Paul VanderMarck

Recently, I talked with George Abbott from The Insurer on #AI risks and #insurance, which resulted in below linked article. Insuring (Gen)AI errors, be it via innovative or traditional insurance products, has the potential of aggregation risk. This is in particular true if we look at (Gen)AI use cases, which are build on foundation models. Foundation models introduce common elements between GenAI and AI use cases, which make their error rates highly positively correlated, depending on the similarity between use cases. Any event, which induces a data drift with error spikes on the foundation model (even a foundation model update), could lead to insured losses across many use cases and insureds. Such risk needs to be considered and managed by insurers. On the positive side, one can reliably quantify the error frequency or hallucination rate of black box AI and GenAI models for many use cases - as my team at Munich Re demonstrated in a research paper we published together with Agni Orfanoudaki from University of Oxford. This insight allows to build a strong quantitative basis to underwrite and price AI error risks - and other AI risks related to AI errors. This in turn empowers insurers to support their corporate clients on their AI adoption journey and helps to bring reliable AI systems / AI agents to the market.

31

1 Comment

The article should have been titled "Why actuarial expertise is now low key for insurers". If you are an auto MGA or carrier in California and you had combined ratio that averaged under 96% the last 3 year then you get it. Others who don't have a combined ratio under 100 appear to grasping at needles in a haystack. Why? One reason is actuarial expertise appears to be grounded in flawed principles. Historically we've calculated premiums based on data assumptions. If a man has drivers license for 20 years he must be a better driver than man with license for 2 years. That could be correct. But it could also cause you to run in the red. Every company know that a "me too" filing doesn't guarantee "me too" profitability. Having a better understanding of the risk and pricing it appropriately can. Actuarial expertise incorporates a deep understanding of our industry. It should also be grounded in a deep understanding of consumer you are serving. The integration of data science will hopefully can take insurers from insurance dinosaurs. The article doesn't really explain how. They would like you to spend money with the article's sponsor to get any real insight. How can change your destiny from a future similar to dinosaurs of the past? Learn how to improve every touchpoint with AI. Learn how to increase employee engagement with AI. How do we start? Examine workflows. Examine high frequency communication with policy holders. Remember everything is a data point, but not all data provides insight. Actuarial expertise is framed in 20th century experience. The chart below was taken from a study done last year by Statista. We have way too much to process each day. Improve your team's ability to impact your organization you will improve your ratios. Improve your client experience you will improve your ratios. Despite some opinions, you can't control severity. You can however use AI to dynamically improve your modeling without the need to constant rate filings.

1

I think this The Wall Street Journal article on #cyber #insurance nicely vindicates the benefits of using Axio 360 scenario-based cyber risk engineering including interesting insights from Kimberly Denbow, VP Security & Operations American Gas Association citing many insurers have become more selective in the clients that they take on where the number of insurance providers willing to write cyber insurance policies has been limited - Terms can differ so wildly between insurers that it is often hard for operators to understand what a policy actually cover - but it doesn't need to be this way: Axio brings greater transparency on risk profiling, if you're 'top right', no need to get lumped into 'bottom left' of the 2x2 box and no need to get stuck with cyber balance sheet protection. We help you: ❶ Determine scenarios ❷ Assess Cyber Risk Posture ❸ Quantify Cyber loss expectancy & economic loss reduction ❹ Stress test the Cyber Insurance policy for efficacy - will it perform as expected?🧐 ❺ Access the most efficient balance sheet protection capital www.axio.com #gas #naturalgas #cyberriskengineering #cyberriskquantification https://lnkd.in/eRQwzQg8

14

Our landing page for Verisk NGM [Next Generation Models] features product materials, published papers & articles on industry best practices, which define our view of natural catastrophe risk. https://lnkd.in/egMmF5TF In the interest of knowledge sharing & peer review, we have done the due diligence to have industry associations & journals take record of innovation by Verisk Extreme Event Solutions. The selected items tell the story how at times of excess #insurance & #reinsurance hazard & #claims #volatility we bring to market #statistical and #actuarial methodologies to equip our clients with the right approaches for the tasks at hand.

