Transformasi Asesmen Matematika pada Era Kecerdasan Buatan: Sebuah Systematic Literature Review tentang Praktik Asesmen, Umpan Balik, Learning Analytics, dan Arah Penelitian Masa Depan

Lisda Ramdhani (Universitas Pendidikan Ganesha)
I Gusti Lanang Agung Parwata (Universitas Pendidikan Ganesha)
I Gede Ratnaya (Universitas Pendidikan Ganesha)
I Dewa Ayu Made Budhyani (Universitas Pendidikan Ganesha)

Mandalika Mathematics and Educations Journal
Vol 8 No 3 (2026): Edisi September

The rapid advancement of Artificial Intelligence (AI) has significantly transformed mathematics assessment by enabling adaptive assessment, automated feedback, learning analytics, and data-driven instructional decision-making. This study aimed to synthesize current evidence regarding AI-based mathematics assessment, feedback practices, learning analytics, and future research directions. A Systematic Literature Review (SLR) was conducted following the PRISMA 2020 guidelines using the Scopus database. The search covered publications from 2015 to 2026 with the query related to mathematics education, generative AI, artificial intelligence, assessment, and feedback. After applying inclusion and exclusion criteria, 33 articles were selected for analysis. The findings indicate that AI has shifted mathematics assessment from traditional summative evaluation toward adaptive, formative, and personalized assessment. AI-generated feedback and learning analytics improve students' problem-solving skills, conceptual understanding, and teachers' instructional decisions. Nevertheless, issues concerning validity, ethical use, data privacy, and teachers' AI literacy remain major challenges. Future studies should develop ethical, explainable, and human-centered AI assessment frameworks that strengthen mathematics learning while preserving pedagogical quality.