Risky Science Podcast

Folgen

• 2026 Hurricane Season: Risk, Markets, Models 17.06.2026 1Std. 6Min.

  This is the audio recording of our June 11 panel on the 2026 Atlantic hurricane season — three people who actually price this risk for a living, in conversation for an hour.The ground they cover: where the cat models miss, how the ILS market is pricing the 2026 season, and how a new prediction-market contract could change what it means to "test" a hurricane model.
• Ebola, Statistics, and What Pandemic Science Can Teach Markets 27.05.2026 54Min.

  The Democratic Republic of Congo is in the middle of its 17th Ebola outbreak since 1976 and the WHO has declared a public health emergency of international concern. There are no approved vaccines. No approved treatments. Hundreds of suspected cases emerged before the outbreak was even confirmed. And the surveillance infrastructure needed to track it is operating in an active conflict zone.For reinsurers and ILS investors, pandemic risk has always been the peril that's hardest to model, hardest to price, and hardest to transfer. The triggers don't work and the data is noisy.. And the market has largely walked away from the problem since COVID.My guest today has spent his career working on exactly that problem from the science side. Dr. Ben Swallow is a lecturer in statistics at the University of St. Andrews, where he works at the intersection of Bayesian inference, uncertainty quantification, and epidemic modeling. He co-led the uncertainty quantification effort for the Scottish government during COVID. He's worked on Ebola outbreak analysis in West Africa. And he's currently part of the Isaac Newton Institute's program on pandemic preparedness.Subscribe to Risk Market News
• Wildfire's Garbage-In Problem With Brian Bastian 20.05.2026 25Min.

  California’s homeowners insurance market, backed by privae capital, is still in retreat. The publicly-backed FAIR Plan is financial buckling. And somewhere in the gap between what the cat models capture and what's actually happening on the ground at the property level, there's a mispricing problem that nobody has fully solved yet.In this episode of The Risky Science Podcast that wraps up my series of conversations from ClimateTech Connect in Washington this past April, I talk with Brian Bastian, Head of Product at Green Shield Risk Solutions — the analytics and MGA operation that's betting mitigation-first underwriting is the answer the admitted market can't quite get to yet. Subscribe to Risk Market News
• Modeling Every Risk for Every Client with Willis' Ben Fidlow 13.05.2026 25Min.

  Ben Fidlow is a Fellow of the Casualty Actuarial Society and leads analytics and risk advisory at Willis. In this episode, he explains how brokerage modeling differs fundamentally from carrier or vendor modeling — it's about what risk means to a specific client, not an aggregate book. He walks through Willis's expanded partnership with Moody's RMS, which lets his team layer their own climate extrapolations on top of current-day model outputs while retaining the ability to explain every step to clients. He also shares where he thinks AI is creating its most immediate value (converting unstructured client data into structured formats at scale), why federal data erosion is a real near-term problem, and what it would take for a direct capital market risk marketplace to eventually disintermediate both insurers and brokers.Subscribe to Risk Market New
• The Wrong Model for the Wrong Job With Roy Wright 06.05.2026 24Min.

  In this episode of Risky Science, recorded at ClimateTech Connect in April, IBHS CEO Roy Wright breaks down why catastrophe models were never designed to price individual risk, why mitigation only works at the neighborhood level, and why insurance markets start to fail when price signals drift away from underlying risk.
• The LA Fires and the Risk Market Value Chain With Joy Chen 28.04.2026 42Min.

  The Eaton and Palisades fires are now the most expensive wildfire disaster in U.S. history — and what's happening in Los Angeles right now is a real-time stress test of the entire insurance value chain. From how models priced the risk, to how policies were written and sold, to how claims are being managed on the ground.Joy Chen is a former deputy mayor of Los Angeles with a finance background, and she runs the Every Fire Survivors Network — 10,000-plus Eaton and Palisades survivors. Her group has spent the last year and a half documenting delays, denials, and underpayments among insured survivors. Among the statistics they point to: a $300,000 median gap between expected insurance payouts and actual rebuilding costs, and a recovery pace slower than any previous California wildfire on record — including the Camp Fire.The question is whether these are simply the normal costs and challenges of a large catastrophe, or market signals about model adequacy — and what happens to market confidence, and ultimately to capacity, when the system fails at scale.Subscribe to Risk Market News
• How Catastrophe Models Work and Where They Fall Short With Anil Vasagiri 22.04.2026 27Min.

