About the Index
A measurement instrument, built in the right order.
Most AI rankings start with a league table and look for numbers to fill it. This one starts with the assessment, and will publish comparisons only when there is verified evidence to support them.
Evidence before ranking
No comparison is published until enough verified assessments exist to support it. Where there is no evidence, the interface says so rather than filling the space.
A published method
Weights, score bands and completion rules are public and versioned. Any score can be reproduced from the answers that produced it.
Deterministic scoring
Scores come from a pure, tested calculation. AI explains and drafts; it never produces, adjusts or reviews a number.
Built for more than one instrument
Government, SME, sector and literacy assessments drop into the same platform without a rewrite — the schema was designed for them from the start.
What the index measures
The AI Readiness Index assesses how prepared an organization is to adopt artificial intelligence — not how much AI it has bought. Readiness is the combination of direction, governance, data, infrastructure, people, operational integration, external capability and consequence. An organization strong in technology but absent in governance is not ready; neither is one with an approved strategy and unusable data.
Assessment runs across eight dimensions. Each is scored 0–100 from the participant’s own answers and combined into an overall readiness score mapped onto a five-level maturity scale. Only the core question set — asked of every organization in identical wording — produces the score, so two organizations remain comparable even when the deeper diagnostic questions they see differ. The full method is published on the methodology page.
What this index is not
A score here is a structured self-assessment, scored by published rules. It is not a certification, an accreditation, an audit opinion, or a statement of compliance with any law or standard — including the EU AI Act, ISO/IEC 42001 and the NIST AI Risk Management Framework, however much the questions draw on the same concerns.
Answers are self-reported. Where a reviewer has opened the supporting evidence and confirmed it, the answer is marked as verified, and every result reports how much of it carries evidence and how much of the question set was answered at all. Those figures sit next to the score and never adjust it: a strong score on thin coverage is a weaker claim than the same score on complete answers, and the index says so rather than smoothing the difference away.
Who it is for
Government ministries and public authorities, state-owned entities, private enterprises, financial institutions, healthcare providers, universities and non-profits — in the Middle East and North Africa first, and beyond it as the database grows. The instrument is deliberately sector-neutral: it asks about capability and governance, which are comparable across sectors, rather than about specific technologies, which are not.
It is designed to be completed by someone with an organization-wide view: a director of digital transformation, a chief information or technology officer, a strategy or governance lead. It takes about twenty-five minutes.
What a participant receives
- An overall readiness score with a maturity level and a plain-language description of what it means.
- A score for each of the eight dimensions, with the relative strengths and priority gaps identified.
- Recommended actions drawn from a published set, matched to the dimensions where the gap is largest.
- A results dashboard that prints cleanly, because it will be shown to boards and ministers.
How the data is handled
Individual responses are confidential. They are visible to the participant, to colleagues they add to the same organization, and to the index review team. They are never published, and they are never sold. Where a participant has explicitly opted in, their answers may contribute to aggregated, anonymized statistics — for example a sector average — in which no organization can be identified.
Full detail is in the Privacy Policy.
Why there are no rankings yet
Collect credible data first. Build rankings from verified evidence. That is the principle this platform is built on, and it is the reason the benchmarking panel on the results page is empty rather than populated with an estimate. Comparative figures will be published when enough verified assessments exist for a peer group to mean something — and each published figure will be traceable to a recorded source with a retrieval date.
Who publishes it
The index is published by Brain Digits. Questions, partnership proposals and data contributions are welcome at aiindex@braindigits.com.