This Week's Major Aha Moment

For those of you who have been along for the ride, we wrote a newsletter back in August about how the future of UI is changing from clicking through menus and tabs on websites to just telling the web what to do.

This week, it happened before our eyes.

First, the backstory (because it's too good not to tell).

An Austrian developer named Peter Steinberger built an AI agent and called it "Clawdbot," which was a play on Anthropic's Claude, inspired by the little lobster mascot you see when reloading Claude Code. Anthropic's lawyers sent a trademark complaint, so he renamed it "Moltbot.” That lasted exactly three days before he decided it didn't roll off the tongue. The final name: OpenClaw. The whole saga played out in about a week. The internet loved every minute of it, especially the part where someone built a social network called Moltbook, designed exclusively for AI agents. On Moltbook, agents post content, comment on each other's posts, argue, joke, and upvote… all autonomously. Humans can observe, but can't participate. It already has over 1.5 million agents on it. Someone at IBM called it "a Black Mirror version of Reddit."

So what actually is OpenClaw? It's an AI agent that lives on your device and actually does things for you, like browse the web, clear your inbox, schedule your calendar, draft replies, book your haircut. You talk to it through WhatsApp, Slack, iMessage, Telegram, or whatever you already use. You text it like you'd text a friend, and it texts back.

Here's the part that makes this different from every other AI tool you've seen: it's open source.

That means the code is public. Anyone can see how it works, modify it, or build on top of it. No company controls it. No subscription locks you in. Your data stays on your device (not on someone else's server).

Open source sounds scary, but it’s actually amazing for innovation, because it means this thing can evolve fast. Thousands of developers around the world are already building integrations, adding features, and fixing bugs (for free!). 

It also means no single company can kill it, paywall it, or change the terms on you. 

YouTube is flooded with videos of people showing how OpenClaw clears their inbox overnight, gives them morning summaries, and books appointments while they sleep.

Cool? Sure. But that's not why we're excited about it.

What we really see here is a first: a genuine digital worker you can treat like an actual employee.

Think about how you'd onboard a first year analyst at your company. You wouldn't hand them the keys to your bank accounts on day one. But you would give them an email, assign them specific tasks, talk to them over Slack, and pay them a salary to get the work done.

That's exactly what's happening here. Except instead of a salary, you're paying in tokens. The cost is about $20 per day (which varies depending on how frequently you use it). Compare that to an entry-level hire at ~$120K a year.

These agents learn as they go. OpenClaw has persistent memory, meaning it remembers your past interactions over weeks and adapts to your habits. The longer you use it, the more useful it gets. Just like real employees who get better the longer they're on the job.

We're calling it now. If vibe coding was the story of 2025, the thing that changed the entire trajectory of the year — this is the moment for 2026.

We're in the early days. The initial use cases are interesting, but we’re still collectively figuring out the true power of this technology. The agents will get smarter. The integrations will get tighter. The costs will probably come down.

If you can’t tell, we’re really, really excited.

Stay curious,

Julia & Russell

P.S. Happy Super Bowl Sunday to those who celebrate.