  This episode is part of a series of live conversations recorded at Climate Tech Connect 2026 in Washington, D.C.Anil Vasagiri, Head of Risk Data Solutions at Swiss Re is a rare combination of technical depth and commercial perspective to catastrophe risk — he came up through Verisk, where he held senior roles in product management and data strategy, before joining Swiss Re in 2020. Since then, he's led the development of some of the industry's most sophisticated tools for understanding physical risk at the location level, including Swiss Re's acquisition of flood modeling firm Fathom.Subscribe to Risk Market News
• Why Mixing Catastrophes With Prediction Markets Is More Dangerous Than It Looks With Jamie Pietruska 15.04.2026 48Min.

  The LA wildfires burned more than a hundred thousand acres. They destroyed thousands of homes. And while they were still burning, people were placing bets on them.Not insurers. Not reinsurers. Not catastrophe modelers running exceedance probability curves. Anybody with a crypto wallet and an opinion.That's the world of prediction markets — platforms like Polymarket and Kalshi, where you can trade event contracts on everything from Fed rate decisions to wildfire containment timelines. The industry calls it speculative finance. Critics call it arson betting.My guest today has been thinking about this longer than most. Jamie Pietruska is a historian at Rutgers University whose work traces the long arc of weather gambling — from illegal temperature pools in American cities a century ago to the prediction market dashboards on your phone right now. Her argument is that what looks new is older than we think, and what looks like progress may be a step backward.Dr. Peitruska's Aeon articleSubscribe to Risk Market News
• AI, Models, and the Limits of Climate Assumptions with Sarah Kapnick 13.04.2026 27Min.

  We sit down with Dr. Sarah Kapnick at Climate Tech Connect in Washington, D.C. in a conversation covers the time-horizon problem at the heart of climate finance, what the PG&E bankruptcy revealed about the gap between credit models and physical risk, and where AI-generated climate insight ends and hallucination begins.Subscribe to Risk Market News
• Can Models Still Work When Everything Changes at Once? With Christiane Baumeister 01.04.2026 52Min.

  This week  I speak with Dr. Christiane Baumeister, a professor at the University of Notre Dame. Her research focuses on global oil market dynamics — disentangling the supply and demand forces that drive prices and developing forecasting models that are designed to perform precisely when markets are most volatile.
• (Preview) China's Growing Risk Data Moat and the US Brain Drain With Hui Su 25.03.2026 5Min.

  A conversation with Dr. Hui Su, a professor at the Hong Kong University of Science and Technology and one of the leading researchers working at the intersection of satellite data, artificial intelligence, and extreme weather forecasting. Become a member of Risk Market News for access to the full member episode.
• Confidence as a Service With Eric Winsberg 18.03.2026 1Std.

  We speak with  Eric Winsberg: a philosopher of science at Cambridge and the University of South Florida, who has thought hard about what happens when models move from the lab into the world and into policy and markets.
• (Preview) The $232 Billion Storm No One Is Pricing With Moody's Chris Lafakis 11.03.2026 10Min.

  Chris Lafakis and his team did the first analysis to combine Moody's catastrophe modeling infrastructure with a full macroeconomic model. The results are eye opening.This is a preview of the Risky Science Podcast Member Edtion.To get access to the full episode sign up to become a free member of Risk Market News.
• How Hurricane Risk Really Gets Priced with Dr. Ben Collier 04.03.2026 55Min.

  Dr. Ben Collier, a professor at the University of Wisconsin-Madison, and his fellow researchers published a recent paper that uses twenty years of Florida data to trace a direct line from cat model revisions to the premiums homeowners actually pay. The finding? A one-dollar increase in modeled expected loss translates to roughly five dollars in higher premiums. That multiplier — and what's driving it — is what we're unpacking today.In the episode we dive deep into the findings.The paper: Pricing Climate Risk: Hurricane Models and Home Insurance Over the Last Two DecadesSubscribe to Risk Market News
• Black Box Problems, Machine Judgment and the Rules Nobody's Written Yet With Daniel Schwarcz 18.02.2026 53Min.

