Las week Linn delivered her master’s thesis, which is a qualitative case study of a major Norwegian news agency’s use of AI in news production.
Linn’s interests were related to responsible use of AI, and in particular GenAI. Through interviews and observations, she has developed an empirically grounded model of AI use in journalism that takes into account different organizational levels and the nature of practices involved. This is a contribution to the growing body of research on innovative applications of AI and its benefits and challenges.
- Thesis title: “I’m Simply Very Afraid That We Will Become Stupid, and Less Skilled Journalists” –Responsible Use of Generative AI in a Norwegian Newsroom.
- Author: Linn Emilie Kalleberg.
- Abstract: This master’s thesis explores how generative AI tools are integrated into journalistic work practices, how editorial control is maintained, and how this integration affects journalists’ professional identity. The study is conducted as a qualitative single-case study at one of Norway’s largest national news organizations, based on semi-structured interviews with five journalists, two in-house AI developers, and an AI lead journalist, supplemented by documents, field notes, and selected chat interactions with AI tools. The findings show that journalists use generative AI tools across four dimensions, ranging from the automation of routine tasks and research support, through document summarization and large-scale data analysis, to using AI as a creative collaborator. Journalists experience a lack of control over generative AI tools and their output, but maintain editorial control over what is published through manual verification and quality assurance of the content. Responsible use is navigated through professional norms rather than formal guidelines, which are largely unknown or rarely used. The journalistic role is shifting, with greater emphasis on human qualities that AI cannot replace and a growing expectation that AI competence is part of the journalistic professional identity. The thesis argues that responsible use of generative AI in the news industry must be understood through a two-dimensional model that maps the dimensions of generative AI use against the scope of responsible use. The model shows that the need for responsible use increases as AI involvement moves from the automation of routine tasks toward the creative process of journalistic work.