Readiness · Primary index
Infrastructure + Talent + Governance + Investment + Economic Readiness0–20 + 0–20 + 0–20 + 0–20 + 0–20 = 0–100Each pillar is rounded to the nearest whole point before the five pillars are added.
Research protocol · Version 3 · 2026 edition
How the index is built
The Readiness Index combines five 20-point pillars into a 100-point score. Trajectory is a separate directional measure. Adoption is a separate lens and never changes the Readiness score.
01
Formula overview
Two analytical lenses, one transparent readiness calculation.
Trajectory · Directional
Seven signals measuring momentum−10 to +10Projected 2028 = Readiness + (Trajectory × 1.5), clamped from 0 to 100.
Adoption · Separate lens
Observed use across five domains0–100Adoption is curated independently and does not feed the Readiness or Trajectory formulas.
02
Pillar-weight table
Every readiness point is assigned to a defined component.
| Pillar | Component | Points | Preferred source | Fallback |
|---|---|---|---|---|
| Infrastructure20 points | Fixed broadband subscriptions | 6 | World Bank Open Data | Use 6/20 of the static Infrastructure baseline. |
| Individuals using the internet | 3 | World Bank Open Data | Use 3/20 of the static Infrastructure baseline. | |
| Mobile subscriptions | 3 | World Bank Open Data | Use 3/20 of the static Infrastructure baseline. | |
| Access to electricity | 3 | World Bank Open Data | Use 3/20 of the static Infrastructure baseline. | |
| Infrastructure quality baseline | 5 | AI Trajectory Index | This designed residual has no live replacement. | |
| Talent20 points | Gross tertiary enrolment | 6 | World Bank Open Data | Use 6/20 of the static Talent baseline. |
| Labour productivity | 6 | World Bank Open Data | Use 6/20 of the static Talent baseline. | |
| Researchers | 2 | OECD Data Explorer | Return this allocation to the static Talent baseline. | |
| Global Innovation Index | 1 | World Intellectual Property Organization | Return this allocation to the static Talent baseline. | |
| Talent quality baseline | 5 | AI Trajectory Index | This designed residual has no live replacement. | |
| Governance20 points | Rule of Law | 5 | Worldwide Governance Indicators | Use 5/20 of the static Governance baseline. |
| Government Effectiveness | 4 | Worldwide Governance Indicators | Use 4/20 of the static Governance baseline. | |
| Regulatory Quality | 3 | Worldwide Governance Indicators | Use 3/20 of the static Governance baseline. | |
| National AI strategy | 3 | OECD.AI Policy Observatory | No strategy receives 0 points. | |
| AI strategy recency | 2 | OECD.AI Policy Observatory | No dated strategy receives 0 points. | |
| AI regulation in force | 1 | OECD.AI Policy Observatory | No regulation flag receives 0 points. | |
| Government AI readiness | 2 | Oxford Insights | Use 2/20 of the static Governance baseline. | |
| Investment20 points | R&D expenditure | 6 | OECD Data Explorer, then World Bank Open Data | Use 6/20 of the static Investment baseline. |
| FDI net inflows | 5 | World Bank Open Data | Use 5/20 of the static Investment baseline. | |
| Government expenditure | 3 | IMF DataMapper | Return this allocation to the static Investment baseline. | |
| Capital ecosystem baseline | 6 | AI Trajectory Index | This designed residual has no live replacement. | |
| Economic Readiness20 points | GDP per capita, PPP | 6 | World Bank Open Data | Use 6/20 of the static Economic Readiness baseline. |
| Private-sector credit | 3 | World Bank Open Data | Use 3/20 of the static Economic Readiness baseline. | |
| Trade openness | 3 | World Bank Open Data | Use 3/20 of the static Economic Readiness baseline. | |
| Services value added | 3 | World Bank Open Data | Use 3/20 of the static Economic Readiness baseline. | |
| Market capacity baseline | 5 | AI Trajectory Index | This designed residual has no live replacement. |
03
Indicator data dictionary
Search the production inputs, codes, thresholds, and source roles.
