Digital platform Avantia has launched a new AI tool designed to better identify fraud and enhance the home insurance claims process.

Built on its existing proprietary decisioning platform, ‘Holmes’ is trained to use a vast amount of data gathered internally through the customer lifecycle and supports handlers by rapidly identifying potential fraudulent activity, as well as flagging straightforward cases for quick and fair resolution.

In initial trials which have taken place in Q2 2025, Holmes exceeded initial expectations, improving fraud detection accuracy and highlighting cases ready for quick settlement. Avantia also found that its new AI tool proved to have much broader benefits than its original use case, making recommendations about claim coverage, payment amounts, data capture and next steps, particularly on highly complex claims. As a result, Avantia has identified significant operational and indemnity cost savings from rolling Holmes out to all claims.

In controlled tests, Holmes not only helped make Avantia’s fraud detection rate six times more effective, it was also able to complete payment calculations with a 98% accuracy and make additional recommendations in 50% of cases which meaningfully improved the outcome of a claim. Holmes is expected to help process live claims from September 2025.

Dan Huddart, CTO at Avantia Group, said: “What we’ve seen from Holmes is proof that AI is a game changer and that forward-thinking insurers should be ready to rip up the rulebook and completely redesign their claims management teams and systems.

“We were early adopters of AI and are now designing our entire operation around it because we can see it gives us and our customers tangible benefits, with improved processing time and consistency in the handling of a claim. I believe this is the most advanced implementation of genAI to date in the home insurance market.”