The Lloyd's Market Association (LMA), in collaboration with Barnett Waddingham, has launched an AI Adoption Toolkit to help managing agents build and strengthen governance frameworks for artificial intelligence (AI) across the Lloyd's market.
The release follows new survey findings indicating that AI and machine learning adoption has shifted from infancy to early-stage deployment across much of the market, and that formal governance frameworks are now becoming the norm rather than the exception.
According to the latest LMA–Barnett Waddingham survey on AI risk management, based on responses representing more than 60% of Lloyd’s market stamp capacity, a clear majority of firms now have, or are developing, a structured AI framework. Many are also mandating human oversight of AI-generated outputs, underlining a focus on augmentation rather than full automation.
The survey pointed to a marked change in the past year, with firms moving beyond isolated pilots towards early production use in functions, such as pricing, exposure management and claims.
Responsibility for AI governance is typically shared between senior technology leaders and dedicated committees, while concerns have shifted towards data privacy, cyber security and third-party risk as adoption scales.
That backdrop has created demand for practical, market-specific guidance. The LMA and Barnett Waddingham have framed the new toolkit as a way to translate high-level principles into processes that can be implemented at syndicate level, particularly in an environment where regulators are stepping up their scrutiny of AI use in financial services.
The AI Adoption Toolkit is presented as principles-based rather than prescriptive. It is intended to be flexible enough to support firms at different stages of their AI journey and to be adjusted to varying organisational structures and regulatory footprints.
The framework is built around five core themes: governance and accountability; risk tiering; data protection, security and intellectual property; training and awareness; and pragmatic adoption.
The launch sits within a broader LMA programme on AI and machine learning in actuarial, risk and exposure management. The association has repeatedly highlighted that members are looking for practical, governance-led support as AI becomes more embedded in day-to-day business, particularly where systems influence reserving, capital, risk selection or client outcomes.
Market participants said the need for this kind of guidance is becoming more acute as firms move beyond proof-of-concept work and start to integrate AI tools into core workflows, including delegated authority arrangements and complex reinsurance structures.
“As AI adoption continues to develop across the Lloyd’s market, the focus is increasingly shifting towards how it is implemented and governed in practice," said Sanjiv Sharma, head of actuarial and exposure management at LMA. “The toolkit is designed to provide practical guidance to help firms embed appropriate controls early, manage risk effectively and scale adoption in a structured way.”
