LORENA KAPXHIU
DOI: 10.70827/REYY5015/Iamh
Abstract
Hybrid foreign language education might be radically transformed by using artificial intelligence (AI) in teaching and testing. Online platforms came to be central to learning during the COVID-19 pandemic as they allowed for flexible schedules for both teachers and students. Nevertheless, the use of AI in the assessment process has raised concerns about the reliability and validity of the results. Comparisons between performanceintraditionalexamsand AI-driven evaluations revealed notableinconsistencies, raising concerns about the effectiveness of AI inhigh-stake assessments. This study involved 30 foreign language students who transitioned from classroom instruction to six months of fully online AI-supported classes. The research combined various methods such as AI-based adaptive assessments, student portfolios, surveys and results from external examinations to compare the effectiveness of automated assessments with traditional ones. The findings show that AI-based systems are useful in providing personalized feedback and promoting higher student engagement. However, these systems face difficulties in evaluating complex skills such as speaking or writing. Combining automated assessments with teacher-led evaluations results in a more balanced and reliable approach. This study suggests the need for integrated assessment models in language education that merge technology and traditional pedagogy in the interest of more equitable and inclusive education.
Keywords: artificial intelligence, combined assessment, hybrid education, language assessment, online exams, personalized learning.