Estonia is currently considering the integration of artificial intelligence (AI) in the assessment of the native language final exams, an initiative being explored by the Estonian Education and Youth Authority (Harno) in collaboration with researchers from Tallinn University. This proposal evolves from a recent research study that revealed that AI-driven language models can provide grades closely resembling those awarded by human evaluators. Brought to light by lecturer Merilin Aruvee, who specializes in Estonian language teaching and applied linguistics, this examination of AI in education highlights a significant technological advancement in assessment methodologies.

Aruvee, alongside doctoral candidate Katarin Leppik, developed new evaluation criteria for the writing component of the 9th-grade final exam after engaging in e-exam trials for some time. This effort led them to question the feasibility of incorporating AI into the assessment process. To refine their criteria, they sought feedback from experienced language teachers, which helped adapt their grading model to align more closely with educational expectations. This exchange underscores the collaboration between academic research and practitioners, aiming to improve educational standards in Estonia.

If implemented, utilizing AI for exam evaluations could streamline the assessment process and ensure a more objective grading system. This approach may also relieve teachers from assessment duties, allowing them to focus on instructional quality. As Estonia contemplates this innovation, it reflects a broader trend in education, where AI is increasingly seen as a tool to enhance efficiency and effectiveness in evaluation while raising questions about the balance between technology and human oversight in academic settings.