The AI competence gap grows where experience is not built

Last week we wrote in AI Value Lab Oslo, building on Erlend Rosseland Stokke's point in digi.no, about an AI shortage that may prove more demanding than many see today.

Not a shortage of technology.
Not a shortage of tools.
👉 But a shortage of people and environments that have made AI work in practice.

Stokke points to a self-reinforcing cycle. Organisations remain stuck in pilot mode because they lack people with real experience. At the same time, experience is built only when someone gets to take AI from pilot to production.

At Digital Norway's tech breakfast on Tuesday, on AI adoption in Equinor and Telenor, I heard the same point confirmed from another angle.

In reality, many organisations follow one of two paths.

🔁 One is pilot mode. Many discussions. Heavy tool focus. New tests. But limited movement. You learn something, but not enough to build confidence, execution capability and practical maturity.

📈 The other is building competence through real use cases. Then learning moves into workflows, customer experience, data, governance, collaboration and delivery. Competence is not just something you talk about. It is built through use, adjustment and responsible implementation.

It was striking how closely this was linked to commercial value.

Telenor highlighted attractive AI areas within:
💥 B2B sales and go-to-market
💥 Customer growth and retention
💥 Customer service and follow-up

These are not peripheral AI areas, but workflows with high feasibility and high value. They are at the core of commercial processes, customer journeys and value creation.

For me this aligns well with my commercial leadership experience and what I explore through practical AI projects.

Another strong point from Telenor was that AI should not simply be layered onto existing workflows. Then gains are often incremental. When workflows are redesigned around AI, gains can become structural.

Telenor showed a simple but precise formula for AI value:

Annual net value = Baseline × AI delta × Adoption × Confidence − Build & run cost

It shifts the discussion from technological potential to operational realism. Adoption, confidence, quality and the cost of building, operating and scaling.

Equinor showed at the same time the importance of structure, sharing, learning and responsible use.

Real AI competence is built when people work concretely with needs, user experience, data, risk, accountability, workflow and business value.

That is also why we established AI Value Lab Oslo, where we build practical understanding of how AI can create real value.

The question is not only whether organisations use AI. The question is which path they are actually building competence in. 💯

#AICompetence #CommercialLeadership #DigitalTransformation #PracticalAI