Only 14% of Icelandic companies have an AI policy in place. In the Nordic countries the figure is 48% (Staða gervigreindar á Íslandi 2026). That is a three-and-a-half-fold difference between us and the countries we most often compare ourselves with. To most managers this sounds like bad news. For you, it could be the best business opportunity of the decade.
The gap in numbers: Iceland versus the Nordic countries
Let us start with the facts. They come from recent Icelandic surveys, not foreign reports that may not apply here.
Icelandic companies with an AI policy
Statistics Iceland 2025Nordic companies with an AI policy
State of AI in Iceland 2026Professionals who use AI at work
Viska 2025Professionals who have received training from their employer
Viska 2025Notice the pattern in the table. Usage is already there: 80% of professionals use AI at work. What is missing is the structure: the policy, the training and the measurement. In other words, Icelandic companies are not behind in interest. They are behind in execution.
That means the gap is not a technical problem. It is a management task. And a management task can be solved in weeks, not years.
The government is waking up: the 2025 to 2027 action plan
The Icelandic government has published an AI action plan for the years 2025 to 2027. Among other things, it provides for skills development in the labour market, a central training portal and a new centre for AI and language technology. The government's own estimate in connection with the plan is that Iceland's gains from the use of AI could amount to as much as 500 billion ISK a year.
Put that figure in context. This is the government's own estimate of what is on the table if the Icelandic labour market gets to grips with the technology. Those gains will not be shared equally. They will flow to the companies that build the skills first.
It is also right to be realistic about the timeline. Public plans take time, and some of the measures have not yet been put into practice. A company that waits for the state to solve its training for it could be waiting a long time. A company that starts on its own now can then use the public resources as a supplement, not a prerequisite.
Why is the latecomer's advantage real?
It sounds paradoxical, but being behind has its advantages, if you act at the right time. Three reasons:
- The technology is more mature now. The companies that blazed the trail in 2023 and 2024 paid for experiments that did not work. You can start directly with the methods that research has confirmed deliver results: 25 to 56% faster completion of individual tasks among trained staff (Harvard/BCG 2023).
- Your competitors are probably behind too. When only 14% of companies have a policy, the odds are overwhelming that your competitors are in the same position as you. Whoever moves first in each industry sets the benchmark: faster quotes, shorter turnaround times, lower cost per unit.
- The advantage compounds. A department trained today finds new use cases every month. A six-month head start in skills becomes a two-year head start in processes. It is far harder to catch up on ways of working than to buy software.
In a small market like Iceland this is amplified even further. In many industries there are three to five serious competitors. The first company to get to grips with AI in its industry does not need to be the best in the world. It just needs to be the first in Iceland.
What do the companies that take the lead do?
The difference between the companies that succeed and the rest does not lie in the tools. It lies in three decisions that cost little money but a lot of focus.
First, they set a policy before usage gets out of hand. Which data may go where, which tools are approved, how results are verified. This is a document that can be finished in days, not months.
Second, they train a broad group, not just the technical people. Research shows that the biggest gains are precisely among those with the least experience, where a 43% improvement was measured (Harvard/BCG 2023). The advantage is created when the whole department works differently, not when two enthusiasts run experiments.
Third, they measure. A baseline before training, the same measurement after 30 days. Then you know whether the investment paid off, with your own numbers rather than statistics from foreign reports.
What you can do right now
- Compare your company with the numbers. Ask yourself three questions: Is there a written AI policy? Have staff received formal training? Is usage measured in any way? If the answer is no to two or more, you are in the 86% group, and that is the position you want to get out of before your competitors do.
- Map your competitors. Look at the websites, advertising and job listings of your three main competitors. Are they recruiting for AI skills? Do they talk about the technology publicly? If not, your window is open. If so, it is closing.
- Choose one department as a testing ground. Do not try to change the whole company at once. Pick one department with many repetitive tasks, for example finance, customer service or marketing, and set it the goal of finding three use cases in the next two weeks.
[ Get in touch ]
Book a free assessment
90 minutes that pay off immediately: we map your AI usage, risks and 3 to 5 automatable workflows, and deliver a report within a week. No commitment.