The Philadelphia Inquirer and other local news outlets are using artificial intelligence to expand coverage and improve efficiency.
The Philadelphia Inquirer hadn’t written much about suburbs like Lower Merion, Pa., and Cherry Hill, N.J. for years. Now it is revisiting those communities — with an assist from AI.
Last year, the Inquirer launched newsletters in four locations, amassing more than 50,000 free subscriptions.
Reporters are using artificial-intelligence tools to scan community meetings for topics that may prompt news, such as a zoning issue related to an ICE detention facility and a proposal for a new data center. The effort is partly funded by a partnership between OpenAI and Microsoft and the nonprofit Lenfest Institute, which owns the Inquirer.
Matt Boggie, chief technology and product officer at the Inquirer, sees AI as an opportunity for growth after years of shrinking the paper’s footprint.
“If we go into these areas and can give context people appreciate, they’re more likely to become paying subscribers to the paper,” he said, adding that the newsletters are “a massive subscription driver” so far. The paper now has plans for eight more AI-assisted newsletters overseen by two new staffers.
Cash-strapped local papers have faced years of print circulation and web traffic declines, and cut staff as revenue dried up. Some newsrooms see an existential threat from the rise of chatbots and AI-powered search engines that answer users’ questions without requiring them to click on article links. But AI could be a potential savior as well: It promises a way to monitor police scanners and town meetings — the time-consuming bread and butter of much local journalism — more efficiently, and even opens the door to expanding coverage.
In Des Moines, Iowa, Axios reporter Linh Ta used to attend three-hour city budget meetings. Now, she feeds the budget into Anthropic’s Claude to highlight notable line items for her newsletter. With the time she saves, she is planning to do more in-depth reporting on state water-quality issues.
“I don’t have to scramble constantly with the daily grind,” Ta said.
Axios, which produces newsletters in 34 communities, eventually wants to expand its reach into the hundreds, said Chief Operating Officer Allison Murphy. OpenAI is helping to fund the expansion.
Outlets including the Andover Advertiser and Hampshire Chronicle in southern England have been using AI to draft entire articles on topics like problems at a local nursing home and the return of a historic locomotive. More than a quarter of the 60,000 articles published in January by USA Today’s local U.K. papers were built by reporters feeding press releases and notes into AI tools, which then spit out drafts for human editing.
Reporters are also using AI to scan meeting transcripts and compile requests for access to government data. The goal is to cover the communities more comprehensively and free up time for reporters to meet with sources, said Henry Faure Walker, chief executive of USA Today’s Newsquest Media Group, which oversees the U.K. outlets.
“It’s a game changer in terms of building out a sustainable local news business,” he said. Digital subscriptions were up by 32% in 2025 from the prior year, and Newsquest is now looking to increase its staff.
Nota, a startup that helps media companies manage their websites with software and AI tools, launched 11 of its own local news sites last year in places including Chesterfield County, Va., and Sutter County, Calif. Nota’s technology scans and ingests public data, then writes drafts of stories about topics including city council votes, budget meetings and weather, in both Spanish and English. Two part-time employees select and edit the content.
Stories cost under $10 each to produce, far less than traditional newsgathering, said Nota Co-Founder and CEO Josh Brandau. The goal is to eventually generate enough revenue from ads and AI content licensing to fund journalists for breaking and investigative work. “Bots can’t do that,” he said.
Cleveland Plain Dealer Editor Chris Quinn has written more than a dozen columns in the past year about how his newsroom is experimenting with AI tools, including having a human “rewrite specialist” use AI to turn reporters’ notes into article drafts. He dismisses complaints that AI-drafted stories may not be written very elegantly.
“We publish news, not poetry. Who died? What restaurant closed? What was the Browns score?” Quinn wrote in one recent piece. “The quicker we get to the point, the happier our busy readers are.”
Still, reporters and editors say AI isn’t ready to fully replace humans in local newsrooms and they need to fact-check the material produced by Claude, Google’s Gemini or OpenAI’s ChatGPT.
The Philadelphia Inquirer apologized last year after publishing a syndicated summer reading list featuring titles that don’t exist. And the Washington Post took heat for errors in its AI-powered personalized podcast, which was launched in a beta test in December; the company has said new products go through research, prototyping and development. Some other early AI experiments, including from USA Today’s local titles, led to corrections in published articles.
There is also a risk that automating more basic tasks will lead to burnout, not better journalism, said Brier Dudley, editor of the Seattle Times’ Save the Free Press initiative, which advocates for local news reforms.
“Is it leading to deeper, more time-intensive reporting or just enabling them to churn out more clickbait?” he asked.
Read more on The Wall Street Journal
