In December 2024 The Beacon joined 27 other nonprofit newsrooms in the American Journalism Project's Product & AI Studio, embarking on a year-long experiment to see how large language models and other AI tools might fit into local newsroom workflows without supplanting journalists. According to The Beacon, the newsroom entered the cohort already working on an AI policy with The Trust Project and with staff informally testing AI for tasks such as extracting information from PDFs and images. [1][3]

The newsroom entered 2025 with a specific editorial aim: to use LLMs to turn public meeting transcripts and civic records from the Kansas and Missouri statehouses into investigative story leads. That ambition, The Beacon reports, encountered practical limits when tried against the reality of committee hearings and complex public-records formats; many of the tools tested were “not quite ready” for the scale and nuance of statehouse coverage. The Beacon says the experiment forced a timely pivot toward more modest, but reliable, uses of the technology. [1]

Working with AJP coach Justen Fox, The Beacon trialled platforms including Everlaw, ChatGPT Enterprise and LocalLens. The newsroom concluded that LLMs were most valuable when deployed to accelerate discrete steps inside larger reporting projects rather than to produce finished reporting. The Beacon describes examples where reporters used AI to speed source-finding, fact-check previous meeting coverage and surface themes in large datasets. The cohort ended in December with virtual showcases organised by AJP where The Beacon presented its findings. [1]

The Product & AI Studio itself was launched by the American Journalism Project in 2023 to help local newsrooms pilot AI-driven tools, backed by support from OpenAI and the Patrick J. McGovern Foundation. The programme has provided grants, coaching and a collaborative forum to test tools, share lessons and surface best practices for responsible AI use in journalism. In 2024 the studio awarded $1.4 million in grants to 28 news organisations to pursue projects ranging from civic-data extraction to revenue experiments. According to AJP materials, the initiative emphasises both product development and ethical guardrails such as clear AI usage policies. [2][4][7]

The Beacon’s practical takeaways align with broader patterns emerging from other cohorts and experiments: AI can free journalists from repetitive work and help with audience or fundraising efforts, but success depends on careful oversight and targeted application. Industry examples show AI-assisted fundraising and membership initiatives produced time savings and higher conversion rates for some local outlets, and AJP-supported cohorts have since expanded to include focused work on membership and revenue operations. The Beacon framed its own use of AI as operational augmentation rather than automated story production. [5][6][1]

The newsroom also used the cohort to cultivate a testing mindset: developing custom GPTs to generate news-quiz questions, produce a health "brief" summarising Missouri headlines for reporters, and identify patterns in large datasets. The Beacon stresses that these prototypes are intended to enhance workflows and reporting capacity, not to write articles wholesale, and that the newsroom will continue refining its AI policy and practices in collaboration with peers and coaches. According to The Beacon, coach Justen Fox compiled a guide for cohort participants outlining pros and cons of the tools trialled, providing a resource for other newsrooms weighing similar choices. [1][2][7]

The Beacon’s experience underscores a pragmatic middle path for local journalism: adopt AI where it demonstrably reduces repetitive labour and surfaces leads, maintain editorial control over reporting, and codify usage in transparent policies. As the AJP programme has shown, funding, coaching and cross-newsroom sharing can accelerate safe experimentation; the challenge ahead is translating pilots into durable practices that protect accuracy, privacy and public trust while bolstering the capacity of local reporting. [1][2][4][7]

📌 Reference Map:

##Reference Map:

  • [1] (The Beacon) - Paragraph 1, Paragraph 2, Paragraph 3, Paragraph 6, Paragraph 7
  • [3] (The Beacon) - Paragraph 1
  • [2] (The American Journalism Project) - Paragraph 4, Paragraph 6, Paragraph 7
  • [4] (The American Journalism Project) - Paragraph 4, Paragraph 7
  • [7] (The American Journalism Project) - Paragraph 4, Paragraph 6, Paragraph 7
  • [5] (BlueLena) - Paragraph 6
  • [6] (Fund Journalism / News Revenue Hub) - Paragraph 6

Source: Noah Wire Services