29 entries
| Indicator | Pillar | Code / unit | Source role | Scoring rule |
|---|---|---|---|---|
| Fixed broadband subscriptionsProxy for high-quality digital access. | Infrastructure | IT.NET.BBND.P2per 100 people | Live sourceWorld Bank Open Data | >35: 6; >20: 4.5; >8: 3; >2: 1.5; otherwise 0.5 |
| Individuals using the internetMeasures broad digital access. | Infrastructure | IT.NET.USER.ZS% of population | Live sourceWorld Bank Open Data | >85: 3; >65: 2.2; >40: 1.5; >15: 0.8; otherwise 0.3 |
| Mobile subscriptionsMeasures mobile network reach. | Infrastructure | IT.CEL.SETS.P2per 100 people | Live sourceWorld Bank Open Data | >120: 3; >80: 2.2; >40: 1.5; otherwise 0.5 |
| Access to electricityCaptures a necessary condition for compute and connectivity. | Infrastructure | EG.ELC.ACCS.ZS% of population | Live sourceWorld Bank Open Data | >97: 3; >80: 2; >50: 1; otherwise 0.3 |
| Infrastructure quality baselineRepresents server density, cloud availability, grid reliability, and related quality signals. | Infrastructure | Maintained fieldNot applicable | Static baselineAI Trajectory Index | Award 5/20 of the static Infrastructure baseline. |
| Gross tertiary enrolmentMeasures the volume of the advanced education pipeline. | Talent | SE.TER.ENRR% gross | Live sourceWorld Bank Open Data | >70: 6; >50: 4.5; >30: 3; >15: 1.8; otherwise 0.8 |
| Labour productivityMeasures whether human capital is productively deployed. | Talent | SL.GDP.PCAP.EM.KDconstant 2017 PPP dollars per worker | Live sourceWorld Bank Open Data | >85,000: 6; >65,000: 4.5; >40,000: 3; >18,000: 1.5; otherwise 0.75 |
| ResearchersAvailable primarily for OECD members and partners. | Talent | MSTI T_RS.10P3EMPFTE per 1,000 employment | Live sourceOECD Data Explorer | >10: 2; >6: 1.5; >3: 1; otherwise 0.5 |
| Global Innovation IndexMaintained annual enrichment, not a live API input. | Talent | WIPO_GIIscore, 0 to 100 | Maintained sourceWorld Intellectual Property Organization | >55: 1; >40: 0.7; >25: 0.4; otherwise 0.2 |
| Talent quality baselineRepresents education quality, ICT skills, researcher capacity, and brain-drain conditions. | Talent | Maintained fieldNot applicable | Static baselineAI Trajectory Index | Award the unreplaced share of the 8-point static Talent allocation. |
| Rule of LawMeasures contract enforcement, rights, and judicial quality. | Governance | GOV_WGI_RL.ESTestimate, -2.5 to 2.5 | Live sourceWorldwide Governance Indicators | Normalize continuously from -2.5 to 2.5. |
| Government EffectivenessMeasures public-service and policy implementation capacity. | Governance | GOV_WGI_GE.ESTestimate, -2.5 to 2.5 | Live sourceWorldwide Governance Indicators | Normalize continuously from -2.5 to 2.5. |
| Regulatory QualityMeasures capacity for market-supporting regulation. | Governance | GOV_WGI_RQ.ESTestimate, -2.5 to 2.5 | Live sourceWorldwide Governance Indicators | Normalize continuously from -2.5 to 2.5. |
| National AI strategyMaintained national-policy field. | Governance | Maintained fieldyes or no | Maintained sourceOECD.AI Policy Observatory | A published national strategy receives 3 points. |
| AI strategy recencyMaintained policy date. | Governance | Maintained fieldyear | Maintained sourceOECD.AI Policy Observatory | 2022 or later: 2; 2019 to 2021: 1; older or absent: 0. |
| AI regulation in forceMeasures existence, not regulatory quality. | Governance | Maintained fieldyes or no | Maintained sourceOECD.AI Policy Observatory | A regulation flag receives 1 point. |
| Government AI readinessMaintained annual enrichment, not a live API input. | Governance | OXFORD_GARIscore, 0 to 100 | Maintained sourceOxford Insights | >=65: 2; >=45: 1.5; >=30: 1; otherwise 0.5 |
| R&D expenditureOECD data takes precedence when available; the World Bank series is otherwise used. | Investment | GB.XPD.RSDV.GD.ZS% of GDP | Live sourceOECD Data Explorer, then World Bank Open Data | >3.5: 6; >2: 4.5; >1: 3; >0.3: 1.5; otherwise 0.5 |
| FDI net inflowsMeasures realised foreign capital attraction. | Investment | BX.KLT.DINV.WD.GD.ZS% of GDP | Live sourceWorld Bank Open Data | Cap at 20; >10: 5; >4: 4; >1.5: 3; >0: 2; otherwise 0.5 |
| Government expenditureReplaces 3 points of the capital-ecosystem baseline when available. | Investment | G_X_G01_GDP_PT% of GDP | Live sourceIMF DataMapper | >=45: 3; >=35: 2; >=25: 1; otherwise 0.5 |
