Newsrooms are using AI to speed up journalism — but avoiding the business-model gamble
Cartoon · AI-generated
A study of 160 AI projects finds news organisations are automating production far faster than they are experimenting with subscriptions, revenue or fundraising.
A major new analysis of AI adoption in newsrooms suggests that journalism organisations are moving cautiously around one of the technology’s most consequential challenges: how to make money from it. The News + AI Field Report, produced by Djordje Padejski for the Walter Cronkite School of Journalism and Mass Communication in conjunction with Arizona State University’s Knight Center for the Future of News, examines 160 AI projects launched by 101 news organisations between late 2022 and the end of 2025. Its central finding is strikingly consistent: newsrooms have been quicker to use AI to make existing work faster than to rethink the economics of their business. According to the report, most projects are concentrated on routine production tasks such as summarisation, drafting, transcription, translation and workflow automation.
That pattern matters because the report also finds a persistent gap between anxiety and action. Newsrooms interviewed for the study expressed deep concern about AI’s potential to disrupt site traffic, audience behaviour and the broader news business model, yet comparatively few projects were aimed at revenue, subscriptions or fundraising. The report’s categorisation of projects shows that business-focused work made up only a minority of the overall total, even though those initiatives were among the areas where many organisations say they feel the most pressure to adapt.
One of the more revealing details is that the field appears to be unevenly distributed. The report indicates that every fundraising project, and more than half of the business-related projects, came from newsrooms participating in the American Journalism Project’s AI programmes. Remove those grantees from the sample and the number of business experiments drops sharply. That finding suggests that outside support may be doing more than funding individual projects: it may be supplying the scaffolding, accountability and encouragement that many newsrooms need before they are willing to test bolder ideas.
The report’s interactive database, which maps the projects by category and use case, underlines how narrow much of the industry’s current experimentation remains. Newsroom-facing chatbots, for example, have become one of the more visible expressions of AI adoption, but they do not necessarily solve a core reader problem or address the commercial weaknesses that preoccupy publishers. The Field Report suggests that the dominant instinct has been to apply AI where it fits most easily into existing workflows, rather than where it might force a more fundamental rethink of products, audiences and revenue.
Taken together, the findings point to a broader structural issue in journalism innovation. Product and technology teams may be well placed to run experiments, but they do not always control the business model decisions those experiments would need to influence. The report therefore raises a question that goes well beyond AI itself: whether newsrooms have the incentives, cross-team collaboration and external support needed to experiment with the future of their business, not just the mechanics of their reporting.
Inspired by headline at: [1]
Sources by paragraph: - Paragraph 1: [2], [3], [4] - Paragraph 2: [2], [5] - Paragraph 3: [2], [6] - Paragraph 4: [1], [4], [7] - Paragraph 5: [1], [3], [6]
Source: Noah Wire Services · Original