AI in London at a glance
We help London banks, insurers, asset managers and advisory firms use Claude, OpenAI, Agentforce and Breeze on document-heavy and client-servicing work. A person reviews every customer-facing output, and data, permissions and UK GDPR obligations are settled before any pilot.
How London firms use AI
London financial firms run on documents. Lloyd's market participants handle submissions, slips and claims correspondence. Asset managers produce due diligence questionnaires, fund documents and investor letters. Wealth managers prepare suitability reports and annual reviews. Claude is good at the first pass on all of these: summarizing a submission, drafting a DDQ answer from approved material, or preparing an adviser's meeting brief from CRM notes and recent correspondence. The professional still owns the judgment and signs off on the output.
The regulatory frame is specific. The Consumer Duty asks firms to put customers' needs first, so any AI that touches client communication needs review steps and records that show the outcome was considered. SM&CR means a named senior manager is accountable for how AI is used in their area. UK GDPR governs the personal data AI reads, including purpose limits and retention. So we start by mapping data and permissions, keep personal data inside approved systems, and design human review into every customer-facing use. Agentforce can then answer routine client service questions from approved content. AI only works on clean data and tight permissions.
AI use cases for London firms
Submission and slip summaries
Claude condenses broker submissions into key terms and open questions for underwriter review.
DDQ and RFP drafting
Draft answers from an approved content library for investor relations teams to edit and approve.
Annual review preparation
Advisers get draft review packs from CRM history, checked before any client sees them.
Client service deflection
Agentforce answers routine servicing questions from approved articles and hands off the rest.
AI platforms that fit in London
Systems the AI works from
Common systems for London firms in this kind of project. Anything else connects through APIs or middleware.
- Salesforce
- Addepar
- Black Diamond
- Seismic
- Snowflake + Fivetran
How a AI engagement runs
We start with an AI readiness assessment covering data sources, lawful basis, permissions and the accountable senior manager for each use case. Pilots begin with internal drafting and summaries, with human review of all customer-facing output. Delivery runs from Dallas and our Sofia practice, which overlaps the UK day.
AI packages and pricing · AI services · Take the AI readiness quiz
Our work in London's leading industries
Client names are anonymized and these projects are not specific to London. All case studies
Industries we serve in London
Banking
AI for banking firms in London.
IndustryInsurance
AI for insurance firms in London.
IndustryAsset Management
AI for asset management firms in London.
IndustryWealth Management
AI for wealth management firms in London.
IndustryFintech
AI for fintech firms in London.
AI in London: questions, answered
How does SM&CR affect our use of AI?
SM&CR doesn't ban AI, but it means someone senior is accountable for how it's used in their area. In practice, each AI use case should have a named owner, a documented purpose, defined data sources and a review process. An AI readiness assessment produces that documentation. We also recommend logging prompts and outputs for customer-facing uses so you can show what happened if a client or the FCA asks. Starting with internal drafting keeps early risk low.
Can we use AI on client data under UK GDPR?
Yes, within the usual principles: a lawful basis, a clear purpose, data minimisation and appropriate security. We keep personal data inside approved enterprise tools with suitable contractual terms, restrict AI features to users who could already see the data, and avoid sending more than a task needs. Your DPO may want a data protection impact assessment for higher-risk uses. We provide the technical detail that assessment needs, including data flows and retention.
Where should a London asset manager start with AI?
Due diligence questionnaires and RFPs are a common first step. Teams answer the same questions repeatedly, and Claude can draft responses from an approved library, flag questions with no approved answer and keep wording consistent. Investor relations staff review and edit everything before it goes out. Meeting prep for distributor and investor calls is another good early use, since it saves time without sending anything directly to clients. Both pilots need clean, current content to draw from.
More for London firms
AI in nearby markets



Sources
- citypopulation.de - Greater London (ONS mid-year estimate, 30 June 2024)
- TheCityUK - Enabling growth across the UK 2025
- TheCityUK - The UK's £10trn asset management sector bolsters global position with strong growth
- Lloyd's - About Lloyd's
- FCA - Consumer Duty
- FCA - Senior Managers and Certification Regime
- FCA - Regulatory Sandbox
- HubSpot Knowledge Base - Cloud infrastructure and data hosting FAQ
