change?

Sam Worthington: Mark Twain once said, “History doesn’t repeat itself, but it rhymes.” Historical data is the foundation of modeling, but climate overlays are now used to simulate potential future scenarios. Comparing multiple models provides a more nuanced understanding than relying on one.

Michelle Gouveia: Speaking of data, how is real-time data, like IoT sensors in buildings, being incorporated into risk modeling?

Sam Worthington: Real-time data is incredibly useful for monitoring and loss mitigation. For example, sensors tracking cargo temperature can prevent losses or speed up claims processing. However, in traditional one-year policies, real-time data has limited application beyond testing risk assumptions.

Tom Wilde: What role do governments and regulators play in supporting insurers and mitigating risks?

Sam Worthington: Governments and regulators play a crucial role in providing scientific data, setting standards, and implementing defenses. The insurance industry uses this information to refine hazard modeling and pricing. It’s a collaborative relationship where both parties influence each other.

Michelle Gouveia: Insurers also seem to be taking a more proactive role in loss prevention. What’s Aon’s approach to helping clients mitigate risk before a claim occurs?

Sam Worthington: Insurers are increasingly involved in risk mitigation. For example, they might warn customers to move their cars during potential flooding or adopt pay-as-you-go insurance models using telematics. It’s a shift from being reactive to proactive.

Tom Wilde: Let’s talk about AI. How is AI shaping the future of risk modeling and insurance, particularly in the context of environmental risks?

Sam Worthington: AI has broad applications in improving data ingestion, claims processing, and policy customization. In climate-related contexts, AI helps analyze datasets and predict risks more effectively. While there’s still room for growth, AI is integral to reimagining how insurance operates, tailoring solutions to individual customers in real time.

Tom Wilde: That’s a great note to end on. We’ve been talking to Sam Worthington, Managing Director of P&C Technology at Aon. Sam, thanks for such an engaging conversation.

Sam Worthington: Thank you, Tom and Michelle. I’ve thoroughly enjoyed it.

Tom Wilde: And with that, I’m co-host Tom Wilde.

Michelle Gouveia: And I’m co-host Michelle Gouveia.

Tom Wilde: Thanks for listening to Unstructured Unlocked.

Check out the full Unstructured Unlocked podcast on your favorite platform, including:

Subscribe to our LinkedIn newsletter.

Resources

Blog

Gain insights from experts in automation, data, machine learning, and digital transformation.

Unstructured Unlocked

Enterprise leaders discuss how to unlock value from unstructured data.

YouTube Channel

Check out our YouTube channel to see clips from our podcast and more.