Abstract
<jats:title>Abstract</jats:title> <jats:p>Generative artificial intelligence (GenAI) refers to artificial intelligence systems that can create or generate content, such as text, code, images, video, and audio. Recent advances in transformer‐based and diffusion models have significantly enhanced the capabilities of these systems (e.g., ChatGPT and Gemini), enabling them to generate human‐like text, produce high‐quality visuals, and engage in naturalistic conversations. These innovations have sparked growing interest among language assessment professionals in exploring how GenAI can be integrated across various stages of the assessment process, including test development, administration, and the evaluation of test‐taker performance.</jats:p> <jats:p>This entry provides an overview of the potential applications of GenAI in language assessment, organized by its primary output modalities: text, image, audio, and video. In doing so, it highlights how GenAI can support feedback generation, automated scoring, item and stimulus creation, and interactive speaking tasks. At the same time, the discussion addresses key challenges related to accuracy, transparency, fairness, data security, and construct validity. As GenAI tools become increasingly multimodal and sophisticated, their potential to enhance the efficiency, accessibility, and authenticity of language assessments continues to grow. However, the realization of this potential requires thoughtful implementation, human oversight, and ongoing empirical research to ensure GenAI is used in ways that are ethical and adequate for the intended interpretations and uses of language tests.</jats:p>