How Axios uses AI to help deliver high-impact local journalism

Axios is a media company delivering vital, trustworthy news and analysis in the most efficient, illuminating and shareable ways possible. It offers a mix of original and smartly narrated coverage of media trends, tech, business and politics with expertise, voice and smart brevity.

We spoke with Allison Murphy, Chief Operating Officer at Axios, about AI supporting high-impact local journalism and serving communities better.

AI is already a huge part of how Axios Local works. At the core, what we’re trying to do is prove that you can run a sustainable, profitable local news model that delivers high-quality journalism to every community in America. That means solving for scale and efficiency—and that’s exactly what AI is good at. So there’s a really natural fit between what OpenAI is building and what we’re building at Axios Local.

We use AI across the whole workflow—from story creation to editing to distribution—but where it’s really made a difference is helping reporters do important work faster. Readers come to Axios for smart brevity, so we built a custom GPT called the Axiomizer. Reporters drop in their drafts and it suggests sharper headlines, clearer “Why it matters,” “What’s next,” and “Between the lines”—basically helping great reporting land even better with readers.

It’s not replacing journalists. It’s taking strong, expert reporting and making it crisper, clearer, and more useful. We’re also adding editing and style checks into the tool so copy editors can focus on what really needs human judgment, instead of spending time on basic fixes or formatting.

The result is that everyone—reporters and editors alike—gets more time to focus on high-impact journalism, while AI handles the busywork in the background.

“[AI] has already become central in how we do the work of Axios Local.”

There are a lot of ways to think about this, but it really comes down to both coverage and how we work. Our goal is to let reporters spend their time doing what only humans can do—talking to sources, digging into data, and telling great stories. Every minute we save them on production, formatting, or busywork is a win.

That efficiency lets us reach more communities. If we can launch a new city with just one amazing reporter—without needing a whole extra layer of production and support—we can go to places we never could before. That’s exactly what we’ve done in places like Boulder and Huntsville, Alabama, which are our first one-reporter cities.

With AI-powered workflows behind the scenes, a single reporter can produce a great local news product. It means more local coverage, in more places, with the same high bar for quality.

At its core, the local news crisis is really an economic one. Great local journalism has to be deeply tailored to each community, which makes it hard to get the cost efficiencies that other industries rely on. You can’t just copy-and-paste a newsroom.

What AI does is change that math. It lets us get more out of our expert reporters and editors, and it strips out costs that don’t actually add value for readers. By improving the economics, we make it possible to do high-quality journalism in more places.

AI is also opening up whole new sources of information. There’s already a huge amount of public data out there—city council meetings, school board recordings, government transcripts—but it’s basically locked away because no one has time to watch or read all of it. With AI, reporters can get quick, reliable summaries and spot what actually matters. Instead of sitting through a three-hour meeting, they can see where the story is moving and know who to call.

That means great reporters can cover more ground, uncover more stories, and serve their communities better—by turning information that was technically public, but practically inaccessible, into something people can actually use.

“We want to make it so that a reporter can spend all of their time doing the unique work that only an expert human reporter can do.”

Human reporters are always going to be at the center of Axios. That’s non-negotiable. They’re what create trust with readers. They’re what make Axios feel like a neighbor in your pocket—someone who knows your community and tells you what really matters. If you lose that human voice, you lose the whole product.

What we standardize is everything around them. We use technology to make the style consistent, and to handle things like formatting, data, and analytics so reporters don’t have to. Readers care deeply about things like housing prices, school performance, and how their community compares to the next one over—but turning raw data into clear, trustworthy, useful insight takes real technical work.

By building tools that handle that for them—clean charts, vetted math, transparent comparisons—we give every reporter access to capabilities that used to be uneven or hard to scale. That way, every community gets the same high-quality data-driven journalism, while the reporting itself stays local, human, and deeply rooted in the place.

One of the things we’ve really focused on is identifying the parts of our newsletters that readers love—and then figuring out how to make those easier to produce.

A great example is our news roundups. These aren’t just lists of links; they’re deeply curated by local reporters who know which neighborhood blogs, regional outlets, and niche sources actually matter in their community. That kind of curation takes a lot of time.

So we worked with our reporters to capture their process—what they read, how they decide what’s worth sharing, which sources they trust—and built that into our AI prompts. Now, instead of starting from scratch every day, reporters get a short, vetted list of links that already reflects their judgment. They just pick what works. What used to take hours now takes minutes, and every city gets a high-quality roundup that still feels local and human.

We’ve taken a similar approach across the newsletter—breaking it into components rather than trying to automate the whole thing at once. The more specific the task, the better the results. That gives us control, consistency, and much higher quality.

Another great example is how we listen to readers. We run quarterly surveys across all our cities, but we only have one audience insights lead. Before, turning that data into something reporters could actually use took weeks. Now, with AI, we can analyze the responses and generate clear one-page summaries for every city in less than a day. That means reporters get real reader feedback almost immediately, and they can adjust what they cover and how they cover it.

It’s not flashy, but it’s powerful. It keeps us tightly connected to our readers—and it helps every reporter deliver a better local product.

“It's absolutely critical that we have AI in the hands of the journalists [...]”

The value of truly original, expert journalism is only going to keep rising. No AI can build a source relationship or break a scoop. That human trust is irreplaceable, and it’s what great reporting will always be built on.

What AI can do is make that reporting go further. First, it unlocks information that’s already public but hard to access—meeting transcripts, records, data—so reporters can ask better questions and find more stories faster. Second, it transforms how journalism reaches people. A single reported story can now become a newsletter, a video, a podcast, or a social clip without needing a whole production team behind it.

That means a great scoop doesn’t just live in one place anymore—it can reach more audiences, in more formats, with far less friction. There will be disruption, of course. Media always has been. But the upside is huge: more questions answered, more communities served, and more high-quality journalism getting to the people who need it.

And from our perspective, that’s exactly what makes our local mission possible. We’re still early, and there will be bumps along the way—but as lo