Our Thesis
The world cannot be reasoned about credibly on vibes. What happens when AI reshapes work, health, food, and society — and what does the evidence actually show?
The question that frames everything we build is simple: what happens when AI starts to take over most jobs — and how does that cascade into health, food systems, social structures, and the way people relate to each other?
That question does not have a settled answer. But it can be investigated with evidence. The signals are already present in the data: AI capability is advancing faster than most models predicted; employment patterns in certain sectors are already shifting; demographic pressures are compounding economic ones; health systems are under strain in ways that intersect with technological and economic trends.
The future cannot be reasoned about credibly on vibes. Decisions about capital, policy, product strategy, and social support all depend on the same underlying signals — they just read them differently.
Apiar Data is the search engine that makes those signals reachable. We index, normalise, and make searchable the world's institutional evidence on how things are changing — across AI, labor, health, population, economy, and environment. Built for AI agents to navigate at scale, with every dataset traced back to its original source.
Our Approach
Selective Sourcing
We only index data from verifiable institutional sources — official statistical agencies, central banks, international organisations, and peer-reviewed repositories.
Standardised Format
Raw data arrives in dozens of formats. We normalise everything to consistent time series structures, making cross-source comparison straightforward.
Continuous Updates
Automated pipelines monitor source agencies for revisions and new releases. Data freshness is tracked and displayed on every dataset page.
Full Documentation
Every dataset is accompanied by methodology notes, source attribution, coverage details, and known limitations — the context that transforms numbers into knowledge.
API-First Access
All data is accessible programmatically via a structured API, enabling integration into research workflows, models, and analytical tools.
Open Licences
Where source licences permit, we release data under open licences that allow reuse, sharing, and derivative work with attribution.
Impact & Reach
These numbers reflect the platform's scale, but the mission is measured differently — in research that gets done faster, in models built on better data, in decisions informed by context that was previously inaccessible.
How We Select Sources
Not all data is equal. Our source selection process prioritises:
- Institutional provenance: Data produced by official statistical agencies, central banks, or recognised international bodies
- Methodological transparency: Sources that publish their collection and calculation methodology
- Revision policy: Sources that communicate revisions clearly and maintain historical revision records
- Licence compatibility: Sources that permit redistribution, aggregation, or derivative work
- Coverage consistency: Sources that maintain consistent coverage over time, enabling meaningful historical analysis
Sources that do not meet these criteria may still be indexed with appropriate quality flags, but are not included in Apiar's primary curated datasets.