Nine in ten Lloyd's of London insurers draw up internal AI governance frameworks as LMA survey reveals sector-wide policy development, coinciding with FCA action
The Lloyd's Market Association (LMA) — the trade body representing managing agents and syndicates operating in the Lloyd's of London insurance market — has reported that nine out of ten Lloyd's insurers are actively developing their own internal governance rules for artificial intelligence. The finding was published on 16 April 2026 and coincided with action by the Financial Conduct Authority (FCA), the UK's financial services regulator, though the full details of the FCA's concurrent step are behind a paywall. The LMA data signals that Lloyd's market participants are not waiting for mandatory regulation before establishing AI governance structures — they are building frameworks proactively, reflecting both commercial pressure (underwriters using AI-assisted risk modelling face liability questions if algorithmic outputs are wrong) and anticipation of incoming regulatory requirements. The FCA has signalled its intention to apply existing Consumer Duty, Senior Managers and Certification Regime (SM&CR), and systems and controls rules to AI-assisted decision-making in regulated firms, rather than building a bespoke AI statute. For law firms advising financial services clients, this creates immediate demand for AI governance policy drafting, technology licence review for AI vendor contracts, and data processing agreement markup where AI tools process personal or commercially sensitive data. The Lloyd's market is a particularly high-stakes environment because AI is being applied to pricing, claims assessment, and catastrophe modelling — areas where a governance failure could produce both FCA enforcement action and large-scale policy disputes.
Why this matters
The LMA survey finding is a leading indicator of a wave of AI governance mandate work hitting financial services legal practices. When 90% of a major market segment is simultaneously drafting AI frameworks, the demand for standardised contractual templates, regulatory impact assessments, and vendor due diligence advice is substantial and near-term. The FCA's concurrent action — using existing regulatory powers rather than waiting for bespoke AI legislation — is the enforcement backdrop that makes this urgent: a firm with a poorly designed AI governance framework that produces a discriminatory pricing outcome or a claims handling failure faces SM&CR personal accountability exposure for senior managers, not just corporate fines. The 'why now' trigger is the combination of AI capability deployment accelerating faster than internal oversight processes and the FCA's explicit supervisory focus on algorithmic systems in regulated activities.
On the Ground
A trainee on an AI governance project for a Lloyd's syndicate would assist with AI governance policy drafting, mark up data processing agreements (DPAs) for AI vendor contracts under UK GDPR, and prepare vendor due diligence questionnaires assessing whether an AI tool's training data and model outputs comply with FCA expectations on explainability and fairness. They would also draft regulatory impact assessment memos mapping the client's AI use cases against FCA guidance.
Interview prep
Soundbite
Lloyd's syndicates drafting AI governance rules now are insuring against FCA SM&CR personal liability before enforcement arrives.
Question you might get
“How does the FCA's Senior Managers and Certification Regime apply to AI-assisted decision-making in a Lloyd's syndicate, and what steps should a Compliance Oversight Senior Manager take to discharge their responsibility for an AI-driven claims assessment tool?”
Full answer
The Lloyd's Market Association has reported that nine in ten Lloyd's insurers are building internal AI governance frameworks, timed alongside FCA regulatory action. This matters because Lloyd's is one of the most AI-exposed sectors in financial services — underwriting, pricing, and claims assessment are all being augmented by algorithmic tools, and every one of those use cases touches an FCA-regulated activity. The legal work is immediate: firms need vendor contract reviews, data processing agreement markups, and governance policy documentation before the FCA's supervisory focus translates into enforcement. The broader trend is the FCA's approach of applying existing SM&CR and Consumer Duty rules to AI outputs, which means senior managers can face personal accountability for model failures — a powerful compliance driver. Firms that build defensible governance frameworks now will be significantly better positioned when the FCA's AI supervision programme intensifies.
My notes
saved