| Capital ecosystem baselineRepresents venture activity, startup density, deal flow, and related capital-market depth. | Investment | Maintained fieldNot applicable | Static baselineAI Trajectory Index | Award the unreplaced share of the 9-point static Investment allocation. |
| GDP per capita, PPPPurchasing-power-adjusted economic capacity. | Economic Readiness | NY.GDP.PCAP.PP.KDconstant 2017 international dollars | Live sourceWorld Bank Open Data | >55,000: 6; >30,000: 4.5; >12,000: 3; >4,000: 1.5; otherwise 0.5 |
| Private-sector creditMeasures financial depth available to firms. | Economic Readiness | FS.AST.PRVT.GD.ZS% of GDP | Live sourceWorld Bank Open Data | >150: 3; >80: 2.2; >40: 1.5; >15: 0.8; otherwise 0.3 |
| Trade opennessMeasures access to international markets and supply chains. | Economic Readiness | NE.TRD.GNFS.ZS% of GDP | Live sourceWorld Bank Open Data | Cap at 200; >120: 3; >70: 2.2; >40: 1.5; >20: 0.8; otherwise 0.3 |
| Services value addedProxy for knowledge-economy capacity. | Economic Readiness | NV.SRV.TOTL.ZS% of GDP | Live sourceWorld Bank Open Data | >75: 3; >60: 2.2; >45: 1.5; >30: 0.8; otherwise 0.3 |
| Market capacity baselineRepresents economic complexity, digital-market capacity, and business conditions. | Economic Readiness | Maintained fieldNot applicable | Static baselineAI Trajectory Index | Award 5/20 of the static Economic Readiness baseline; subtract 1 when IMF debt exceeds 100% of GDP. |
| Government debt modifierA modifier, not a separately weighted component. | Economic Readiness | GGXWDG_NGDP% of GDP | Live sourceIMF DataMapper | Above 100 subtracts 1 point from the market-capacity allocation. |
| GDP per capita growth trajectory inputRegistered here for data-dictionary completeness; it does not affect Readiness points. | Economic Readiness | NY.GDP.PCAP.CD% change | Live sourceWorld Bank Open Data | Used only in the separate Trajectory score. |
| High-technology exports trajectory inputRegistered here for data-dictionary completeness; it does not affect Readiness points. | Investment | TX.VAL.TECH.MF.ZS% of manufactured exports | Live sourceWorld Bank Open Data | Used only in the separate Trajectory score. |
04
Source precedence
The scorer follows one deterministic path for each input.
- 1Live public input
Use the latest available World Bank, IMF, or OECD observation.
- 2Maintained enrichment
Use WIPO, Oxford Insights, or maintained national-policy data only where the formula defines it.
- 3Proportional fallback
When a preferred value is unavailable, use that component's share of the static pillar baseline.
- 4Pillar rounding
Sum component contributions, round the pillar, and clamp it from 0 to 20.
- 5Readiness total
Add the five rounded pillars and clamp the result from 0 to 100.
Stanford HAI, Tortoise, ITU, and WEF provide useful context and calibration references, but they do not directly contribute points to the production score.
05
Missing-data policy
Coverage is disclosed separately from a country's performance.
Substitution rule
Missing values receive the proportional share of the country's static pillar baseline. The underlying static baseline is never rewritten during a live request.
Coverage labels
- High
- At least 75% observed and no more than 30% fallback allocation.
- Medium
- At least 50% observed and no more than 55% fallback allocation.
- Limited
- All other source-availability profiles.
What coverage means
Coverage measures source availability. It is not statistical confidence, a margin of error, or a judgment about data quality.
06
Limitations
Use the index as a structured comparison, not a precise forecast.
- Threshold sensitivity. Small input changes near a band boundary can change component points.
- Uneven observation years. Public indicators update on different schedules and may lag current conditions.
- Baseline judgment. Static residual components encode expert judgment where comparable global data is incomplete.
- Projection limits. The 2028 value is a directional scenario based on current momentum, not a probability-weighted forecast.
- Adoption separation. The Adoption lens is curated from a different evidence base and should not be read as part of Readiness.
07
Downloads and citation
Reuse the method with the version and canonical URL intact.
CSVMethodology data dictionaryComponent keys, source URLs, point budgets, rules, and notes.CODESource and READMEScoring implementation, data clients, and full technical documentation.
Suggested citation
Mishra, A. (2026). AI Trajectory Index (Version 3). https://ai-trajectory-index.vercel.app
Canonical URL: https://ai-trajectory-index.vercel.app/methodology