6

To become an underwriting trailblazer, P&C insurers must adopt AI. In an interview with @ReinsuranceNews, my colleague @Adam Denninger shared how AI adoption in underwriting will be incremental, serving as a valuable tool for outperforming mainstream carriers. https://lnkd.in/dRtAHGjX

31

Growing attention to the intersection of insurance and climate risk. This time it hits home. https://lnkd.in/e28wkimR Nice to hear more attention to these esoteric metrics like “combined ratio” and a little known ratings agency called“AM Best “. I’m expecting some key insights from InnSure , Charlie Sidoti , Christopher Lowell, CPCU, MBA. Also for people who want much more on this topic check out https://10across.com/ and their in-depth analysis on impacts of climate change on insurance markets across the country. #insurance #climatechange #infrastructure #homeowners

27

4 Comments

I enjoyed interviewing Adam Rimmer of FloodFlash to hear his thoughts on parametric insurance! Check out his interview for Insurance Thought Leadership to learn what he thinks about the future of parametric, how it is filling gaps left by traditional coverage, and the surprises between the UK and US markets. Thanks to Adam for sharing his thoughts! #abileads #insurance #risk #riskmanagement #parametric #cpcu #parametricinsurance https://lnkd.in/evVUEh_P

14

How can AI help underwriters? By increasing speed and effectiveness of insights -- Jennifer Hoolehan Kyung, VP Underwriting at USAA previews her thoughts from the upcoming webinar Pioneering Ethical AI: P&C Insurers to Play a Critical Role. Full show on September 12, 1p EDT. Jen is joined by Matt McHatten, Iris Devriese and myself, moderated by Peter Miller. If you are an underwriter or an insurance professional looking ahead, join us. Link to registration site in comments. SAS Munich Re The Institutes CPCU Society. #AI #insurancetechnology #analytics

16

2 Comments

Outstanding article by our friends and Alan Demers, CPCU, AIC and Stephen Applebaum, "Can Climate Tech Save Insurance" and the paradigm shift from "repair and replace" to predict and prevent. In this issue of Connected, Climate Tech is defined as technology enabled tools and solutions that can avoid, limit, mitigate and remediate the impact of extreme weather events on policyholders and their property, whether man-made or natural. This includes emergency response and triage services. I will also say, Connected is my favorite newsletter! It is a 3 minute scan of our industry, filled with well sourced and very insightful content. Give them a read! https://lnkd.in/ed2WjrdW

25

8 Comments

Berkshire Hathaway's reinsurance businesses reported a net underwriting result of $2.3 billion for Q2 2024. Berkshire Hathaway has achieved an 82% year-on-year rise in net underwriting earnings across its insurance and reinsurance businesses for the second quarter of 2024, as the firm’s P&C reinsurance operation again generated more than $1 billion in underwriting profit. #Reinsurance #Insurance #BerkshireHathaway #BusinessSuccess #InsuranceIndustry

34

This is a good report with nice visual presentation of data from our colleagues at Gallagher Re. Putting the insured and economic loss together gives a strong perspective of the #insurance coverage gap. For readers outside of the industry & for #publicsector & political leaders it is a good data point of motivation in the continuous effort to reduce the gap. https://lnkd.in/eiMgu-ng

1

Our updated US #Wildfire model comes at a needful time. The market at large is looking for social awareness of the true & immediate hazard for property owners. #insurers & #reinsurers demand a stable and coherent view of the fair technical price of #catastrophe risk. To meet demand our updated model features . enhancements to the fire spread model, especially in the wildland-urban interface . insurance loss modeling of smoke damage outside of the wildfire perimeter, basic principles, https://lnkd.in/ePkR9B8j . the incorporation of data from recent fire seasons, including fuel information supplemented with insights from Fireline® . accounting for coverage and spatial dependencies in all accumulations, see https://lnkd.in/eBvWg7QM https://lnkd.in/eVwzckVj #californiainsurance, #extremevents, #riskmanagement, #actuarialscience

