← Back to blog
Tool Friday EP. 18

Tool Friday #18 — Arc: They Stopped Building It. I'm Still Not Leaving.

The workflow
01
One Space per context
Work, personal, one per active client. Different tabs, different favorites, different vibe. Switching context is a swipe, not a window-management exercise.
02
The command bar is the only navigation
Cmd+T, type, go. New tab, switch tab, search, run a command — same key. I haven't touched a bookmarks bar in two years.
03
Vertical tabs that archive themselves
Tabs auto-archive after 24h of neglect. Tab hygiene stopped being a thing I do. It just happens.
04
Little Arc + split view for reference work
Open a link in a throwaway mini-window, or pin two tabs side by side. Reading docs while writing stopped being alt-tab roulette.
Matteo Lombardi
May 29, 2026
Try Arc free → Try it

The Browser Company stopped building Arc in May 2025.

In September 2025, Atlassian — the Jira and Confluence people — bought the whole company for a reported $610 million. Every engineer, designer and PM is now working on Dia, an AI-first browser that launched on Mac last October. Earlier this year the founder, Josh Miller, confirmed Dia would inherit Arc’s greatest hits: the sidebar, the vertical tabs, the shortcuts.

Translation: Arc is done. Not shut down — it still launches, still works, still gets security patches as of this week. But it will never get another feature. Ever. The team that made it is, by their own words, somewhere else now.

I open it every morning anyway. And I’m not switching.

TL;DR: Arc is frozen — security patches only, no new features, the team’s on Dia now. As software it’s a 9/10 and nothing has dislodged it. The only question is whether you need a browser that keeps evolving or one that’s already done. I needed done. If a frozen tool gives you the ick, go to Dia — no hard feelings.

Why everyone says to leave

Let me argue the other side first, because it’s a strong one.

You don’t usually adopt abandoned software on purpose. The standard advice — and most of the “Arc is dead in 2026” posts say exactly this — is: a tool with no roadmap is a liability. The security patches are a courtesy that can end whenever Atlassian decides. Mobile is frozen. Windows was always a half-finished beta and now it’s a frozen half-finished beta. The company’s actual product is Dia, which is a different browser with a different philosophy (Chromium under the hood, an AI chatbot bolted to the address bar). If you build your daily workflow on Arc today, you’re building on a foundation the builders walked away from.

All true. I’m not going to pretend any of it isn’t.

Why I’m still here

And yet. Here’s the thing the “leave Arc” crowd skips: most software updates make the software worse.

Think about the apps you use daily. How many of this year’s “new features” did you actually want? Arc being frozen doesn’t mean it’s missing something — it means it stopped at a point where it was already complete, and now no product manager will ever ship a redesign I have to relearn, a sidebar I have to re-hide, an AI feature I have to turn off. The thing that’s a liability for most abandoned software is, for a tool that was already complete, weirdly a feature.

What Arc actually got right, and what I’ve never found anywhere else in one place:

  • Spaces. A separate environment per context — work, personal, one per client — each with its own tabs and favorites. Not “windows.” Not “profiles” hidden three menus deep. A swipe. This alone reorganized how I work.
  • The command bar. Cmd+T does everything: new tab, switch tab, search, run a command. One muscle memory for the entire browser. I never touch a bookmarks bar.
  • Tab hygiene that’s automatic. Tabs archive themselves after a day. I used to be a 60-tab person. Arc quietly cured it without me deciding to be cured.
  • It gets out of the way. Vertical tabs, a collapsible sidebar, full-content view. The chrome of the browser disappears and you’re left with the page. After a week on Arc, every other browser feels like it’s shouting at you.

None of that needs an update. It was already finished. That’s the whole point.

What I’d actually want to know before installing it

Not “where it shines” — you can get that from any review. The things I’d want a friend to tell me:

The learning curve is real, and you’re investing it in a frozen tool. Arc reorganizes how you think about tabs and spaces. It takes a week to click. Going in, know that you’re spending that week on software that will never reward you with new tricks. For me it paid off years ago. If you’re starting today, that’s a fair thing to weigh.

The security patches are the load-bearing question. Right now, Arc still gets them. A browser without security updates is genuinely dangerous — it’s your most exposed app. The day The Browser Company (now Atlassian) announces patches are ending is the day this recommendation expires. I check for that news the way some people check the weather. Adopt Arc with that one alarm set.

Your migration path is Dia, and Dia is not Arc. When the day comes, the official off-ramp is Dia — same lineage, inheriting the hits, but Chromium and AI-first by design. It’s not a like-for-like. Don’t fall so in love with Arc that you have no exit. I have Dia installed. I just don’t open it yet.

It’s a RAM hog, and that’s the one real complaint. All those live Spaces and tabs have a cost — Arc is heavy on memory, and if you run it next to other hungry apps you’ll feel it. For me the experience is worth it every time. But it’s the one honest mark against it, so go in knowing your machine needs the headroom.

Mac or nothing, basically. If you’re on Windows, the Arc experience was never the full one and now never will be. This whole post assumes a Mac.

Who should install a dead browser in 2026

Install it if: you live in the browser all day, you value organization over novelty, you’re on a Mac, and the idea of “finished software” appeals to you rather than scaring you. You’ll get a 9/10 tool for free that will never get worse.

Skip it if: you want AI-native browsing, you’re on Windows, you need a long-term roadmap to feel safe, or “the company abandoned it” is a dealbreaker on principle. Those are all legitimate. Go to Dia. No hard feelings.

The verdict

Score: 9/10 — with one asterisk the size of a billboard: it’s done.

Arc is the best browser I’ve used, and it’s been discontinued, acquired, and replaced. Both of those are true at the same time, and the tension between them is the entire review.

Would I install it again today, knowing it’s frozen? Yes. Am I recommending it without the asterisk? No — the asterisk is the recommendation. Will I switch when the patches stop? I’ll have to. Until then, you’ll have to drag me.

A finished tool that already does everything beats an evolving tool that does it worse. That’s not nostalgia. It’s just the math, run honestly. Arc is free at arc.net. Set the one alarm. Enjoy the quiet.


Tool Friday is a weekly series where I review one tool I actually use. This week’s tool: a browser the makers walked away from, that I refuse to leave, and the honest case for — and against — running finished software in 2026.

Verdict

Arc is in maintenance mode: security patches, zero new features, the team is on Dia under Atlassian now. As software, it's a 9. As a bet, it comes down to one question — do you need your browser to keep evolving, or do you need it to already be done? I needed done. If you do too, install it with your eyes open: a great tool that will never get better, and might one day stop getting patched. I'll take that trade. You might not.

9
/10
Free
Free. Mac-first (Windows was always a beta, now frozen). No paid tier, no subscription, no upsell. The catch was never the price — it's that the product is finished.