Norway is moving to keep generative AI out of its youngest classrooms. According to The Decoder, the government is largely banning AI tools in elementary schools and tightening their use in secondary schools, with the new rules taking effect when the school year starts in late August.
Prime Minister Jonas Gahr Stoere laid out the reasoning on Friday. “The most important thing in school is that our children learn to read, write, and do math,” he said, warning that “uncritical use of AI causes students to skip important learning steps.” Stoere pointed to a slide in learning outcomes since around 2015 and put part of the blame on smartphones, screens, and algorithms.
What the rules actually say
The policy sets different limits by age:
- Grades 1 through 7 (ages 6 to 13): students generally won’t be allowed to use AI at all.
- Lower secondary (ages 14 to 16): AI tools are permitted cautiously, under teacher supervision.
- Older students: the focus shifts to learning how to use AI properly.
There’s also a physical-materials angle. The Decoder reports that the government plans to pass a law requiring municipalities to provide physical teaching materials, which in practice means more printed books back in classrooms. Stoere said previous governments leaned too hard on digital media.
Part of a bigger pullback
This isn’t a one-off move. Norway has already banned smartphones in schools, handed teachers more authority in the classroom, and is planning a social media ban for children under 16. The AI restriction fits a consistent thread: pull screens back, put fundamentals first.
The concern isn’t pulled from thin air. As early as 2024, Swedish researchers studied the link between AI use and students’ ability to learn. Their findings showed both opportunities and risks, which is roughly the tension every education system is now trying to manage.
Why this matters
What stands out here is the age cutoff. Norway isn’t debating whether AI belongs in education at all. It’s drawing a hard line at the point where kids are still building the basic skills, reading, writing, arithmetic, that everything else sits on top of. The argument is simple: if a chatbot does the early reps for you, you never build the muscle.
That’s a meaningful signal for the AI industry. Education has been one of the loudest growth stories for tools like ChatGPT, and a national government just told the vendors that the under-13 market is off-limits for now. Edtech companies building for primary schools should expect more of this, not less.
The global split
Norway is part of a wider tightening, and The Decoder lays out where other countries stand:
- Japan issued guidelines back in 2023 urging special caution with children under 13 and treating AI-generated schoolwork as cheating.
- In the U.S., a 2024 court ruling said schools can penalize unauthorized AI use.
- UC Berkeley Law School will ban AI for nearly all graded assignments starting in summer 2026, allowing it only for research.
Others are running the opposite play. The United Arab Emirates is making AI a required subject from kindergarten through 12th grade starting in the 2025-26 school year. In Germany, the Conference of Ministers of Education wants AI woven into lessons and called an outright ban “unrealistic and untenable.”
So there’s no global consensus. One camp says protect the fundamentals first and add AI later. The other says fluency with these tools is itself a fundamental, and kids fall behind if you wait.
What to watch next
Norway’s late-August rollout will be an early real-world test of the protect-first approach. Watch two things: whether learning outcomes actually move, and whether other European education ministries follow Oslo or side with Berlin. Either way, governments are no longer treating classroom AI as an open question. They’re picking sides, and they’re doing it fast.
More details are available at the original report from The Decoder.