Quick Takes: Predictions for 2026 (That I'll Regret Making)

I hate predictions. They’re usually either so obvious they shouldn’t need saying, or so specific they’re wrong by February. But here we are in January, and everyone expects takes, so here are mine.

AI agents won’t actually automate that much. Not because they can’t — they can handle specific, boring, structured work just fine. But the setup, maintenance, and debugging will take longer than the work they replace for 80% of the people trying them. The 20% who have a genuine operational bottleneck will save real time. Everyone else will buy a subscriptions they use once, feel guilty about it, and then abandon it. Hype death by spring.

Personal newsletters will get boring before they get interesting. Right now, newsletters are cool because they’re still countercultural. But the moment everyone starts one, they become just another content channel fighting for attention. The ones that survive will be the ones that don’t chase growth metrics. Most will be abandoned by July.

“Quiet quitting” gets rebranded as “healthy boundaries” and suddenly becomes respectable. People didn’t stop resenting overwork. They just got better at naming it. Expect less performative hustle and more talk about “sustainable pace.” Whether companies actually change their behavior is a different question. I’d bet against it.

The AI arms race slows because the real business problem isn’t the AI — it’s how to make money with it. The companies shipping impressive demos aren’t the ones making profit. This year will be about AI, but it’ll be the unglamorous stuff: cost optimization, margin improvement, boring implementation. Not revolution. Just reality.

More people will quit their side projects than start them. The startup energy has matured into burnout. You’ll see a flood of “I’m shutting down my SaaS” posts. It turns out that building something on the side while working a job is harder than it looks, especially after a few months when the novelty wears off. The ones surviving will be the people who treated it like a business, not a hobby.

Remote work stays, but “async first” becomes a luxury. Tight budgets mean synchronous meetings are back for many teams. Small startups and bootstrapped companies will fight for the talent by staying truly async. Large companies will quietly return to scheduling hell while claiming they’re “flexible.”

Crypto stays quiet. The FOMO cycle is broken. Until prices shoot up again and someone invents a new use case everyone forgot about, crypto gets filed under “solved problem nobody actually needed solved.” The energy has moved to other shiny things.

The real winners aren’t using the newest tools. They’re using three tools really well instead of trying to stay on top of every launch. Focus wins. Every single year, focus wins, and every single year, people ignore this and chase novelty anyway.

Here’s the meta-prediction: most of this will be wrong. By December, I’ll have missed the actual story while focusing on the obvious extrapolations. That’s how predictions always work. But I’d rather be confidently wrong and learn from it than stay silent and wonder.

Check back in a year. I’ll probably need to do what I do best — admit I got things backwards. Until then, if you’re thinking about predictions and trends in your own work, skip the hype machine and focus on what actually moves the needle for you.

The real trend of 2026? Skepticism becoming mainstream again.