How FOIL and Reassured built an industry-first AI solution to tackle non-disclosure risk
Non-disclosure is one of the most persistent and costly challenges in the UK protection industry. When medical information is incomplete or inconsistent at the point of application — even in small ways — the consequences at claim stage can be significant for customers and insurers alike.
Reassured, the UK's largest life insurance broker, identified this as a problem worth solving properly, with a production-ready AI system built to operate at scale across their business.
That system is SAINT.AI — and it was built with FOIL.
The Challenge
With around 70% of Reassured's customers new to protection, the medical disclosure process is often unfamiliar territory. Customers navigating complex questions under pressure can inadvertently provide incomplete or inconsistent information, creating risk that only surfaces later — at exactly the moment it matters most.
The challenge was building something that could detect inconsistencies reliably, at volume, using the messy, unstructured data that real customer interactions produce.
What We Built
SAINT.AI is a bespoke generative AI model that analyses unstructured data from call transcripts and insurer application records, identifying inconsistencies and flagging them for human review.
The system doesn't replace the judgement of Reassured's teams — it augments it. By surfacing potential discrepancies before they become claim challenges, SAINT gives Reassured's operations an additional layer of accuracy assurance across every policy they write.
Developed over six months and supported by Microsoft, the solution has been built, tested, and refined in close collaboration with Reassured's teams. It is now being implemented across their operations.
Mish Naik, Head of Strategy at FOIL, explains the technical challenge the project required us to solve:
The Impact
For Reassured, SAINT.AI directly supports their commitment to customer outcomes — particularly for the 30% of their customer base drawn from groups least likely to have protection cover.
Laura McGraw, Co-Founder and Chief Customer Officer at Reassured, put it plainly:
The industry has taken notice. SAINT.AI has been recognised as a finalist in the Generative AI Initiative of the Year category at the British Data Awards.
Mike Pritchard, Programme Manager at L&G, also acknowledged the significance of the approach: tools that help surface potential inconsistencies earlier can support better decision-making and greater confidence for customers — provided they complement, rather than replace, responsible human engagement.
Why This Matters
SAINT.AI is a precise example of what we mean when we talk about AI that earns its place in production. It addresses a specific, high-stakes business problem. It works with the data organisations already have. And it keeps human oversight better supported.
FOIL partnered with Reassured to design, build, and deploy SAINT.AI in collaboration with Microsoft. To discuss how FOIL can help your organisation move from AI experimentation to production, get in touch.