  A conversation with Daniel Schwarcz, professor at the University of Minnesota Law School, where he teaches insurance law, contract law, tort law, and financial regulation and his academic work sits at the intersection of AI governance and insurance regulation. (00:00) - Introduction (00:17) - Guest background: From P&C attorney to insurance law professor (02:13) - AI in insurance today: back-office efficiency vs. underwriting and claims (10:06) - Is AI "locked and loaded" for underwriters and claims departments? (12:24) - The 50-state regulatory problem and its compounding complexity (22:05) - Catastrophe modeling and AI in property underwriting (30:19) - Why disclosure usually forestalls regulation rather than protecting consumers (38:40) - Schwarcz's proposed fix for shadow insurance (43:40) - "Obamacare for Homeowners Insurance": the case for insurance exchanges (48:56) - Five-year outlook: where is the insurance industry headed?
• AI Risk, Markets and Modeling the Unknown With Daniel Reti 11.02.2026 42Min.

  In this episode of the Risky Science Podcast we are joined by Danie Retil, co-founder of Exona Labs, a startup building AI risk modeling and quantification tools. 
• Prediction Markets, Parametrics and Rethinking Weather Risk With Dr. Partick Brown 04.02.2026 53Min.

  For decades, insurers, reinsurers and energy companies have relied on models, parametrics, and traditional hedges to manage hurricane and weather exposure. But what if markets could continuously price those risks — in real time — and let anyone transfer or hedge them instantly?In this episode I’m joined by Dr. Patrick Brown, Head of Climate Analytics Interactive Brokers to talk about modeling the models, forecast contracts, and whether prediction markets could become the next tool in the risk-transfer stack for the institutional market.
• Greenland, Venezuela and the New Political Risk Model Reality with WTW’s Sam Wilkin 21.01.2026 46Min.

  In this episode of the podcast, we speak with geopolitical risk expert Samuel Wilkin of Willis Towers Watson about why political risk is moving from a background concern to a front-line business problem. Sam breaks down the rise of “gray zone” attacks in the space between war and peace—from covert sabotage to infrastructure disruption—and explains why these threats are so difficult to model and insure. He also argues that the future of political risk management is less about perfect forecasts and more about scenario discipline, exposure mapping, and governance structures that can keep up with a faster, messier geopolitical cycle.
• Cyber Risk in 2026 and Why Near Misses Matter More Than Losses With Morgan Hervé-Mignucci 14.01.2026 41Min.

  In our first episode of the New Year we are focusing on cyber risk in 2026, a peril that looks increasingly systemic, yet remains poorly understood when it comes to how losses actually materialize.Over the past decade, cyber risk modeling has matured rapidly. But as cloud concentration deepens, dependencies multiply, and “near miss” events become more frequent, a central question remains unresolved: what does a truly systemic insured cyber loss actually look like—and are markets prepared for it?In this conversation with Coalion’s Dr. Morgan Hervé-Mignucci,Head of Risk Modeling at Coalition ,the discussion focuses on how cyber models have evolved, where they still fall short, and why many high-profile disruptions generate far less insured loss than the headlines suggest.
• Climate, Markets and the Limits of Insurability with Dave Jones 24.12.2025 51Min.

  In the last episode of Risky Science, we examined skepticism around climate-conditioned catastrophe models with Roger Pielke Jr.—questioning how much weight long-range climate assumptions should carry in near-term insurance and capital decisions.Today’s discussion is a direct counterpoint.My guest is Dave Jones, former California Insurance Commissioner and now director of the Climate Risk Initiative at UC Berkeley Law. His recent article argues that insurance itself has become the clearest early-warning signal of climate risk—describing property insurance as the “canary in the coal mine,” and warning that the canary is already dying.This conversation is timely because the stress is no longer theoretical. Catastrophe losses are accelerating, insurers are pulling back from high-risk regions, and residual markets are expanding rapidly. Jones argues that neither deregulation nor rate increases will be enough if the underlying drivers of loss continue to intensify.We’ll examine California and Florida as live case studies, what mitigation and modeling can realistically achieve in the near term, and where the practical limits of insurance may already be coming into view.