From personal feeds to shared projects
Bluesky CEO Toni Schneider introduced the feature at a conference. He described Attie as a way to make “social programming” available on demand: projects built together, or made for a group rather than an individual.
Attie is a place for Bluesky to test ideas involving AI. When someone starts a project, the app links it to the creator’s account on atproto and suggests ideas based on their public Bluesky profile.
Users can connect their own AI model or choose one with open weights. Bluesky COO Rose Wang said finished websites and games could reach users through the network. In her view, building a project is only part of the work; creators also need people to use it. Attie, she said, gives projects access to Bluesky’s audience rather than leaving them to find users on their own.
That audience is large on paper, but the source offers no active-user figure. Some people who tried Bluesky have returned to X, making distribution a promise to test, not a result the announcement establishes.
A wider bet on the network
Attie is part of Bluesky’s effort to revive the network. Over the coming months, the company plans to focus on both developers and communities. One requested change is support for private data in the protocol. It is now in alpha testing, with a launch planned before the end of the year.
Schneider previously led Automattic and was a partner at True Ventures. After former Bluesky head Jay Graber left that role in March, he led the company temporarily; Graber became chief innovation officer. Schneider now heads Bluesky as it moves into AI.
He expects Attie to help people create many small social projects for friends, arguing that people do not need to communicate with the whole world all the time.
I think the sharper test is not whether Attie can generate a website or game, but whether a project built for a small group can bring people back to Bluesky. The announcement describes a route to an audience; it does not show that those audiences are active enough to sustain the projects.
Daily AI news
Every day we pick what actually matters in AI and explain it plainly — no hype, no filler. Subscribe if you want to follow where the industry is going.
Only what matters — every day
Follow on X