The Caribbean's First AI Summit Just Wrapped in Trinidad. Your Insurer Didn't Wait for It.

What Just Happened in Trinidad

CAIF 2026 was organized by the Caribbean Telecommunications Union in partnership with UWI St Augustine and the university's Artificial Intelligence Innovation Centre. CTU Deputy Secretary-General Nigel Cassimire framed the stakes at the opening session: AI, he said, must be harnessed to reduce the region's vulnerabilities and amplify its strengths, which requires standards fitted to the region's scale, governance that protects citizens, and financing that lets pilot projects survive past their first year. Dr Craig Ramlal, who chairs the Caribbean AI Task Force and directs the AI Innovation Centre, put it more starkly: the Caribbean cannot afford to sit out the AI transition happening around it.

The CAITF report itself supplied the numbers behind that urgency. Latin America and the Caribbean account for 6.6% of global GDP but attract only 1.12% of global AI investment. A PwC survey of 154 company directors, published in March 2026, found that only 6% believe their boards spend enough time understanding AI's impact on the business, and only 9% believe their boards receive enough information to manage the risks that come with it. Financial institutions across Jamaica, Trinidad and Tobago, and Barbados are already using AI for fraud detection, credit scoring, and customer service automation, the report noted, while acknowledging that oversight of those systems ranges from thin to nonexistent.

AI Already Prices and Settles Policies

None of this is a warning about a future risk. Underwriting is the process an insurer uses to decide whether to accept an application and what to charge for it, and across the region that process increasingly runs through a model rather than solely through a human underwriter's judgement. A model trained on historical claims data can price a policy in seconds where a manual review once took days. It can also encode whatever bias sat inside the historical data it learned from, quietly and without anyone deciding to put it there.

The same shift is under way on the claims side. Automated triage tools sort incoming claims by likely cost and complexity, routing straightforward ones toward fast settlement and flagging others for a closer look, a look that itself is often assisted by a fraud-detection model rather than performed by a human alone. Two Caribbean examples, one from Jamaica's largest domestic financial group and one from the region's largest catastrophe risk pool, show how far that shift has already gone.

Sagicor's Prize-Winning Claims Tool

On April 21, 2026, at the AC Marriott Hotel in Kingston, eleven finalist teams from the University of Technology, Jamaica pitched products at the 2026 Sagicor Innovation Challenge. Team Connective, made up of Keira Uter and Jocade Hudson, took the J$500,000 grand prize for Data Bridge, a platform built to extract, organize, and transfer information across systems without manual data entry, cutting the document-processing time that slows down large financial institutions. Team SurPath, Rianna Henningham and Jemar Plummer, took second place and J$250,000 for a climate-risk intelligence platform mapping hazard exposure across Jamaica's fourteen parishes. Team Nexus, Denisha Barker and Gabriele Russell, took third and J$100,000 for an AI-powered tool that automates insurance claims processing.

Sagicor has put more than J$125 million into its Innovation Lab since 2019, roughly J$25 million a year, according to reporting on the challenge in the Jamaica Observer and the Jamaica Gleaner. That is a training pipeline for exactly the kind of tool the group says it wants. Sagicor Group Jamaica's president and CEO, Christopher Zacca, said in 2025 that the group is moving to digital automated underwriting across all of its operations, a five-year plan now in its third year under chief technology officer Joanna Banks, with the later years built around customer-facing releases rather than internal plumbing.

A student prize is not the same as a deployed production system, and it would be wrong to claim Data Bridge or the Nexus claims tool are already running Sagicor's live book. But a major Caribbean insurer funding, judging, and rewarding an AI claims-automation tool built by its own market's students is a clear signal of direction, not a hypothetical one.

CCRIF's Record $1.57 Billion Pool

The parametric side of Caribbean insurance tells a parallel story. CCRIF SPC, the Caribbean Catastrophe Risk Insurance Facility, confirmed on June 5, 2026, that its coverage limits had grown 9% to US$1.57 billion for the current policy year, up from US$1.44 billion, with a 17% increase across its Caribbean portfolio and 18% across its Central American portfolio. CEO Isaac Anthony welcomed new members including the Jamaica Public Service Company and the Nevis Electricity Company Limited to its electric utilities book, and noted that Jamaica had purchased the COAST product, CCRIF's parametric cover for the fishery sector, for the first time.

