Application of AIGC in multilingual teaching in higher education

This study systematically evaluates the technological realization path and social benefits of generative artificial intelligence (AIGC) in multilingual teaching and learning in higher education. Based on current advanced multilingual model architectures and cross-modal content generation technologies, an automatic generation system for teaching resources supporting 12 languages is constructed. Experimental data showed that the system led to a 41.2% increase in learning efficiency in a Tamil course at the Indian Institute of Technology (IIT) (p<0.01, n=320), while a test of teaching Chinese as a second language at the Beijing Language and Culture University (BLCU) showed that AIGC-generated personalized exercises reduced the learner error rate by 33.7%. The study further revealed that (i) the cost of generating instructional materials for low-resource languages (e.g., Kiswahili) decreased to 18.5% of the traditional approach; and (ii) through an authoritative equity assessment framework, it was confirmed that the technological solution could narrow the digital education divide between developing and developed countries by up to 27.3 percentage points. This provides actionable technological support for the equity goals of the UN's Education 2030 Agenda.