Why we're building Pique

Early in my career I worked in traditional media — local newspapers and local television — at exactly the wrong time, which turned out to be exactly the right time to learn something. I had a front-row seat to the collapse of the business model. And the thing that stuck with me, the thing I have been turning over for years since, is that nothing about the journalism failed. Readers never stopped caring whether their taxes were going up, whether the development on the corner got its variance, whether their kid's school was on the closure list. What failed was the cost structure underneath the journalism. The presses, the towers, the trucks, the buildings — and eventually the reporters — were priced for an advertising market that left and never came back.

The demand for local accountability never went anywhere. The economics of supplying it did.

This week, after years of circling that gap, we formed a company to close it. It's called Pique. This post is the why, written plainly enough that you can judge for yourself whether we're right.

What actually disappeared

When a local newsroom shrinks, the loss is specific. It's the reporter in the room at the zoning board on a Tuesday night. It's the person who reads the four-hundred-page council agenda packet before the meeting instead of after the vote. It's the phone call that starts, "I noticed something odd in the budget."

Here's the strange part: the raw material never went away. Cities publish almost everything they do — agendas, minutes, budgets, permits, contracts, press releases. The public record is right there, growing every week, and it is overwhelmingly unread. Local journalism wasn't out-reported by a better competitor. The watchdog was priced out of the room, and nothing came to take its place.

What changed

For the first time, software can read everything a city publishes — every agenda, every set of minutes, every filing — every day, at a cost that fits local economics. That sentence was not true three years ago. It is the reason this company exists now and not in 2015.

I want to be careful here, because "AI media company" can mean a lot of things, most of them bad. So let me say exactly what we mean. At Pique, AI is a tool in the toolbelt. It watches the routine: the agendas, the minutes, the procedural churn that used to consume a reporter's week before the real work could start. It flags what changed, what's anomalous, what's quietly important — and it shows precisely where in the record it found it. The journalist does the part that requires a human: showing up, working sources, asking the question that isn't in the packet, getting to the bottom of the story the documents only hint at.

The backlash against AI — strongest among younger audiences — is real, and frankly, it's earned. People have watched AI used to flood the internet with content nobody stands behind. Our answer to that is transparency. Everything we publish traces back to its source in the public record, and we'll show you the machine's work the same way we show you the reporter's. And here is a prediction I'm willing to be graded on: done right, this grows journalism jobs. When the machine handles the routine, a reporter's hour goes further — and reporting that hasn't been affordable anywhere in twenty years becomes affordable again.

The other inverted pyramid

Every journalist learns the inverted pyramid on day one: the most important facts at the top of the story, supporting detail below. Ours is the other one. It orders the company instead of the story.

At the base sits deep, original accountability reporting — long-form, document-driven, the kind of work that reads like traditional news because it is traditional news. The audience at the base is small: the deeply, civically engaged. It is also where all of the value is created.

Stories graduate up the pyramid as their appeal broadens. In the middle, conversation and audio — the council story becomes the thing you listen to on the drive home. At the top, short-form video: mass visibility, more passive consumption, the version of the story that finds you instead of waiting to be found.

The pyramid works in both directions. Everything at the top inherits the sourcing of the base — and every viewer can climb down. If a thirty-second video stops your scroll with a startling claim about your city, you don't have to take it on faith or go do your own research. The full reporting that produced that clip is sitting directly underneath it, one tap away. Most viral content is a dead end; ours has a basement.

And the pyramid no longer costs a fortune to build. Legacy media carried two crushing costs: the newsroom and the pipes — presses, towers, and later, app teams. AI collapses both. Standing up a new channel — a purpose-built app, an interactive, an experience tuned to how a particular audience actually consumes — used to mean a product organization and a year. Now it's a small team and a sprint. The experience a twenty-two-year-old expects has almost nothing in common with the one their parents expect. We don't have to bet on one format and force it on everyone. We can meet every audience in its native habitat, fed by the same verified base.

None of this makes the work easy. Attention is the scarcest commodity in media, and we'll be earning it against everyone else who wants it. The pyramid is how we intend to compete: win attention at the top, build trust at the base, and keep the two connected.

What we believe to be true

1. Local accountability journalism didn't lose its audience. It lost its cost structure. The demand is intact — it shows up every time a school closes or an assessment spikes. The supply side broke. Fix the economics and the journalism comes back.

2. Trust compounds from the base of the pyramid; reach compounds from the top. You need both — built in that order. Reach without a trusted base is just more noise. A trusted base without reach stays a newsletter.

3. AI belongs in journalism — in the open, with transparency. The backlash is earned, and the answer is provenance, not promises. Every claim traceable to its source, every machine contribution visible, nothing published that a human doesn't stand behind.

4. Every audience deserves a native habitat. The cost of bringing a new channel online has collapsed. Forcing one format on every generation is now a choice, and it's the wrong one.

5. Done right, AI grows journalism jobs. When machines absorb the routine, the reporter's hour gets spent on the work only humans can do — and coverage that's been economically impossible for two decades becomes a business again.

6. An AI-native foundation is the advantage. It isn't that others don't see the opportunity — it's that media companies built before this moment face real headwinds to operating from an AI-native stance: cost bases, workflows, and habits formed in a different era. Starting from a clean sheet is rare, and we intend to use it.

Day one

I've been thinking about this for years. The company is a week old. Closing the gap between those two sentences starts with a partner who sees the opportunity the way we do, and I want to thank Drive Capital for being first in the room. Drive built its firm on the conviction that great companies get started in places others overlook — and there's a fitting symmetry in that firm backing a thesis about the most overlooked ground in media. We're grateful they saw it early.

From here, this blog will document what we build — including, openly, how we use AI and where we draw the lines, because a company built on showing the work should expect to have its own work examined.

We're starting where the need is most concrete: cities like the ones we grew up in, where the council meets on Tuesday, the packet is four hundred pages, and right now, nobody's reading it.

If this resonates, follow along. If you want to build it, we're hiring.