If your insurer doesn't know about your AI, your claim could be cut.

Business insureds must make a fair presentation of the risk under the Insurance Act 2015, disclosing every material circumstance they know or ought to know. How a business uses AI can be material. If the presentation is unfair, the insurer's remedy depends on what it would have done had it known, from charging more to refusing the claim.

Under the Insurance Act 2015 a breach of warranty suspends cover until it is remedied, rather than ending it. And under section 11, if a term is designed to reduce a particular kind of loss, the insurer cannot rely on its breach if you show the breach could not have increased the risk of the loss that happened. Evidence of your controls decides both.

Yes, for AI used for risk assessment and pricing of natural persons in life and health insurance (Annex III, point 5(c)). Those systems must meet the high-risk requirements, including logging and human oversight.

EIOPA's Opinion on AI governance and risk management (August 2025) explains how existing insurance law applies to AI not classed as high-risk: data governance, record-keeping, fairness, cyber security, explainability and human oversight, applied in proportion to the risk.

It depends on your disclosure, your warranties and the policy wording. Many policies are silent about AI. A sealed record of what you disclosed and what your AI did is the strongest position to argue from.

It seals every AI decision so you can prove what happened, issues an Insurability Certificate underwriters can check, and logs incidents in a claim-ready file the moment they happen.