Posts on what we actually think.
Essays on conformal prediction, social inflation, and where generative AI helps and where it doesn't. Written by the team.
When Settlement Data Is Sparse, Triangulation Beats Imputation
Training a predictive model purely on public jury verdicts guarantees catastrophic failure. To set accurate reserves, insurers must stop imputing missing settlement data and start triangulating the true boundaries of the claim.
Negotiating When Your Comp Database Is Trial-Biased
A jury verdict is the definition of an outlier—the artifact of a case where the settlement process failed. Yet claims teams still bring binders of trial outcomes to mediation, anchoring their negotiation strategy to extreme anomalies.
Policy Limits in the Comparable Set
Policy limits artificially cap historical settlement data, creating a massive blind spot for predictive models that fail to track collectability.
Liability Probability Dominates the Number
A plaintiff attorney demands ten million dollars, forcing the defense to fixate on the gross damages ceiling. But the mathematical reality of litigation risk is driven almost entirely by the probability of liability.
Want to talk to an executive?
Press, partners, investors, candidates — the inbox is monitored. Tell us who you are and we'll route it to the right person within two business days.