Search wider
Use A.I. to interrogate large collections of sources and identify connections that a single-subject reading may miss.
Method · Sources · Interpretation
RESEARCH
STANDARD
The research exists to make the world easier to navigate. The Almanac uses A.I. as an active research instrument to help process, compare and challenge large bodies of evidence, while keeping source checking, uncertainty and editorial judgement visible.
Why use A.I. this way?
Because the problem is too much information, not too little.
Important changes now unfold across technology, economics, politics, health, culture, education and social life at the same time. A.I. can help connect those fields, search for contradictions and keep a large evidence map current. It is a tool for orientation, not an authority.
Compare faster
Test how the same issue looks across countries, institutions, datasets and time periods without assuming one system is universal.
Challenge claims
Use contradictions as prompts for deeper checking rather than smoothing them away into a convenient answer.
Return to the reader
Ask what the evidence changes for work, money, housing, relationships, health, trust, opportunity and everyday decisions.
Source hierarchy
Start with evidence closest to the event.
A.I.-assisted discovery is never treated as evidence by itself. Where possible, claims begin with primary statistical, regulatory, institutional or technical material and are then cross-checked against independent and international sources.
| Layer | Typical sources | Purpose |
|---|---|---|
| Primary | National statistics, regulators, legislation, public datasets, company filings, technical documentation | Establish what was measured, announced or formally changed. |
| International | OECD, ILO, WHO, World Bank and comparable multilateral bodies | Improve comparability and provide external benchmarks. |
| Academic | Peer-reviewed research, major research institutes and longitudinal studies | Test mechanisms, causal claims and longer-run effects. |
| Independent analysis | Credible specialist reporting, sector analysis and transparent datasets | Challenge official interpretation and add context unavailable in primary sources. |
Editorial tests
Four questions before a claim survives.
Not every interesting statistic belongs in the Almanac. Evidence is retained when it survives tests of relevance, comparability, corroboration and material significance.
1. What exactly was measured?
Definitions, sample, geography, period and methodology matter. Similar labels can conceal different measures.
2. Does another source agree?
Agreement increases confidence. Disagreement is not discarded; it becomes part of the analysis.
3. Does comparison add anything?
A second country is included only when it changes, qualifies or strengthens the interpretation.
4. What changes for people?
The final test is material consequence: work, income, housing, family life, health, trust, opportunity, autonomy or meaning.
Evidence labelling
Fact is not forecast.
Fast-moving subjects become misleading when observation, inference and prediction are written in the same voice. The Almanac keeps those categories distinct so the reader can tell what is known, what is inferred and what remains a bet on the future.
Observed
Directly supported by current evidence or a documented event.
Inferred
A reasoned interpretation supported by evidence but not directly measured as stated.
Forecast
A forward-looking judgement that may be useful, but remains contingent and revisable.
International comparison
The 2027 edition introduces a recurring China and East Asia lens. It does not assume that Western evidence generalises globally, nor that differences elsewhere automatically invalidate the Western case.
Asia is not treated as a single unit. China, Japan and South Korea form the default East Asian comparison group; Taiwan, Hong Kong and Singapore are added when statistically or culturally relevant, with India and Southeast Asia used where the subject requires them.
Non-dilution
Broader geographical coverage must not produce shallower research.
The Almanac therefore favours deep comparisons over national-statistic collections. A country appears because its evidence helps explain the subject and improve the reader's map, not because a quota of territories needs to be represented.
Corrections and revision
An annual almanac is a snapshot, not a claim to permanent certainty. Later evidence may revise earlier conclusions. The series is designed so that those changes can be shown rather than hidden.