.png){kind=logo}
Soteris, a Y Combinator-backed machine learning company, launched an AI-powered product to help property and casualty insurers and managing general agents identify unprofitable policies already in their books. The company said the tool can flag money-losing policies without changing rates, policy forms, regulatory filings, or staffing levels.
The launch comes after a strong year for the US property and casualty insurance sector. The industry recorded a USD 60.9 billion net underwriting gain in 2025, nearly three times the USD 22.1 billion recorded in 2024, according to AM Best’s latest financial review. The industry’s combined ratio improved 3.7 points to 92.9.
Separately, Verisk and the American Property Casualty Insurance Association estimated the 2025 underwriting gain at about USD 63 billion. They also reported that the combined ratio improved to 92.9% from 96.6% in 2024. Fitch Ratings described the performance as the industry’s strongest in two decades, with the combined ratio reaching its best level since 2006.
Soteris’ original loss-ratio product has been used by carriers and MGAs since 2020. The company said it has scored more than 100 million policy submissions covering over USD 180 billion in premiums.
Also Read: China Probes DeepSeek, Moonshot AI Over Claude Data Routing Claims
Soteris said its new product addresses the gap between predicting a policy’s loss ratio and determining how much profit it actually generates. The economics of one policy can be divided among several entities, including companies handling sales and servicing, state licensing and capital backing.
“Every insurer knows they’re writing policies that will lose them money. They can’t find those policies with the resources currently at their disposal,” Soteris founder and CEO Sunit Shah said. He described this as the ‘blind spot’ the company built Soteris to address.
Insurers commonly use spreadsheets and pivot tables to assess policy performance. Soteris said its approach can generate millions to billions of segment combinations by examining multiple policy characteristics simultaneously and treating each policy as its own credible segment.
Implementation takes under 90 days, the company said. Once active, the system delivers insights in under 250 milliseconds through an API at any stage of a policy’s lifecycle, including quote or binding.
Soteris said users of its original loss-ratio product recorded improvements of five to 15 percentage points within a year. In proof-of-concept work for the new product, the company observed book EBITDA increases of 70% to 125% among participating insurers.
Soteris has raised more than USD 8 million in seed funding from Spider Capital, Khosla Ventures, Intact Ventures, Amplify Partners, Foundation Capital and the Webb Investment Network. The company was part of Y Combinator’s 2019 cohort.
Founder Sunit Shah previously built pricing models for life insurance and spent two years building a USD 750 million P&C insurer within hedge fund Pine River Capital, according to his Y Combinator profile.
