Gradient AI
Predictive risk models for insurance underwriting and claims
Overview
Gradient AI is a Boston insurance software company, spun out of Milliman in 2018, whose models predict underwriting and claims risk for group health, workers' compensation and P&C insurers. Its products include SAIL and Renewal Analytics for group health and the Total Incurred Prediction model and ClaimVoyant triage for workers' comp claims, with integrations into Guidewire ClaimCenter.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 14 Oct 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
- SAIL for new business
- Renewal Analytics for existing business
- Risk Management Life-Cycle combining both
- Underwriting risk scores and severity predictions
- Claims triage and intervention
- Portfolio stratification
What you can produce with Gradient AI
- Risk scores on group health submissions to price new business (SAIL)
- Twelve-month group risk projections turned into renewal prices, with IBNR adjustments and peer benchmarks
- Early estimates of a workers' comp claim's total incurred cost to support reserving
- Triage flags on claims likely to be costly or slow to close, from first notice of loss (ClaimVoyant)
- AI risk alerts embedded in Guidewire ClaimCenter without a separate portal
- Portfolio views of risk concentrations across a book of business
Gradient AI covers both underwriting and claims, with distinct products for group health (SAIL, Renewal Analytics) and workers' comp (Total Incurred Prediction, ClaimVoyant), delivered inside Guidewire ClaimCenter and other core systems. A named carrier, MEMIC, reports better reserving accuracy, and the company is funded through a 2024 Series C and 2026 growth capital from CIBC.
Its headline results are vendor figures, pricing is not published, and its data and lines are US-focused. Best suited to US carriers, TPAs and self-insured employers prepared for an enterprise sales cycle; ask for references in your own line.
What we found
Gradient AI sells predictive models to insurers for underwriting and claims. It began in 2012 as a practice inside Milliman, the actuarial consultancy, and became an independent company in 2018 when Stan Smith, its founder and chief executive, bought the business out. It is based in Boston and works in group health and stop-loss, workers' compensation and wider property and casualty, selling to carriers, MGAs, TPAs, PEOs and self-insured employers.
The product line is more specific than earlier versions of this page suggested. In group health, SAIL scores new-business submissions, and Renewal Analytics projects each group's next twelve months of risk and turns it into a renewal price; a September 2026 update added IBNR adjustments, peer benchmarking, a group-termination model and client-ready PDF reports. In workers' comp, the Total Incurred Prediction model estimates a claim's final cost early, and ClaimVoyant triages claims at first notice of loss. Gradient's claims tools run inside Guidewire ClaimCenter, with ClaimVoyant due to join them this autumn, and Duck Creek lists Gradient as a partner.
Most of the evidence comes from the vendor. Gradient cites more than 300 enterprise deployments, client retention above 95%, an 80% cut in quote turnaround and lost-time claims shortened by 44 days. The clearest named result is The MEMIC Group, a workers' comp carrier, which reported in a May 2026 joint press release that reserves became more accurate and timely after a six-month study of the prediction model. No outside measurement of these figures was found, and earlier versions of this page wrongly said no customers were named.
On funding, Gradient raised a $56.1 million Series C led by Centana Growth Partners in July 2024, and CIBC Innovation Banking provided growth capital of an undisclosed amount in March 2026. No acquisition has been reported. Pricing is not published; contracts start with a demo, and buyers should ask for references in their own line of business.
Compiled by RECATOOLS from public sources: vendor documentation and pricing, code repositories and published reporting.
About this listing
This entry was compiled from publicly available data including Gradient AI's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with Gradient AI unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to Gradient AI directly →
Spotted something out of date? Suggest an update →
Related reading
Alternatives to Gradient AI
More in Business & Finance
