​

I watched all of Severance without watching a single episode.

22-minute YouTube recap. Commentary included. Told people at work I'd seen it. Moved on.

I'm not embarrassed by the hack. But I am interested in what it says about where attention is going.

We want the value without the time cost. The insight without the friction. The watercooler moment without the 9 episodes.

And that instinct is bleeding into how most professionals think about building their careers.

We're all watching the recap. Very few people are doing the actual work.

This matters more right now than it ever has. Because the window to build something real, something AI cannot replicate, is closing fast.

Kane Kallaway, a sharp content operator, said something in a recent 2.5-hour interview that stuck with me:

His argument is simple. Free distribution at scale has existed roughly from 2007 to 2028. A 20-year window. The last phase of that window, the algorithmic "for you" era, started around 2020.

After that, AI floods the feeds. Organic reach dies or goes pay-to-play. The cost of starting from zero gets dramatically higher.

I don't know if the timeline is exactly right. But I believe the direction.

The professionals who will look back at this period saying "I wish I had started sooner" are not lazy. They are making a more specific mistake.

They are building the wrong kind of capital.

Career Capital Isn't Content. It's Positioning.

Seth Godin has a concept in This Is Marketing that keeps coming back to me as I read through it: the smallest viable market.

Not the biggest audience. The smallest one that actually matters.

His point is that most people go wide because wide feels safe. More reach, more options, more potential upside. But wide is where you disappear. The people who win find the smallest corner of the map where they, and only they, are the perfect answer.

The same logic applies to your career capital right now.

The professionals gaining real ground aren't posting daily. They've picked a specific lane, a specific audience, a specific problem they solve, and they're going deep on it. They're not trying to be known. They're trying to own a term.

Think about it this way. If someone says "atomic habits," you think of one person. If someone says "self-improvement," you think of no one, or a hundred people. The narrower the term, the faster you own it. The more durable that ownership becomes as AI gets better at generating generic content.

That tweet lands differently when you realize Clear didn't build a following. He built ownership of one idea, repeated and deepened over years. That's the career capital that survives the next wave. Not a strategy. A daily practice of going deeper into one specific thing.

What's Actually Getting Replaced First

Here's where I'll be direct, because I think a lot of people are building false moats right now.

Entertainment content dies first. The hot takes on industry news. The reaction posts. The "here's what I think about X thing that happened this week." Kallaway was blunt about this: the pure media game, attention-for-CPMs, is a losing game to start in 2025 or 2026. AI will produce that content cheaper, faster, and better-optimized than any individual can.

Hollow "human" content dies second. This one is sneakier. There's been a real backlash against AI slop on LinkedIn, which has pushed a lot of creators toward personality-driven posts. Venting about the job market. Jokes about the algorithm. Memes that perform well but teach nothing. It feels authentic. But Arvid Kahl said it perfectly:

Human slop is still slop. Authenticity without value is just noise with a face on it.

What survives? Specific expertise applied to a specific audience's specific problem, delivered by someone with lived proof that it works.

That's the moat. Not your personality. Your precision.

What I'm Actually Building

Here's my honest answer to the question I've been sitting with for a few weeks.

I started writing more this year. Not for virality. To sharpen my thinking. Because by the time AI fully catches up to generating generic business content, I want to own a lane specific enough that the algorithm has nothing better to serve in its place.

I'm also learning video. Short form first, long form eventually. Kallaway's point about video being the last frontier resonates: as writing gets commoditized, the personal story told on camera, with your specific voice and your specific lens, becomes the hardest thing to replicate. Not impossible. But hard.

Codie Sanchez said it plainly this week:

The career capital you build in the next 18-24 months won't just protect your job. It'll define what you're known for on the other side of the AI transition.

The question isn't whether to build it. It's whether you're building the right thing.

The Practical Play

Step 1: Pick the lane before you pick the platform.

Before you post anything, answer this: what is the one intersection of your specific experience and your audience's specific problem that nobody else is covering the way you can? Not your industry. Not your job title. The specific corner of the map that's empty.

Start there. Everything else, the platform, the format, the cadence, is secondary.

Step 2: Build depth before distribution.

Most people have the order backwards. They want the audience first, then figure out what to say. Flip it. Spend 60-90 days going deep privately. Write. Test ideas in conversations. Document what you're actually learning in the work.

Distribution is easier when you have something worth saying.

Step 3: Make one rep. Iterate off it.

Don't optimize for the perfect post. Optimize for the next one being slightly better than the last. Kallaway's biggest lesson from building a content machine from scratch:

That gap closes with reps. Not with waiting.

The window is open. It won't be forever.

​