AI maturity
AI maturity measures a company's ability to adopt AI and get measurable value from it, across six dimensions: strategy, data, technology, governance, skills and results.
Updated October 11, 2026
The reality
of organizations use AI in at least one function, but only 39% see an impact on their operating income.
Source: McKinsey, The state of AI 2025
of Tunisian companies consider advanced technologies essential to innovation, yet 19% report having innovated.
of agentic AI projects will be canceled by the end of 2027, according to Gartner: costs, unclear value or poorly managed risks (forecast).
Source: Gartner, June 25, 2025
What we measure
The Gartner, Cisco and MITRE frameworks all assess several dimensions: a company can have plenty of tools and still be immature. Here are the six questions we ask.
Do you know exactly where AI creates value in your business?
Is your data accessible, reliable and usable by AI?
Do your tools and systems let you deploy AI in your processes?
Do you have usage rules, an owner and risk management in place?
Do your teams know how to use AI, and are they ready to change how they work?
Do you measure what AI actually delivers?
The five levels
A framework inspired by the reference models. Each level leads to a concrete next step, not a score to show off.
A few individual uses, no shared rules.
Next step: Raise leadership awareness and set usage rules.
See our trainingIsolated trials, no priorities and no owner.
Next step: Pick two or three priority use cases.
Agentic AI trainingPilots underway and the first rules in place.
Next step: Move a use case into production with KPIs.
Get supportAI is embedded in several processes.
Next step: Strengthen governance, data and value measurement.
See our trainingAI is changing how the company operates.
Next step: Oversee autonomous agents at scale.
Talk to a consultantThe levels above are i2Business's own framework, built from the Gartner, MIT CISR and MITRE models. They are not those models' official labels.
What the frameworks say
Leading consulting firms and research centers reach the same conclusion: the gap is widening between companies that structure their approach and everyone else.
Stages 1 and 2 perform below their industry average, stages 3 and 4 above it.
13% of companies are “Pacesetters” (ready), 48% “Followers.”
In Tunisia
According to the ITCEQ survey (May 2026), here is the share of companies citing each obstacle.
For most companies, ITCEQ recommends a hybrid path: introduce AI while strengthening the digital foundation, with gradual investments.
Self-reported data. Source: WMC / TAP citing ITCEQ, June 3, 2026.
Our method
10 questions to position your company on the key dimensions.
See where you are strong and where AI is still fragile in your business.
A concrete action plan: where to start, and in what order.
European regulation
Since February 2, 2025, Article 4 of the EU AI Act has required providers and deployers of AI systems to train their staff. The text was softened in July 2026: the obligation is now to support the development of this literacy, without guaranteeing a specific level for each person. It can apply to companies outside the European Union whose AI outputs are used in the EU.
General information, not legal advice. Sources: White & Case · PwC.
FAQ
It is its ability to adopt AI and get measurable value from it, across several dimensions: strategy, data, technology, governance, skills and results.
By assessing each dimension based on facts (tools in production, written rules, measured results) rather than intentions. A short questionnaire is enough for an initial baseline.
It depends on the model: Gartner describes five, MIT CISR four, MITRE five. i2Business uses a five-level framework, from discovery to transformation.
According to ITCEQ (May 2026), 86% of companies consider advanced technologies essential to innovation, but 19% report having innovated. The barriers cited are financial resources, skills and resistance to change.
With a baseline assessment, then two or three priority use cases and employee training. For most companies, ITCEQ recommends a hybrid path: AI and the digital foundation in parallel.
Gartner cites rising costs, unclear business value and poorly managed risks, and predicts that over 40% of agentic AI projects will be canceled by the end of 2027. Measuring your maturity before investing reduces that risk.
Go further
An initial score and two priority areas to decide where to invest in AI, before you spend.