Prompt me Journalism: When Newsroom meets GenAI

The paper examines how generative AI reshapes journalistic labour by introducing prompting as
a new form of work and communication in public service media (PSM). As journalists increasingly
collaborate with language models, they are required to formulate precise prompts that define
tasks, specify tone, and anticipate machine logics. In doing so, they do not merely delegate work
but actively translate editorial intentions into machine-readable commands.
Given the current lack of insights into this journalist-machine collaboration, this paper explores
research question arises: How does prompting (re)define journalistic work and working
conditions in the PSM newsroom? Drawing on ethnographic observations, interviews, and
analysis of prompt archives from an ongoing case study at a European PSM organisation, the study
investigates how prompting is introduced, learned, embedded, and distributed in everyday
newsroom routines. For this purpose, both the themes and structures of prompts themselves and
the social processes of their creation are examined, through training seminars, prompt archives,
and affordances of working tools.
First findings suggest that prompting is introduced through bottom-up experimentation, framed
as a creative “playground” for innovation, while at the same time shaped by external pressures
pushing journalists into the role of the “human-in-the-loop”. In this emerging configuration,
prompting becomes both a site of professional empowerment as well as a reflection and a
mechanism through which AI infrastructures rearticulate the organisation of journalistic work.
Prompting consequently acts as a layer of algorithmic labour, redistributing expertise, control, and
creativity within the newsroom. It redefines work organisation through a continuous process of
prompt “optimisation” as an emerging form of algorithmic management where journalists must
constantly refine their communicative engagement with the system.