Robots in the News and Newsrooms: Unpacking Meta-Journalistic Discourse on the Use of Artificial Intelligence in Journalism

Abstract

As journalism has grappled with the potentials and boundaries of AI within the industry, journalists have produced plentiful articles detailing experimentation and potential consequences of AIdriven journalism (see, GPT-33, 2020). Accordingly, this article analyzes media coverage (N ¼ 95 articles) of AI in journalism over a 5-year period, starting in 2016 and ending in 2020, to examine prominent themes related to uses, roles, and concerns regarding AI in the newsroom. We sample coverage from 20 US and UK news media outlets representing a diversity of media with regards to media type and partisan leaning. We employ a thematic analysis on the media coverage of AI as it relates specifically to its use and application in journalism. Our exploration uncovers a tension between the industry and profession of journalism in highlighting the hopes and pitfalls of AI. It also allows for a discussion on assessing the place of AI in news making, especially with regard to the economic and contextual complexity in which news stories operate and the normative ideals of journalism in the digital era.

Without journalists, there is no journalism: the social dimension of generative artificial intelligence in the media

“…Las denominaciones han sido diversas, ya sea "periodismo escrito por máquinas" ( Van-Dalen, 2012), "periodismo algorítmico" (Anderson, 2012), "periodismo robotizado" (Clerwall, 2014) o, más comúnmente, "periodismo automatizado" (Graefe, 2016;Moran;Shaikh, 2022), dentro del marco más amplio que ofrecen las prácticas del periodismo computacional y del periodismo asistido por ordenador (Codina;Vállez, 2018;Parratt;Mayoral;Mera, 2021).…”

Fostering Artificial Intelligence to Face Misinformation: Discourses and Practices of Automated Fact-Checking in Brazil

“…But that outcome is fraught with tensions because actors outside the journalistic field have largely supported these technologies with demands foreign to fact-checkers' standards. As this research has shown, when journalists abdicate an active role in the development of AI products, they end up acquiescing to a conflict of interest between their professional needs and those coming from the tech industry, a trend that has also been discussed in recent work on automated journalism (Ford & Hutchinson, 2019;Moran & Shaikh, 2022;Simon, 2022).…”

“…As new robots have been put in motion by fact-checkers, even in countries struggling with an acute economic crisis (such as Brazil), it is critical to ask how professionals can keep their autonomy while facing automation demands. Although fact-checking has been linked with innovative trends in digital journalism toward machine-readable forms of writing (Graves & Anderson, 2020;Johnson, 2023), this does not imply that practitioners must embrace every new technology as inevitable to improve their daily work (Moran & Shaikh, 2022). National and regional fact-checking communities should engage in ongoing discussion (beyond global fact-checking annual meetings) regarding the pertinence of introducing new AI products into their workflows, weighing their pros and cons concerning the systems' training demands and the staff and skillset required.…”

“…In the fact-checking industry, calls to implement AI to automate processes underline how machinery can ease the workflow without undermining fact-checkers' authority as "truth-tellers" (Johnson, 2023). This optimistic outlook has mirrored the technologists' discourses that automation can "free up" reporters from routine tasks to pursue in-depth coverage (Moran & Shaikh, 2022;Simon, 2022).…”

“…Within that framework, AI has gained new momentum in the journalistic field (Moran & Shaikh, 2022). AI is usually understood as “the activity of computationally simulating human activities and skills in narrowly defined domains, most commonly through the application of machine learning approaches, a subfield of AI, in which machines learn from data or their own performance” (Simon, 2022, p. 2).…”