There is no research ethics law in Trinidad and Tobago. There is no research ethics law in Jamaica. Of all the CARPHA member states in the Caribbean, only Guyana has enacted comprehensive legislation governing research with human participants (CARPHA/CANREC). And yet the region’s institutional review committees meet monthly, train annually, and follow US regulatory precedent as though the law existed. The question is not whether the Caribbean takes research ethics seriously. The question is what happens when you build an AI system that governs itself more rigorously than any jurisdiction requires.
The Gap
In 2022, Aguilera, Carracedo, and Saenz published a systemic assessment of research ethics systems across Latin America and the Caribbean in the Lancet Global Health, and the findings were stark. Jamaica, Nicaragua, and Trinidad and Tobago have not established the national oversight body that the assessment’s indicators require. Jamaica’s research ethics policies are not legally binding. Trinidad and Tobago has five institutional research ethics committees, trained by CARPHA in a January 2016 workshop, but no statute that compels their authority. CARPHA established CANREC, a regional network of 21 research ethics committees, in March 2016 at a three-day workshop in Barbados, but CANREC is coordination, not legislation.
The University of the West Indies at St. Augustine has a functional Campus Research Ethics Committee. It meets on the last Wednesday of each month, requires CITI ethics training for all named investigators, accepts applications through the UWIScholar platform on a first-Thursday-of-the-month deadline, and offers four review tracks: full, expedited, exemption, and waiver. The NCRHA Ethics Committee in Trinidad oversees compliance with the Ethical Guidelines for Biomedical Research on Human Subjects across the regional health authority system, staffed by a chairperson, a secretary, and thirteen members including ethics experts, scientists, clinicians, a legal expert, and a social worker. Both committees follow review procedures that parallel 45 CFR 46, the US federal regulation for the protection of human subjects.
But parallel is the operative word. Trinidad and Tobago follows US regulatory precedent informally. The committees exist. The procedures exist. The legislation does not.
The Ledger
CaribChat is a governed AI cancer navigation service for Caribbean populations, launching at the CAOH 2026 Annual Scientific Conference in Port of Spain this July. Since March 2, 2026, the community learning ledger has captured over fifty-five sessions from real patients, caregivers, and clinicians navigating cancer care across the Caribbean. A caregiver asks where to get screened in Port of Spain. A clinician queries prostate cancer prevalence in Jamaica. A grandmother asks whether soursop tea is safe during chemotherapy.
The ledger captures questions, not people. No names, no email addresses, no IP addresses, no demographics. The architecture was designed from inception so that identity is never collected, which is a fundamentally different posture from de-identification, where identity is collected and then removed. The ledger is append-only and cryptographically verified. The governance framework enforces these constraints architecturally, not through policy manuals that investigators might overlook or administrators might waive. The system cannot violate its own protections because the protections are the system.
The Governance
The CANONIC governance framework measures structural completeness across every service in the portfolio. CaribChat meets the full governance standard. Every conversation turn is ledgered. Every clinical claim traces to a governed evidence source drawn from CAOH conference proceedings, CARPHA cancer surveillance, NCCN resource-stratified guidelines, and seven other curated layers. The system cannot diagnose, cannot prescribe, and cannot present US-only guidelines without Caribbean resource stratification context. These are not aspirational policies; they are architectural constraints enforced at the code level.
The result is a community learning system whose data protections exceed anything that legislation in the Caribbean currently requires, because no legislation currently requires anything. But the system was not built for regulatory arbitrage. It was built because structural governance is better than procedural governance, regardless of jurisdiction.
The First
This is, to our knowledge, the first distributed ledger of federated community learning in the Caribbean, and possibly the first anywhere. Not a blockchain. Not a token scheme. A governed, append-only record of what real patients, caregivers, and clinicians actually ask when they navigate cancer care, captured without identity, verified cryptographically, and compounding into community intelligence that improves with every session. The ledger belongs to the community that generated it.
The IRB we are building is modular and open, designed to extend across a global federation. The Caribbean is the first node. Florida is the second, where MammoChat already captures community learning for breast health navigation in the United States. Europe, Latin America, Africa, Asia, the Pacific: each node joins the same governed framework, the same structural protections, the same ledger architecture, adapted for local evidence layers, clinical context, and healing traditions. One IRB protocol covers all of them because the governance is identical everywhere it runs.
