AI in Insurance Broking: Managing Risks and Costs
A recent Moterra briefing highlights that while most UK insurance brokers have not formalised an AI strategy, they are already relying on public AI tools for routine tasks such as renewal summaries, market submissions, policy wording comparisons and claims updates. The reliance is driven by workload pressures rather than a deliberate compliance plan, creating a gap between perceived and actual operations. A Bank of England and FCA survey released on 26 October 2026 shows that 86 % of UK financial services firms now use AI, up from 75 % in 2024, underscoring the speed of adoption across the sector.
Regulators are watching this trend closely. The FCA has indicated that existing supervisory frameworks will cover AI use, meaning brokers that place business with EU insurers must also satisfy DORA requirements. Public AI tools fall short of these standards because client data leaves the firm, processing locations are unknown and there is no audit trail of queries or reviews. Banning the tools outright is ineffective; staff simply shift to personal devices, further eroding visibility.
Moterra proposes a different approach: deploying Anthropic’s Claude model inside the broker’s own AWS or Azure cloud environment, encrypted with the firm’s keys and excluded from model training. This architecture keeps client data within a UK-region cloud that already meets resilience and third-party oversight standards. Vytenis Sakalas, CEO and co-founder of Moterra, notes that brokers need AI they can stand behind, enabling a single-sentence response to regulators: the AI runs inside the firm’s cloud under its controls, with a named person signing off on client-facing output.
Cost control is another advantage. Traditional licences charge per seat regardless of usage, quickly outweighing benefits as demand spreads across the office. Moterra’s usage-based pricing eliminates per-seat fees, allowing firms to set team spending limits and monitor adoption, savings and expenses on a unified dashboard. Deployments typically go live within one to two weeks, letting brokers shift from informal to controlled AI use before the next renewal peak. This reduces compliance risk and provides finance and boards with clear spend approval and outcomes.
Why it matters
Regulators expect transparent AI use, and a secure, firm-hosted solution lets brokers meet compliance while controlling costs.
FAQ
What percentage of UK financial services firms use AI according to the latest FCA survey?
86 % of UK financial services firms use AI, up from 75 % in 2024.
Why do public AI tools fail regulatory requirements for brokers?
Client information leaves the firm, processing locations are unknown, and there is no record of what was asked or who checked the answer.
How does Moterra’s solution keep client data secure?
It deploys Claude inside the broker’s own AWS or Azure account, encrypted with the firm’s keys, and never uses the data to train the model.
What is the typical deployment timeline for Moterra’s AI solution?
A standard deployment is typically live in one to two weeks.
Source: Insurance Business UK. See the original source for full details.
Scan with your phone
Scan with your phone