Insurance is data-rich and decision-heavy — strong AI territory, but with sharp fairness and conduct obligations. Here’s how UK insurers can adopt AI in 2026. (dgm implements osFoundry as an independent partner; regulatory responsibility stays with the insurer.)
Where AI helps
- claims processing and triage;
- underwriting support;
- document and submission handling;
- fraud detection; and
- customer service.
AI accelerates these — but pricing, underwriting and claims decisions affecting customers carry fairness obligations.
The regulation
Insurers are FCA-regulated (and PRA for prudential matters). The key frameworks:
- Consumer Duty — fair outcomes, no foreseeable harm (especially relevant to pricing and claims);
- SM&CR — accountability;
- UK GDPR and the DUAA’s automated-decision safeguards.
Fairness is the central risk
AI in pricing and claims must avoid unfair or discriminatory outcomes, and insurers must evidence good outcomes. Significant decisions affecting customers should keep meaningful human involvement — both for Consumer Duty and because the DUAA requires safeguards for significant solely-automated decisions about individuals.
Manage it by bias testing, human oversight, documented outcomes, and transparency — fairness and explainability designed in from the start.
Where osFoundry and dgm fit
dgm builds fairness-aware, data-controlled AI on osFoundry: explainability, audit trails to evidence outcomes, human oversight for significant decisions, and bias-awareness. Data control via self-hosting or an EU region (it publishes US/EU/JP regions, not a UK one) and bring-your-own-key.
dgm is an independent integration partner with zero integrations so far — no client claims. Regulatory responsibility stays with the insurer. To scope a fairness-aware insurer AI project, book a consultation with dgm. Not regulatory advice.