Wanich Risk Lab is building the frontier risk engine.
Insurance is one of the largest industries in the world, and at its core it does one thing: it puts a price on the probability that something goes wrong. Everything else the industry does, the distribution, the paperwork, the claims machinery, exists to service a number that a small group of people set once a year using tools built before the internet. The number is almost always wrong in a specific direction, and the entire industry is organized around absorbing the cost of being wrong instead of getting closer to being right.
We think the frontier of insurance is not a better interface or a faster quote, it is a fundamentally better model of risk. A model that reads every signal the world now produces about a person, a business, a building, or a fleet, and turns that signal into a price that updates as the world does. A model that learns from every risk it writes and every claim that settles, until the gap between what it charges and what it pays is the tightest gap in the market.
We are building that model. An engine that understands risk better than any institution alive today, and becomes the way risk is priced in the century ahead.
If this excites you, contact us.
Latest from the lab
All articlesBy Mark Wasuwanich · October 15, 2026
We ran the same audit categories repeatedly on two independently built AI agents. One category moved by 18.5 to 25 points on a 1-100 scale across runs, well past the tolerance we set, while several others held stable. Here is what that means for evaluation evidence in AI-agent underwriting submissions.
By Mark Wasuwanich · October 14, 2026
Pure premium is supposed to equal frequency times severity. We measured how often multi-task pricing models violate that identity across nine model families on two public motor datasets, and found a simple fix that removes the gap at no cost to accuracy.
By Mark Wasuwanich · September 5, 2026
A read through what people are actually building in and around insurance pricing right now, and what it points to for the model that eventually replaces the annual number.
Wanich Risk Lab, 2026.