KJCKorea
Journal Central

This study explores gentrification beyond physical displacement and examines it as a new dimension of linguistic and informational inequality. Focusing on Seosulla-gil in Seoul, it compares the amount and accessibility of information in offline (storefront signs) and online (Instagram posts) linguistic landscapes. In the offline sphere, policy restrictions on signboard size have reduced information beyond store names, turning informational absence into aesthetic symbolism. Conversely, online platforms provide extensive contextual information—brand stories, philosophies, and community activities—creating an informational gap between the two spheres. However, this expansion presupposes digital access and excludes groups that lack digital literacy from fully engaging with such information. To mitigate this exclusion, this study proposes two complementary guarantees: the guarantee of digital accessibility, emphasizing institutional responsibility for digital literacy education, and the guarantee of informational accessibility, encouraging shop owners’ voluntary information disclosure and language policies based on the right to access information. Together, these dual guarantees suggest that the exclusion and marginalization caused by gentrification can be transformed into the restoration of publicity and inclusivity.

This study examines how Large Language Models (LLMs) recognize and refuse unethical directive speech acts by analyzing their responses to indirect and evasive user requests. Based on the Cross-Cultural Speech Act Realization Project (CCSARP), directive prompts were constructed by varying degrees of indirectness to evaluate the models’ pragmatic inference abilities. The study was conducted in two stages. First, a high rate of information leakage was observed for indirect directives using ChatGPT-4o (February 2025 version). Second, newer models—GPT-5, Claude Sonnet 3.7 and 4, and Gemini 2.5 Flash—were tested across four categories of unethical directives through multiturn dialogues. Logistic regression with Benjamini–Hochberg FDR correction revealed that although newer models displayed improved refusal performance overall, they remained vulnerable to highly indirect and non-conventional directives, particularly those related to discrimination and harmful behaviors. These results suggest that current AI safety systems rely heavily on surface-level keyword filtering, indicating the need for models to better learn diverse directive strategies and expressions in Korean. Moving beyond technology-centered safety evaluation, this study experimentally analyzes AI pragmatic response mechanisms and proposes directions for fostering ethical communication in future human–AI interactions.

This study investigates the language attitudes of English-track international students at a Korean university operating under an English-medium instruction (EMI) system. Survey data from 195 students were analyzed quantitatively and qualitatively to assess attitudes toward English (L2) and Korean (L3) across cognitive, affective, and behavioral domains. The finding shows that the students regarded both languages as important, but attributed different functional roles to each: English operated as academic symbolic capital, whereas Korean was essential for daily life. English consistently received high scores across all domains, whereas Korean showed a marked discrepancy between strong cognitive recognition and lower affective and behavioral attitudes. Furthermore, attitudes toward Korean declined with longer residence (p < .01), likely reflecting limited institutional support, discontinuities in Korean language education, and broader tensions arising from conflicting language policies. These results suggest that EMI programs should incorporate plurilingual pedagogical approaches that support the flexible and integrated use of students' linguistic resources.