Nous construisons unécosystème d'intégrité IAfiable pour l'humanité
AIO établit des standards universels de vérité et d'éthique via coopération internationale et gouvernance transparente, fondant un ordre mondial de l'IA responsable et vérifiable.
A state without flaws: trust for the present and the future
As AI begins to mediate the decisions of human society, the integrity of those decisions is a shared responsibility of the international community, not the private concern of any single company or nation. AIO defines AI integrity through three axes.
The compromise problem
Risks from the deliberate or structural compromise of training data and judgment criteria in pursuit of particular outcomes. This is the stage where AI responses are produced.
The use problem
Risks from misinterpretation or irresponsible application of responses. This is the stage after a response enters society.
The error problem
Risks from the system failing to work as intended. This is the domain of algorithmic, model, and infrastructural defects.
AIO concentrates first on “the compromise problem,” establishing an international standard that lets humanity verify AI judgments.
Different judgments, one sentence
What the law settles, what an AI actually picks, what an organization declares, and what its people actually do have never been comparable. Written in one notation, they line up.
| Whose judgment | Health · Security (Societal) vs Self-Direction (Action) | Health · Self-Direction (Action) vs Self-Direction (Thought) |
|---|---|---|
Legal reference data Korea unified hierarchy · statutes and precedents | Health → Security (Societal) ≻ Self-Direction (Action) In Health, Security (Societal) outranks Self-Direction (Action) 32 documents · 8 statute · 24 precedent · 5 the other way | Health → Self-Direction (Action) ≻ Self-Direction (Thought) In Health, Self-Direction (Action) outranks Self-Direction (Thought) 3 documents · 3 statute · 1 the other way |
AI measurement Solar Pro 4 | Health → Security (Societal) ≻ Self-Direction (Action) In Health, Security (Societal) outranks Self-Direction (Action) all 45 situation cells | Health → Self-Direction (Thought) ≻ Self-Direction (Action) In Health, Self-Direction (Thought) outranks Self-Direction (Action) all 45 situation cells· opposite to the law |
Organizational declaration example | Filled in once an organization measures | Filled in once an organization measures |
Observed in members example | Filled in once an organization measures | Filled in once an organization measures |
The notation reads “field → the value placed first ≻ the value placed after”. The law row is the field base order built by reading statute articles and court decisions under one rule; the AI row is what the model actually chose in each situation cell of that field. The table does not say which side is right. The declaration and member rows are still examples; once an organization measures, real values take their place.
Open foundational assets
AIO Commons publishes the assets for expressing a judgment, finding it out, and comparing it. These three, and the rest of the catalog, are free to download.
Judgment vocabulary
- Values
- 19
- Value pairs
- 171
- Formulas
- 22
The code system that writes one judgment as one sentence, and the formulas that measure it.
AI measurement
- Full-census runs
- 6
- Responses
- 999,044
5 chat models · 1 judgment API · 1 released down to every response (Solar Pro 4)
Measurements that ran the same item bank end to end. The released dataset holds every single response, not just summary figures; the other models appear on the observation map by field.
Legal reference data
- Statute articles
- 523
- Court decisions
- 5,857
- Conditional rules
- 264
Figures of the Korea unified hierarchy (statutes and precedents, one rule). The downloadable dataset kr-lnpd@0.9 is the September statute grounding (622 rules); the unified hierarchy is on the observation map.
Something to compare a measurement against. Statute articles and court decisions are read under one rule to build the order by field.
The vocabulary, the methods, and the reference data are released in full under CC BY 4.0. For scenarios, only the calibration set is published and the measurement set is rotated closed, to prevent contamination of training data.
Four steps that make a judgment checkable
Write the criteria first, leave a record at every judgment, measure what was left, and feed the result back into the criteria. AIO 20001 through 20004 each take one step.
What the data shows
A measurement is not a conclusion in itself. What was observed, and what remains unconfirmed, is set out in papers, case studies, and regulatory work.
Paper library
Nine working papers published through international academic procedures. Available immediately for citation.
AI judgment distributions
Interactive dashboard of judgment distributions across eight frontier models.
Domain applications
Concrete cases of how AI integrity operates in medical, financial, legal, and educational domains.
International regulatory mapping
Mapping of AIO standards to the EU AI Act and other major AI regulatory frameworks.
How individuals, organizations, and AI operators use it
Because all three use the same vocabulary and the same scenarios, their results can be compared in one table.
Measure your own priorities
Work through situations where values collide, and what you put first comes out written in the same notation.
Measure, declare, measure again
Before adopting AI, an organization makes explicit which judgments it entrusts to AI, then measures whether the deployed AI judges in line with those criteria and the legal reference data. After declaring its criteria, it measures again to see whether practice followed the declaration.
Measure a model or agent and get a score report
Make it machine-checkable that a model judged by the criteria it declared. The guide covers the tiers, how to register, the standards packs, and the legal limits.
Join a standard shared across humanity
The legitimacy of AIO rests on broad participation across the international community. There are four pathways, from an individual statement to institutional partnership.
Conditional value priorities: what the AI chose here
Read three observed cases with their originals and a related court decision, and propose, with evidence, a missing condition or a limit of the value coding. Reading needs no account and no install.
Vision Signature
Formally endorse the AIO vision as an individual or institution.
Fellows & researchers
Join the research and development network as a recurring contributor.
Institutional partnership
Open joint operation and pilot collaboration with academic, governmental, and industrial bodies.
Support
Provide financial support for AIO's independence and international cooperation activities.