Students using AI to cheat will face stricter rules, says Danish government
The Danish government has announced stricter measures to combat AI-assisted cheating in upper secondary schools, requiring students to give an oral defence of essays written at home and mandating that schools monitor computer use during written exams. Education minister Magnus Heunicke said urgent action was needed, describing AI cheating as a growing problem undermining students' ability to think for themselves, and the changes take effect immediately as pupils return from the summer break.
Around 9,000 students who write the major "større skriftlige opgave" assignment each year will now have to defend it orally, while schools must install monitoring tools and firewalls to control computer use during exams. Schools are also being encouraged to have pupils complete more written work under supervised conditions rather than at home. The move follows a similar reckoning in Sweden, where suspensions and warnings for unauthorised AI use reportedly rose by 688% between 2023 and 2025, prompting a government review due to report in April.
- Denmark orders schools to tackle AI cheating with immediate effect.
- Students must now orally defend essays written at home.
- Schools must monitor computers and install firewalls during exams.
Both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
Academic integrity requires students to develop their own critical thinking and writing abilities; allowing AI to write essays undermines this core purpose of education and leaves students unprepared for independent work. Without intervention, the practice will proliferate unchecked, creating an unfair system where qualifications no longer reflect genuine capability and undermining educational standards across the system.
The case against
Rather than relying on surveillance and prohibition, schools should teach students to use AI tools responsibly and ethically—skills they'll need throughout their careers—distinguishing between legitimate assistance and cheating through education rather than enforcement. Overly restrictive rules ignore the reality that AI literacy is now essential and risk fostering distrust, whereas integrated frameworks for responsible technology use would better prepare students for a world where these tools are ubiquitous.