Jack Dorsey's Buzz Didn't Try to Out-Feature Slack — It Targeted the One Thing Slack Structurally Can't Fix

The Play

On July 21, 2026, Jack Dorsey announced on X: "we're launching BUZZ! a new groupchat platform for teams of people and agents of all sizes." Built by Block on Nostr, the decentralized protocol Dorsey has backed for years, Buzz combines channels, threads, direct messages, code repositories, and automated workflows — with AI agents built in as full participants rather than bots that only respond to commands. It's open source under Apache-2.0, available hosted or self-hosted. The launch reached the Hacker News front page at 179 points and 172 comments.

Why it worked

The comment section was not an easy room. "No one is going to churn Slack for this just because it is 'Agent First'. Slack is more than good enough," one commenter wrote. Another: "This is just bots in chat rooms, right? Am I the crazy one?" A former Slack employee raised a real architectural concern: multiplayer agent setups mean "you end up having to write and maintain complex rulesets about what specific resources agents have access to."

But the reply that mattered most wasn't from Block — it was from another commenter, arguing back against the "Slack is good enough" dismissal: "Slack's approach is NOT good enough, because the version of the Agent that lives in Slack cannot see the context of the Agent that lives inside Cursor." That's the actual value proposition, stated more sharply by a stranger in a thread than Block's own copy managed. And it matches what Block said in its launch post before anyone had even pushed back: "The most productive work doesn't happen when someone asks AI for help. It happens when humans and agents are in the same room, working on the same thing, with shared context... There was no one platform designed for that, so we built one." Block's Head of AI Capabilities, Bradley Axen, added the framing that turns "open source" from a technical detail into a growth argument: "Every company is going to need a place where humans and agents work together. The question is whether that place is proprietary or open."

The growth play to steal

1. Don't try to out-feature an entrenched incumbent on the terms it already wins on — find the one structural limitation it can't retrofit, the way Buzz's defenders pointed to agent context getting siloed per-tool as something Slack's architecture can't solve.

2. State the problem in one plain sentence the way Block did in its own launch post — "there was no one platform designed for that, so we built one" — a single-cause framing travels further in a comment section than a feature list.

3. Ship it open source and self-hostable to make the pitch to a technical audience about credibility, not lock-in, the way Bradley Axen framed it: "the question is whether that place is proprietary or open."

4. Expect the "this is just bots in a chatroom" dismissal and don't argue with it directly in your own copy — let a technical user articulate the real differentiator inside the thread, the way the Cursor-context reply did more work than any press release.

5. Launch loud, through a known founder on their own platform, so the technical scrutiny happens in public — even the negative threads (complaints about the site "maxing-out-8-cores") kept the story circulating and pulled in more comments.

Bottom line

Buzz's win wasn't converting the people who think "Slack is more than good enough" — plenty of the room still thinks that, and said so. Its win was giving the people who don't think that a specific, articulable reason to say so in public, inside a 172-comment thread of technical scrutiny: an agent in Slack can't see the context of the same agent inside Cursor. That's a growth play built to survive the harshest possible audience — expect to be called "bots in chat rooms," and make sure someone in the room can explain, better than you can, why that's wrong.

Source: https://news.ycombinator.com/item?id=48995213