AI Governance Readiness Assessment
How ready are your AI systems to prove, on any given day, that their controls are working? Answer ten questions and get a readiness score for each of the BMA's five proposed supervisory outcomes, with your priority gaps in order.
The questions follow the Laurenceo Production Control Specification, the ten controls set out in Laurenceo's independent response to the Bermuda Monetary Authority's consultation on the responsible use of AI in financial services. Control depth scales with materiality, so the controls asked depend on your answers below.
1Your AI Profile
2Your Controls See results
For each control, choose the option closest to how it works today across your AI use cases.
How the Score Works
Each control is scored from 0 to 3: not in place, documented and run by hand, partly automated, or evidence by default, where the control runs in the system and writes its own evidence as it operates. A control only counts if it applies at your materiality tier. Outcome scores average the controls mapped to that outcome in the specification. Foundational controls that apply at every tier, and controls made more important by vendor AI, personal information or agents, rank higher among the gaps.
The bands are Early (below 40%), Developing (40 to 69%), Established (70 to 89%) and Evidence by Default (90% and above).
What This Is Not
This is not an official BMA tool, and it is not a compliance determination. The specification is illustrative: the consultation paper allows entities to adopt alternative controls that achieve the same outcomes, and the BMA's guidance was still under consultation when this tool was built. It is general information, not legal, regulatory or financial advice.
Your answers are kept only in this browser and in the page address, so you can bookmark or share a result. Nothing is sent to Laurenceo or anyone else.