The Charge
Why Reef exists
AI is arriving in Caribbean banks, ministries, classrooms and clinics faster than the guardrails around it. Reef builds those guardrails. We study how AI systems fail, how they are abused, and how they can be made safe for populations and institutions that rarely sit at the table when the technology is designed.
Reef exists to give the Caribbean a shared, practical safety practice: how to test AI before it ships, how to report and learn from AI incidents, and how to hold systems to a standard the public can trust.
The Name
A reef takes the force of the open ocean so the coast behind it stays whole. Reef is the region's first line of defence against AI harm.
What we work on
Focus areas
Model testing & red-teaming
Probing AI systems for failure and bias before they touch the public.
AI incident reporting
A regional register of AI harms, near-misses, and lessons learned.
Fraud, deepfakes & scams
Defending citizens against AI-enabled impersonation and financial crime.
Safety standards for public-sector AI
Minimum bars any government or bank AI must clear.
Consumer & child protection
Keeping the most exposed people safe as AI reaches everyday life.
Who joins Reef: Risk officers, security researchers, auditors, compliance leads, safety engineers, and anyone who has watched a system fail and wants to stop the next one.
Join Reef
Take your place in the Reef Think Tank
Sign up to join the group, or tell us you want to help lead it. There is no fee, and membership is open across the Caribbean and the diaspora.
