Over the last decade, AI has increasingly been adopted by newsrooms in the form of different tools aiming to support journalists and augment the capabilities of the profession. The main idea behind the adoption of AI is that it can make journalists’ work more efficient, freeing them up from some repetitive or routine tasks while enhancing their research and storytelling techniques. Against this idea, and drawing on the concept of “hype”, we employ a critical reflection on the lens often used to talk about journalism and AI. We suggest that the severe sustainability crisis of journalism, rooted in growing pressure from platforms and major corporate competitors, changing news consumption habits and rituals and the growing technologization of news media, leads to the obsessive pursuit of technology in the absence of clear and research-informed strategies which cater to journalism’s civic role. As AI is changing and (re)shaping norms and practices associated with news making, many questions and debates are raised pertaining to the quality and plurality of outputs created by AI. Given the disproportionate attention paid to technological innovation with little interpretation, the present article explores how AI is impacting journalism. Additionally, using the political economy framework, we analyze the fundamental issues and challenges journalism is faced with in terms of both practices and professional sustainability. In the process, we untangle the AI hype and attempt to shed light on how AI can help journalism regain its civic role. We argue that despite the advantages AI provides to journalism, we should avoid the “shiny things perspective”, which tends to emphasize productivity and profitability, and rather focus on the constructive synergy of humans and machines to achieve the six or seven things journalism can do for democracy. Otherwise, we are heading toward “alien intelligence” which is agnostic to the core normative values of journalism.
Stored only when both papers are in the corpus — 13 citations exist upstream.
Imported from OpenAlex (CC0) — nothing here is generated.