Thanks for subscribing to the free version of Risk Market News. Consider becoming a paid subscriber for access to all content.
Palantir Wants to Target Insurance and Risk Models As It Goes Public
A new tech unicorn is relying on insurtech for its initial public offering.
Palantir Technologies Inc. issued it S-1 disclosure for its planned $20 billion IPO yesterday and the Peter Thiel-backed data analytics company made sure to focus on its private market prospects, especially in the risk finance and modeling industries.
According to the S-1, the company is working on new products in several industries, including insurance, and added that it “will have a significant impact on our business moving forward.”
Palantir detailed its previous success working with insurers, saying in the filing that it had already inked a joint partnership with Japanese-based insurer SOMPO Holdings that helped the tech startup to fuel its commercial and government business in the Japanese market.
In June Sompo announced that it had invested $500 million with the Silicon Valley startup to help launch a so-called Real Data Platform for Security, Health and Wellbeing platform. Not details are available regarding what the Sompo product actually does.
The filing also highlighted another insurance success story:
A multinational insurer sought to build and apply machine-learning models to surface fraudulent insurance claims, while ensuring that processing was sufficiently transparent, interpretable, and accountable to decision makers and oversight authorities. We helped to configure and implement a number of supporting privacy-enhancing features, including pseudonymization processes to minimize data exposure, rigorous documentation of machine learning model features and parameters, and auditing tools for users and regulators.
In addition to the IPO disclosure, Other tech and finance firms are also mentioning Palantir and risk modeling in the same breath,
Palanitir’s insurance and modeling focus was mentioned in Dan Loeb’s second quarter letter to Third Point investors. Loeb, also oiwner of Third Point Re, said that one a portfolio company had partnered with Palantir “to use AI for further [wildfire] risk mitigation and network efficiency.”
“Forward-Looking” Models Need to Focus On COVID-19 Comorbidities, Population Differences
Understanding of particular population segments “unique vulnerabilities”, and incorporating them into exiting models, is the best way understand the spread and mortality of COVID-19, Swiss Re said in a report released this week.
A better understanding of the heterogenous effects of the disease and how individuals respond will improve our prevention measures; will strengthen re/insurers [modeling] of the progress of the pandemic; will provide stronger foundations to target therapeutic responses; and may hasten progress in finding a vaccine.
According to the report, risks associated with the COVID-19 virus can be split into three types; risk of infection, lethality and population mortality rate. The severity of these risks needs to be understood and properly modeled at the level of particular populations in order to understand the overall heterogeneous spread.
For example, women of working age are more likely to test positive with COVID-19 because the greater number of women employed in the health and care sectors. Mortality is highest in areas with “frail” elderly populations where they are exposed both to visiting staff and those discharged prematurely from “hotzone” hospitals.
Insurers and reinsurers in particular need to model these unique traits to gauge the impact to their risk to their book of business, with a particular focus on the differences between insured and general populations and future mortality and morbidity assumptions, Swiss Re added.
Risk Reads
Cascadia Quake Risk Always In the Background
But for most residents of the region, that threat is mounting far offstage. As life in Cascadia carries on—strong rain, stronger coffee, Birkenstocks aplenty—the COVID-19 pandemic has become public enemy No. 1. And for good reason: It’s hard to focus on an invisible fault line fracturing when a global contagion is actively killing people.
A Massive Earthquake Is Coming to Cascadia—And It Can’t Be Stopped, Atlast Obsura
Modeling the COVID-19 Spread Post Hurricane
Researchers built a predictive model combining evacuation patterns from Hurricane Irma — which forced some 7 million people to flee their homes as it moved up the coast of Florida and toward Georgia in September 2017 — and recent county-level coronavirus transmission rates. They then simulated various scenarios of how the movement of 2.3 million evacuees from four coronavirus hotspots might increase Covid-19 cases in both the origin and destination counties. The goal was to determine how emergency managers might better plan evacuations that would limit the spread of the disease.
How Hurricane Evacuation Can Spread the Coronavirus, Bloomberg
Coming Up Short on COVID-19, Verily Pivots to Insurance
Insurance is something of a departure for Verily. A previous project from the company is Project Baseline, a massive medical study to better understand the human body. Project Baseline is also the banner Verily has used for its COVID-19 screening website, which President Donald Trump incorrectly said was backed by Google in March. Verily has also made a health-tracking smartwatch intended to be used for research studies.
Verily, Google’s health-focused sister company, is getting into insurance, The Verge