Category: AI in insurance · Reviewed by Chrissie Anderson, Client Executive · Last reviewed 2026-06-10
ChatGPT is OpenAI’s consumer-facing large language model interface (released November 2022) and the wider set of GPT-based products available via the OpenAI API and Microsoft Azure OpenAI Service. In United Kingdom insurance, ChatGPT and its peers have been adopted across broker, underwriter, claims and customer-service workflows, subject to the FCA Consumer Duty, the ICO’s guidance on generative AI and the supervisory expectations set out in DP5/22 and FS2/23.
This entry catalogues the practical use cases for ChatGPT-style generative AI in UK insurance. They fall into three broad categories:
broker-side internal use (drafting, summarisation, research, code);
insurer-side internal use (underwriting copilots, claims summarisation, complaints drafting); and
customer-facing use (chatbots, voice agents, intent classification).
Legal / Regulatory basis
FCA Consumer Duty (PS22/9) — particularly relevant to customer-facing use; consumer-understanding and avoidance of foreseeable harm.
FCA & PRA DP5/22 and FS2/23 — supervisory expectations on AI governance.
ICO generative AI consultation responses (2024) — emphasising lawful basis, accuracy, transparency and individual rights.
UK GDPR — particularly Article 5 (accuracy), Article 28 (processor terms with the LLM provider) and Articles 44–49 (international transfers).
FCA Senior Managers and Certification Regime (SMCR) — designated SMF accountability for material AI use; commonly mapped to SMF-3 or SMF-24 depending on the firm.
FCA Financial Crime Guide — where ChatGPT-based tools touch sanctions or AML workflows.
FCA SYSC 4 and SYSC 7 — overarching governance and risk control.
How it works in practice (use-case catalogue)
Broker-side internal
Proposal-form and presentation drafting — generating first drafts of submission letters from structured client data, reviewed by a broker before issue.
Policy summary drafting — producing layperson-friendly summaries of policy wording for client meetings (subject to verification against the wording).
Sanctions and AML narrative drafting — generating internal narratives for compliance review (not a substitute for sanctions screening systems).
Market intelligence — internal Q&A over the firm’s research and broker notes via RAG.
Operational copilots — coding, spreadsheet logic and process drafting for operations teams.
Insurer-side internal
Claims file summarisation for handler hand-off and FOS preparation.
Underwriting copilots for retrieval over guidelines and prior placements.
Complaints drafting — first-draft FOS responses for handler review.
Regulatory horizon-scanning — summarising Handbook changes and supervisory publications.
Customer-facing
Self-service chatbots for FAQ, policy-status enquiries and renewals; with clear human-handoff for vulnerable customers and consequential decisions.
Quote journeys with conversational interfaces (with structured fall-back when out of scope).
Claims FNOL chat / voice — intent capture, guidance and photo upload, integrated with Computer vision claims.
Common variations / Subsequent developments
Microsoft Copilot integration in M365 tenants used by UK insurers and brokers — generating governance questions around data residency, retention and audit.
Azure OpenAI Service in the UK region as the preferred enterprise endpoint for many UK firms.
Open-source alternatives (Llama, Mistral, Mixtral) self-hosted for sensitive workloads.
Specialist insurance copilots (vendor and in-house) layered over ChatGPT or Claude APIs.
Voice agents integrating ChatGPT-style models with real-time speech for customer service.
Example
A UK Lloyd’s broker authorises internal use of ChatGPT Enterprise (via Microsoft Azure OpenAI Service, UK region) for drafting submission letters and internal research. The firm prohibits client-data input outside the enterprise endpoint, prohibits use for any consequential customer decision without human review, and documents the use case in its AI register. The Compliance Oversight SMF-16 holder owns the policy; the firm’s Consumer Duty board report records that no customer-facing automated responses are issued without sign-off, and reviews exceptions quarterly.
FCA & PRA, DP5/22 / FS2/23 — AI and Machine Learning, October 2022 / 2023.
ICO, Generative AI consultation responses, 2024. https://ico.org.uk
FCA, PS22/9 — Consumer Duty, July 2022.
FCA Handbook, SYSC 4, SYSC 7, PRIN 2A, FCG.
UK GDPR; Data Protection Act 2018, https://www.legislation.gov.uk
HM Government, A pro-innovation approach to AI regulation, March 2023 White Paper, and February 2024 Response.
Microsoft, Azure OpenAI Service — Data, privacy and security documentation, https://learn.microsoft.com (provider documentation referenced by enterprise governance teams).
This entry is part of the Apex Insurance Wiki. Last reviewed by Matt Bartlett on 2026-06-10. Next review: 2026-12-10.
Apex Insurance Brokers Limited. Authorised and regulated by the Financial Conduct Authority, FRN 724952. Registered in England and Wales, Companies House 07014570. This entry provides general information about UK insurance concepts and is not regulated advice. Consult your insurance broker on your specific position.
Our service promise. We acknowledge every quote request the same working day. For straightforward risks, indicative terms typically follow within five working days. Complex risks — higher-risk buildings, cladding, mid-term proposals requiring fresh underwriting — may take longer; we’ll send you a progress note by the end of the fifth working day in those cases.