언론인들의 인공지능(AI) 기술 수용에 영향을 미치는 요인 = Determinants of Journalists’ Acceptance of Artificial Intelligence

국문 초록 (Abstract)

이 연구는 국내외 저널리즘 분야에서 인공지능(AI) 기술 도입이 빠르게 확대되는 일련의 상황들을 고려하여, 언론인들의 인공지능 기술 수용에 영향을 미치는 핵심 요인들을 파악하고자 시도했다. 특히, 방법론적으로는 언론인들의 인공지능 수용 과정에서 결정적인 역할을 하는 주요 선행변인들과 기술수용모형(technology acceptance model [TAM])의 인지된 유용성·용이성을 내생변인으로 고려하여 구조방정식모델링(structural equation modeling [SEM])을 통해 모델 적합도 검증(model fitting test)을 수행했다. 주요 분석데이터는 전문조사업체를 통해 실시한 설문조사에서 성실하게 응답한 언론인 300명을 최종 분석대상으로 삼았다. 분석 결과, 내생변인 중 ‘인지된 유용성’이 ‘AI 사용 의도’에 가장 큰 영향을 미쳤으며, 언론인이 주요 연구 대상인 이유로 선행변인 중 ‘저널리즘 품질개선 기대’가 ‘인지된 유용성’에 가장 큰 영향력을 보이는 것으로 확인되었다. 또다른 내생변인 중 ‘인지된 용이성’에 가장 큰 영향을 미치는 선행변인은 ‘인공지능 리터러시’인 것으로 확인되었다. 한편, 선행변인으로 투입한 ‘인공지능 효능감’은 분석 결과에서 가설과는 달리 ‘인지된 유용성’에 부적인 영향(-)을 미쳤는데, 이는 조직적 차원의 지원체계가 적절히 갖추어지지 않은 현재의 상황에서 인공지능 도입이 오히려 언론인들의 업무 과정에 불편을 초래하는 기술로 인식될 수도 있음을 시사한다. 이 연구의 구조모형분석에서 모든 적합도 지표들은 권장 기준을 충족하여 연구모델의 타당성이 충분히 확보되었고, 가설 9개 중에서 8개가 지지되었음을 확인하였다. 이러한 결과를 고려할 때, 언론사들은 AI 기술 도입 활성화를 위해 향후 ‘언론인들의 리터러시 교육 확대 및 성공적 활용 사례 공유, 조직적 차원의 지원체계 구축 확보 등’ 다양한 노력을 실천해야 할 필요가 있다.

다국어 초록 (Multilingual Abstract)

This study investigated the determinants of journalists’ acceptance of artificial intelligence (AI) technology, reflecting the rapid and widespread integration of AI tools into both domestic and international journalism practices. As news organizations increasingly apply AI to reporting, editing, and distribution, it is critical to identify the drivers of acceptance and the barriers to use for both theoretical advancement and professional practice. Drawing upon the Technology Acceptance Model (TAM), this study conceptualized perceived usefulness and perceived ease of use as core endogenous variables and incorporates theoretically grounded antecedents. Based on prior theory and a comprehensive literature review, the antecedents include perceived AI quality, expectations for improved journalism quality, AI literacy, and AI efficacy, which are modeled as exogenous constructs. While journalists’ acceptance of AI technology can be influenced by various contextual factors, this study focuses on the most salient antecedent and endogenous variables aligned with its research objectives. The proposed research model was then tested using structural equation modeling (SEM) to evaluate overall model fit and to validate the hypothesized relationships among the constructs. The empirical data were collected through a professional survey of 300 journalists who provided valid responses, thereby ensuring both reliability and representativeness. The analysis confirmed that perceived usefulness exerted the strongest influence on the intention to use AI, underscoring its centrality in shaping adoption behaviors. Among the antecedent variables, expectations for improved journalism quality emerged as the most significant predictor of perceived usefulness, reflecting the belief that AI can enhance journalistic output and audience value. AI literacy was identified as the primary determinant of perceived ease of use, highlighting the importance of journalists’ technical competencies in lowering barriers to adoption. Contrary to the original hypothesis, AI efficacy demonstrated a negative effect on perceived usefulness. This unexpected result suggests that in contexts where organizational support systems remain underdeveloped, journalists may interpret AI not as a resource but as a perceived burden that disrupts established workflows. The SEM exhibits satisfactory fit across multiple indices, and eight of nine hypotheses are supported. The results carry both theoretical and practical implications. Theoretically, this study extends the TAM framework by integrating journalism-specific antecedents, thereby contributing to a more contextually grounded understanding of technology adoption in newsrooms. Practically, the findings highlight the urgent need for news organizations to cultivate conditions conducive to AI adoption. Recommended strategies include expanding literacy training tailored to journalists, systematically disseminating best practices and successful applications, and constructing organizational support infrastructures that reduce perceived risks while enhancing perceived benefits. Through such efforts, journalism can leverage AI not merely as a technological innovation but as a strategic resource for enhancing efficiency, strengthening credibility, and improving the overall quality of news production.