21

Do you know the 3 components required for a #wildfire? Zurich North America's new overview on wildfires is a helpful primer on a rising risk for businesses. https://lnkd.in/gvJX53Ya

1

What's the real game changer for InsureTech firms..,....? InsureTech is evolving rapidly, but not all innovations are game-changers. In two decades in the Insurance industry, I've seen countless technologies come and go. Many promise transformation, but few truly deliver. So, what’s the real game-changer in InsureTech today? Let’s break it down: 1. API Integrations = Seamless Connectivity - APIs connect systems, making data flow effortlessly. - Think of it as the backbone of modern InsureTech. - It’s not just a tool; it’s a necessity for digital transformation. 2. Machine Learning = Smarter Underwriting - ML analyzes vast data to predict risks accurately. - It’s transforming how we assess and price insurance. - The result? More accurate, fair, and efficient underwriting. 3. Digital Platforms = Customer-Centric Experience - Customers demand digital-first interactions. - Platforms offer 24/7 access, transparency, and ease. - It’s not just about selling policies; it’s about enhancing the entire customer journey. Big Picture: Focus on these three - APIs, ML, Digital Platforms. They’re the real drivers of change in InsureTech today. Your Thoughts? What’s your take on these InsureTech trends? Anything you’d add or see differently

22

At this year's National Bureau of Economic Research meeting of the #insurance work group the importance of price-making in distressed markets was the big topic. While building Next Generation Models [NGM] at Verisk we wrote two articles (with Society of Actuaries, Casualty Actuarial Society review) for colleagues in #underwriting & #riskmanagement. The papers outline the value of our view of natural #catastrophe risk for metrics in spatial #diversification, https://lnkd.in/e2ebr5aG & for competitive #premium pricing, https://lnkd.in/e66HebMu. These have brought about some excellent discussion on the #insurancecoverage gap & a suggestion to look at the theoretical volume of #reinsurance coverage needed for distressed markets to function properly. If interested to contribute with data or research, please reach out, #floridainsurace, #californiainsurance

1

5 ways AI will transform insurance underwriting: (and who will win the race) AI is infiltrating insurance, promising to slash risk assessment from hours to minutes. But the tech isn't foolproof yet. Insurers must tread carefully to avoid liability minefields. In this week's LION Specialty newsletter, we predict the 5 main ways insurers will deploy AI in underwriting: 1. Learning risk appetites and underwriting protocols 2. Aggregating data more efficiently 3. Generating assessment memos faster than humans 4.Expediting risk decisions 5.Turbocharging speed and efficiency, especially for larger deals We also forecast who will lead the AI underwriting race (hint: it might not be the mammoths). But that's not all. The latest edition also covers: • Berkshire's Ajit Jain warning insurers aren't making money on cyber • Private equity's low-risk lending push (and the underwriting ripples) • RenRe's view on the improving Florida reinsurance market Like this post and comment "newsletter" or DM us to get the next edition delivered straight to your inbox. Follow Mark Flippen, Natasha I. Kiemnec, ARM and LION Specialty for more exclusive content

21

22 Comments

Well worth a listen this one, partiuclarly if you're a CISO or Insurance Risk Manager or Client Manager Lot's of themes here from Axio CEO Scott Kannry including ESG and the real life example concerns from a CISO at an Energy company around threat actors infiltrating operations to cause an explosion/vapour cloud and resultant impact to the local town culminating in a $6-8Bn loss potential, none of which was insured 😫 And one my favourites, the fire comparison - if you don't think you'll have a fire, why still bother with fire risk-reduction engineering; #cyber is no different Axio helps client imagine the unimaginable; rather than build the [security] wall as high as possible, being prepared for what happens when someone gets in and enacting the Business Continuity Plan is essential to protect your business, just the same as any other ERM risk #cyberriskmanagement #cyberriskquantification #cyberriskengineering #cyberinsurance

6

With the impact of climate change, the withdrawal of insurers from particular markets will only increase. This is a great article highlighting the issues.

5