Peer-reviewed journals require IRB approval or an exempt determination before they will consider a manuscript for publication. The ICMJE is explicit: “All authors should seek approval to conduct research from an independent local, regional or national review body.” That is the structure of scientific publishing, and it does not matter whether the jurisdiction where the data originates has research ethics legislation or not. The journal requires it. Full stop. We are filing through a US IRB, because the Caribbean has no legislation that would make a local filing anything other than arbitrary. The exempt determination files under 45 CFR 46.104(d)(4)(ii): secondary research use of information recorded without identifiers, where the investigator does not contact subjects and will not re-identify them.
The Precedent
Trinidad and Tobago will eventually enact research ethics legislation. When it does, the legislators and regulators who draft that law will look to existing frameworks for precedent. The US Common Rule is the obvious model, and the Caribbean’s institutional committees already follow it informally. But there is now a second model on the table: structural governance, where the protections are not policies that people follow but constraints that systems enforce. A governed ledger with no personally identifiable information, cryptographic integrity, and continuous compliance is not a substitute for legislation. It is a complement that makes legislation enforceable at a level that procedural review alone cannot achieve.
The IRB we are filing has no law behind it in the Caribbean. That is not a weakness. It is the entire point.
Join Us
We are building a global federation of governed community learning. If you are a clinician, researcher, cancer society advocate, or public health leader who wants to join, talk to us. If you are a regulatory scholar interested in structural governance as a complement to legislative frameworks in jurisdictions that have none, talk to us. If you are a patient or caregiver who has navigated cancer care and wants your experience to help the next person, CaribChat is live and every question you ask becomes part of the community’s intelligence.
The CAOH 2026 Annual Scientific Conference convenes July 17-19 at the Hilton Trinidad in Port of Spain. Come find us. Or
Sources
| Claim | Source | Reference |
|---|---|---|
| Jamaica, T&T lack national oversight body | Aguilera B, Carracedo S, Saenz C. Lancet Glob Health 2022;10(8):e1204-e1208 | pubmed.ncbi.nlm.nih.gov/35716677 |
| Jamaica’s research policies not legally binding | Aguilera et al. 2022 (same) | DOI: 10.1016/S2214-109X(22)00128-0 |
| Only Guyana has comprehensive legislation among CARPHA states | CARPHA/CANREC; Dr. Aarons green paper to CARICOM Ministers of Health (2015) | carpha.org/CANREC |
| CANREC: 21 RECs, est. March 2016, Barbados workshop | CARPHA CANREC | carpha.org/CANREC |
| T&T has five RECs, CARPHA training Jan 2016 | CANREC Bulletin Vol. 1 | researchgate.net/publication/337740189 |
| UWI STA: last Wednesday monthly, CITI required, UWIScholar, first Thursday deadline | UWI St. Augustine Campus Research Ethics Committee | sta.uwi.edu/research/campus-ethics |
| NCRHA: Ethical Guidelines for Biomedical Research on Human Subjects, 15-member committee | NCRHA Ethics Committee | ncrha.co.tt/ethicscommittee |
| Both committees parallel 45 CFR 46 procedures | Aguilera et al. 2022 (T&T follows US precedent informally) | DOI: 10.1016/S2214-109X(22)00128-0 |
| Journals require IRB approval for publication | ICMJE Recommendations: Protection of Research Participants | icmje.org/recommendations |
| Exempt category 45 CFR 46.104(d)(4)(ii) | HHS OHRP, Federal Policy for the Protection of Human Subjects | hhs.gov/ohrp/regulations |
| 55+ community learning sessions since March 2, 2026 | CaribChat community learning ledger | caribchat.ai |
| CaribChat meets full governance standard | CANONIC governance framework | canonic.org |
| 10 governed evidence layers | CaribChat governance contract | caribchat.ai |
| CAOH 2026 Conference, July 17-19, Hilton Trinidad | CAOH | caohcaribbean.org/conference |
| OHRP International Compilation of Human Research Standards | HHS OHRP (2024 Edition) | hhs.gov/ohrp/international |
| *BLOG | IRB WITHOUT A LAW | 2026-03-18* |