CAIARESEARCH ATLAS

The Caribbean work atlas

Choose your work. Click a place. Explore the next five years.

Select a place
0
◌ Regional baseline

Local research where present. Labelled regional baselines elsewhere. All results are conditional.

Full study ↗

Your profile & assumptions

Adjust what matters. Empty fields use a baseline.

With no pay entered, the pay index starts at 100. Pay-bump chance 20%, bump size 5%, annual growth 0%; graduate offer chance 40%. These are illustrative assumptions, not observed averages or calibrated probabilities. Current-worker nominal base pay cannot fall. The graduate adjustment affects the starting offer once.

Your inputs stay in this browser tab. Country, role and industry are not used to invent personal wage or hiring probabilities.

Method & sources

How regional baselines work

For each missing economic indicator, use the unweighted mean of the non-missing country values in the released EAEI snapshot. Keep each available local indicator. Calculate the score using the original EAEI formula. The regional inputs are proxies, not measurements of the selected place.

For a missing national employment mix, use the unweighted average of the 12 complete CAIRMC country mixes, normalised to 100%. Keep local internet use where present; otherwise use the mean of the 16 country access values. Run the unchanged country scenario formula with these inputs. No national model is fitted for the missing place.

An unmatched job uses the unweighted mean exposure of all 427 source occupation groups. This is not an employment-weighted regional average. Blank, invalid or out-of-range numerical entries use the displayed planning baseline, with the fields identified in the result.

How to interpret the map

The work map shows displacement-equivalent pressure per 100 positions in the selected broad sector. Detailed industries map to agriculture, industry or services; the underlying study does not estimate each detailed industry separately. Years one to four are a linear interpolation of the published five-year endpoint, not new annual model runs. Role exposure and the role-specific scenario are displayed separately.

The economic map shows the EAEI foundation score using 2022–2024 indicator snapshots. It stays fixed when you change the planning year. The selected year updates the separate work-value scenario. The EAEI score is not converted into GDP growth or a wage forecast.

Personal scenarios

Pay(t) = starting pay × [(1 + annual growth) × (1 + annual pay-bump chance × bump size)]ᵗ. The chance of at least one bump or offer over t years is 1 − (1 − annual chance)ᵗ. Constant annual probabilities and independent annual opportunities are assumptions. Graduate offer outputs describe at least one offer, not continuous employment. No personal probability has been calibrated.

Annual gross work value = annual payroll × capacity gain × (year ÷ 5) × useful-value capture. Cumulative value sums annual gross value less annual costs, then subtracts setup cost once. It is payroll-equivalent useful work, not cash savings or GDP. Country scores, experience and organisational level do not secretly change the wage or value assumptions.

Sources

All professions

Independent CAIA research by Adrian Dunkley. No UN, ILO or World Bank endorsement is claimed.

Attribution guide

Cite: Dunkley, Adrian. “Caribbean AI Research Atlas.” Caribbean AI Association, updated 15 November 2026. Include the page URL, access date, country, occupation, horizon and assumptions when sharing results. Label regional proxies and user-adjusted scenarios. Do not present these results as individual forecasts or imply ILO, World Bank or UN endorsement.

Occupation evidence: ILO Working Paper 140 (2025). Country inputs and formulas: CAIRMC and EAEI, linked under Method & sources. Source licences continue to apply. Try the displacement scenario orb →