Get the Ultimate Guide
to Applying for a Master's
Download for free
Download Your Master's Admissions Guide
From August 2026, the EU AI Act will classify AI systems used for admissions, grading, and exam monitoring as “high risk”. Universities will need to demonstrate transparency, human oversight, and staff AI literacy to remain compliant. Find out what this means for institutions, researchers, and master’s students across Europe.
Europe’s AI laws are changing. As of August 2, 2026, the EU AI Act’s latest changes are set to take effect. It will affect various industries, including the higher education and research sectors. In this article, we will explore what exactly these new AI rules will mean for universities, researchers, and students across Europe.
The EU AI Act is the world’s first comprehensive legal framework on AI. First implemented on 1 August 2024, the framework implements rules on AI use to foster trustworthy AI within Europe. The policies cover AI innovation, risks, and investments. With AI changing rapidly, this AI legislation aims to offer users protection from AI’s risks and challenges.
This risk-based system aims to create clarity around AI usage for AI developers, deployers, and users. With the EU’s AI timeline beginning in April 2018, the AI laws have been continuously changing and developing alongside the evolution of AI.
Ultimately, the AI Act is intended to promote AI research and innovation without compromising safety and trustworthiness.
As of August 2026, the rules regarding the “high risk” level will take effect. This will directly impact universities. This is because the “high risk” level includes any AI systems used for admissions decisions.
For universities and research institutions, the most relevant section is Category 3, which covers “education and vocational training”. This includes assessment tools. For example, any AI systems that evaluate learning outcomes, automate scoring, or assess student performance.
AI specialist and educator Daria Gavrilova shared a comprehensive LinkedIn post that clearly explains what universities can expect. She shared that the high-risk level includes using AI for “evaluating learning outcomes, for assessing what level of education a person can access, and for detecting cheating during exams.”
Moreover, Daria says that the AI literacy obligation will be paramount for universities. This means schools must ensure staff can evaluate output, recognize AI’s limits, and understand how to exercise human oversight. If any school implements AI tools without proper training, this is considered a ‘compliance gap’.
There are also new laws about AI transparency. As of August 2, 2026, AI systems and AI-generated content must be easily identifiable and clearly labelled. For universities, this could include everything from their website chatbots to a professor’s AI-generated course materials. Additionally, all university staff must understand how AI systems are being used and be able to clearly inform students about the methodologies behind the systems.
Article 14 mandates that any AI tools and processes must be overseen by a human. Institutions must include human checks in any AI workflows. This means humans must be reviewing and approving any AI output. For example, AI grades or assessments. AI-automated assessments or grading are not valid without human approval. This also means that AI use must be logged and documented at universities.
Failure to adhere to these new rules can result in heavy fines.
The AI Act primarily affects AI providers and deployers. According to KPMG Law, the AI Act excludes AI use that is specifically for scientific research or testing. Scientific research is exempt from these regulations insofar as it is not on the market, not commercially viable, and not used by external users. This only includes research that does not involve real-world testing. This exemption aims to ensure scientific and research freedom.
Ultimately, higher education institutions have ‘institutional responsibility’. However, staff, researchers, and students have ‘individual responsibility’. If they are using AI tools or products, they should have a sufficient individual level of knowledge to avoid potential risks.
Daria suggests that university leadership start with three simple questions:
If nobody can answer these questions, that silence is itself a warning sign. It is an important sign that changes need to take place to ensure compliance with the new AI Act rules. EU-based universities will need to have correct governance processes in place and keep up to date with potential regulatory trends to avoid future complications.
Students applying to European universities can expect clearer communication about how their university is implementing AI in the classroom and in grading. By understanding the EU AI Act, they know which protocols their university is bound by and the rights they have concerning AI.
It is also important for students to understand their own ‘individual responsibility’ when it comes to AI use. In addition, each university has its own individual AI policy on what type of AI is or isn’t allowed in the classroom or exam room, which students need to adhere to.
These stringent new AI regulations send a clear message that clarity around AI is necessary and mandatory. Universities across Europe will be held to the same AI laws and regulations. This means that universities need to prioritize AI regulation and cement their AI policies. AI staff training and clear university-wide AI policies will be paramount to ensure smooth AI use across the entire university ecosystem.
The universities that are investing in AI staff training, finalizing their AI policy statements, and ensuring transparency throughout their communities will be the ones that avoid hefty fines and come out ahead of the AI revolution.
The European Union offers digital guidelines on AI and information for educators.
Stay ahead of industry trends with insights from master’s students around the world.
Artificial Intelligence is redefining higher education worldwide. From AI-powered tutors to personalized learning paths, business schools are embracing this revolution. But can AI truly replace the mentorship, networking, and leadership development of traditional education? Explore how AI is transforming postgraduate business education, and why the future lies in partnership, not replacement.
