It has become much easier to buy a licence, open a conversation and get help from AI. It is still difficult to turn the time saved into better services, higher revenue or lower costs across the organisation.

The adoption is real

In a survey conducted by Samfunnsøkonomisk Analyse for NHO, 55 per cent of Norwegian organisations said they used AI in 2025, up from 24 per cent in 2023. Official statistics from Statistics Norway also show that three in ten enterprises with at least ten employees used one or more AI technologies in 2025. The figures do not measure exactly the same thing, but they point in the same direction: use is increasing rapidly.

The large majority have nevertheless not reached scaled transformation. SØA/NHO classifies 19 per cent as frontrunners and 17 per cent as advanced users. More than half are still explorers. AI is being tested, but it has rarely become a systematic part of how work actually flows.

Individuals feel the benefit first

It is logical that personal benefits arrive before organisational ones. An employee can produce a better first draft, analyse faster or save an hour in the very first week. In Microsoft’s global 2026 Work Trend Index, 66 per cent of AI users say they have more time for high-value work, and 58 per cent say they produce work they could not have delivered a year earlier.

But time saved does not automatically mean costs saved. McKinsey’s research shows that the time often goes to new tasks or more of the work the employee already had. For the individual, the result may be better quality, less backlog or a slightly calmer working day. All of this has value. It simply does not necessarily appear in the accounts.

“In many cases, people are ready. The systems around them are not.”

Microsoft Work Trend Index 2026

This is the value gap: the employee changes practice while the organisation continues to plan, measure, staff and reward as before. Microsoft finds that culture, leadership support and people practices have more than twice the correlation with reported AI impact as individual effort alone.

The technology can be adopted in a day. A new way of working must be learned, embedded and governed.

What distinguishes those who succeed?

The most consistent finding across Norwegian and international studies is not a particular model or platform. It is that the winners approach implementation differently. McKinsey finds that organisations with the greatest documented impact are almost three times as likely to have redesigned workflows. They also have clearer leadership ownership and are more likely to scale what works.

SØA/NHO finds the same pattern in Norway: frontrunners report the greatest productivity effects, and value arises primarily when AI is integrated into work processes – not when the technology is merely tested or used inconsistently.

What is holding Norwegian organisations back?

The most common barriers are less dramatic than the debate may suggest. Among organisations that do not use AI, the main issues are a lack of insight into which problems the technology can solve, uncertainty about benefits, and insufficient capability to understand, test and use it. As organisations mature, concern shifts towards data security, privacy and regulation.

Time is a real part of the explanation. Most people still have to serve customers, invoice, manage operations and keep the working day on track. Transformation competes with today’s production. That makes leadership prioritisation decisive: without allocated time, learning becomes a spare-time activity for the most interested employees.

Is the public sector in the lead?

The answer is both yes and no. The public sector has scale, data, strong professional communities and considerable capacity to build frameworks. Well over half of public-sector organisations now report using AI, and many report faster task completion, freed-up time and better quality. The OECD also notes that Norway has established a strategy, governance and a broad portfolio of use cases.

But documented value also lags here. Digdir shows that only one in four public-sector organisations has converted efficiency gains into reduced costs or resource use. The Office of the Auditor General of Norway has also identified shortcomings in coordination, data, infrastructure, legal clarification and capability. The public sector is therefore sometimes ahead in experimentation and responsible frameworks – but not necessarily in realising value.

It is good that the public sector leads where trust, language, shared infrastructure and essential public services are at stake. But it should also be a demanding first customer for Norwegian technology companies. The government’s own AI strategy recognises that young, innovative companies often struggle to win public contracts. A state that develops everything itself may end up weakening the market it wants to build.

From individual benefit to organisational value

Every transformation requires effort, and AI is no exception. Personal returns can appear within days or weeks. Organisational returns normally take longer because capability, processes, data, accountability and measurement must work together.

This does not mean the organisation should wait for a perfect strategy. It should start smaller and more specifically.

1. Choose one workflow with clear friction.Describe current time use, quality, waiting, errors and control requirements before choosing a solution.
2. Give the value an owner.Someone must be accountable for the outcome, the change in ways of working and how released capacity will be used.
3. Build capability through real work.Short courses are useful, but learning sticks when the team uses AI on its own tasks, sources and decisions.
4. Test simply and measure soon.A first version should generate learning. Compare it with the baseline and decide whether to improve, scale or stop.
5. Make the learning an organisational asset.Document effective ways of working, quality requirements, roles and controls. The value then becomes more than a private benefit for a few champions.

This is also at the heart of Naviqon’s approach: start with the work. Develop leadership, capability and technology around a prioritised workflow. Measure what matters and build on what the organisation actually learns.