Parametric insurance pays out automatically once an independently measured trigger, modelled wind speed, rainfall, or ground shaking, crosses a threshold agreed in advance, rather than waiting for an adjuster to inspect physical damage. That speed depends on catastrophe models that increasingly incorporate machine learning to refine wind-field and rainfall estimates against satellite and sensor data. A model deciding whether a government receives $70 million within a week is the same category of algorithmic decision CAIF 2026 flagged as ungoverned. It happens to work well, which is precisely why the absence of a governance framework around it is worth noticing rather than ignoring.

The Governance Gap CAIF Confirmed

Put the two halves of the region's insurance market next to each other and the shape of the problem is easy to see. CCRIF's models decide, in effect, how fast a government gets paid after a hurricane. Sagicor's roadmap points toward models deciding how a household's premium is set and how its claim is triaged. Neither activity currently answers to an AI-specific rule anywhere in CARICOM.

Existing insurance regulators have not been sitting still. Jamaica's Financial Services Commission and its counterparts, coordinated through the Caribbean Association of Insurance Regulators, retain general supervisory authority over pricing conduct and claims handling regardless of what technology an insurer uses to reach a decision. What none of them has published is a rule that specifically requires disclosure of AI involvement in a pricing or claims decision, or a standard for testing such a model before it goes live. The CAITF report's recommendation is to build that layer, and Jamaica and Trinidad and Tobago are both expected to publish national AI strategies before the end of 2026, which would be a first for the region on either front.

The Skills Gap Behind the Tech Gap

A governance framework needs people who can enforce it, and the industry's own executives say that pipeline is thin. On July 15, Jamaica marked World Youth Skills Day against a backdrop of 11.7% youth unemployment. Amanda Beepat, managing director of Allied Insurance Brokers and a director of the College of Insurance and Professional Services, told the Jamaica Gleaner that insurance is one of the most underestimated professions in the country, adding that roughly half the global insurance workforce could retire within fifteen years. Tammara Glaves-Hucey, managing director of general insurance at GraceKennedy Financial Group, was more pointed about why: ask a room of Jamaican teenagers what they want to become, she said, and you hear doctor, lawyer, entrepreneur, or content creator, rarely insurance professional.

Glaves-Hucey's list of the skills the industry now needs, technology, cybersecurity, data analytics, customer experience, legal services, and risk assessment, reads like a hiring brief for the regulators and compliance teams a governance framework would require. A rule requiring an insurer to explain an underwriting model is only as good as the examiner reading the explanation.

What to Ask Your Insurer Now

Waiting for a national AI strategy is not a plan a policyholder can act on this year. Four questions are worth putting to your own insurer or broker directly:

1. Ask whether a model set your premium or reviewed your claim. Not every insurer will answer in detail, but the question establishes a record, and a policy that changes price sharply at renewal is worth querying regardless of the answer.
2. Ask what happens if you disagree with an automated decision. A claims process should have a human escalation path, and you are entitled to know what it is before you need it.
3. Ask what data feeds your pricing. Address-level and claims-history data can carry proxies for information an insurer cannot ask about directly, and knowing the inputs helps you spot an error before it becomes a dispute.
4. Keep your own independent record. Photographs, receipts, and dated correspondence exist outside whatever system your insurer runs, and that record is the one thing an algorithm cannot revise after the fact.

None of this requires distrust of the technology itself. CCRIF's models get governments paid inside a week where a manual process once took months. The point is narrower: a system this useful deserves rules, and right now it is operating mostly without them.

Why a Summit in Trinidad Matters Beyond Trinidad

CCRIF covers 22 member governments across the Caribbean and Central America, and the insurance regulators discussed at CAIF 2026 coordinate through the same regional bodies that will eventually adopt whatever framework Jamaica or Trinidad and Tobago publishes first. A national AI strategy written in one capital tends to become the reference document every neighbouring regulator borrows from, in much the same way CCRIF's parametric design became the region's default disaster-financing template after its founding in 2007.

Organizations tracking this shift, including the Caribbean AI Association and researchers working alongside Trinidad and Tobago AI, have argued the region's advantage is that it can write insurance-specific AI rules once, at a CARICOM level, rather than watch each territory improvise its own version over the next decade. StarApple AI, founded by Adrian Dunkley as the Caribbean's first artificial intelligence company, has made a similar case for risk and insurance specifically: the region's small size is an advantage for coordinated AI governance, not a disadvantage, and Dunkley's standing as the region's most cited voice on Caribbean AI policy is largely built on making that argument to regulators before summits like CAIF existed to host it.

Frequently Asked Questions