About
Research shaping the future of AI governance.
AIGX Research™ is an independent global research organization. It publishes governance frameworks and industry outlooks, and is currently developing the AIGR™ AI Governance Ratings methodology, for enterprise, government and regulated industries.
FormationAIGX Research™
The organization was established in January 2026 to address the distance between Responsible AI as stated intent and as demonstrable practice. Principles of fairness, transparency, human oversight, contestability and safety are articulated consistently across international instruments, national strategy and corporate policy. AIGX Research publishes frameworks, evidence requirements and assessment criteria before they are applied, so that a governance commitment can be examined rather than assumed.
Market timingResponsible AI moved from principle to obligation, and the instruments did not follow.
For most of the last decade, Responsible AI existed as a body of principles — fairness, transparency, human oversight, accountability — published widely, agreed broadly and enforced rarely. Organizations adopted them sincerely and had no way to demonstrate adherence beyond restating them.
That position has changed on four fronts at once:
Front 01
Binding regulation
The EU AI Act converted principle into statutory obligation, with duties phasing in across 2026 and 2027. Other jurisdictions are legislating or preparing to. Obligations now attach to named systems, not to organizational intent.
Front 02
Certifiable standards
ISO/IEC 42001 established AI management systems as certifiable, and the NIST AI Risk Management Framework became the common reference in the United States. Both describe what good governance looks like; neither produces a comparable measure of how well a given organization is doing it.
Front 03
Procurement
Buyers in regulated sectors began asking suppliers to evidence AI governance before deployment rather than attest to it afterwards. Questionnaires multiplied; answers stayed unverifiable and non-comparable between vendors.
Front 04
Deployment outpacing control
Generative and agentic systems entered production faster than the control environments around them matured, widening the distance between what an organization has deployed and what it can account for.
Each front raises the same question and none of them answers it: how does an organization demonstrate that a specific AI system is governed? Certification confirms a management system exists. An audit opinion addresses financial statements. A questionnaire records assertions. None of them produces a comparable, dated, contestable position on a named system.
That is the instrument AIGX Research builds, and the reason the work began when it did.
MissionTo advance Responsible AI from declared principle to demonstrated practice.
AIGX Research works to establish the evidence by which governance commitments may be examined; to reconcile those commitments with the standards, regulations and duties already in force; to render the resulting judgements comparable across sectors and jurisdictions; and to ensure that no organization, regulator or affected individual is asked to accept assurance on trust alone.
The commitments below define how that purpose is pursued. Each operates first as a constraint on this organization, and only then as an expectation of those it assesses.
Commitment 01
Aligned
Responsible AI principles are resolved into named controls, required artifacts and accountable owners. A principle that cannot be located in a control is a statement of intent, and we treat it as one.
Commitment 02
Integrated
Controls are mapped to the compliance obligations an organization already carries — AI management systems, risk frameworks, sectoral regulation and statutory duties — so governance work is reconciled with existing accountability rather than added beside it.
Commitment 03
Evidenced
A control is credited when an artifact demonstrates it operating, not when a policy states that it should. Every position carries a recorded path to challenge it; an opinion that cannot be argued with is an assertion.
What we produceDistinct outputs, one evidence model.
Output
Sector frameworks
Controls, risk classes and evidence requirements for health, finance, agentic AI and the public sector.
Output
Assessments
Scoped reviews of a named system against the applicable framework, recorded as findings, reviewer states and open gates.
Output
AIGR™ methodology
A categorical governance designation, scoped to a named system and date. The methodology is in development; no ratings have been issued.
Output
Published research
Market outlooks and benchmarks, each carrying a reference, a version and a review state.
IndependenceThe people who help an organization improve are not the people who approve its outcome.
Assessment, remediation support and rating issuance are separated functions. Criteria are published before findings are scored against them. Commercial terms are fixed before assessment begins and are not contingent on the outcome. Where a conflict cannot be managed through separation of function, the engagement is declined or the rating is withheld.
These are not aspirations. They are the conditions under which a governance opinion is worth anything at all, and they are set out in full on the methodology page.
PositionWhat this organization is, and what it is not.
| AIGX Research is | AIGX Research is not |
|---|---|
| An independent global research organization | An accredited certification body |
| A publisher of governance frameworks and criteria | A standards-setting authority |
| An issuer of scoped, dated governance opinions | An auditor, or a provider of assurance under any professional standard |
| A source of decision support for boards, buyers and risk teams | A regulator, or a route to regulatory approval |
EnquiriesFramework access, assessment scope and research collaboration.
Enquiries are read by the research team. The more precisely an enquiry states the system, the sector and the stage it has reached, the more directly it can be answered.
Disclaimer
AIGX™ assessments and ratings are independent governance evaluations and do not constitute certifications, regulatory approvals, legal opinions, or attestations of compliance. References to third-party standards, regulations, and frameworks are provided for alignment purposes only and do not imply affiliation, endorsement, sponsorship, or certification by their respective owners.