Bluesky Is Building an AI Assistant to Let Users Control Their Own Feeds

Bluesky is developing a new AI assistant, Attie, designed to give users more direct control over what they see on social media. Unlike traditional platforms where algorithms decide content based on engagement signals, Attie allows users to build their own feeds using simple, natural language. Instead of relying on opaque systems, users can describe the kind of content they want — for example, posts related to specific interests, communities, or themes — and the assistant generates a customised feed in response. The tool is powered by AI technology and built on Bluesky’s open infrastructure, which already allows developers to create and share custom feeds. Attie effectively lowers the barrier to entry, removing the need for technical knowledge or coding skills. This marks a notable shift in how social media platforms approach content discovery. For years, major platforms have relied on centralised algorithms to curate user experiences, often optimising for engagement and time spent on the app. While effective from a business perspective, this model has drawn criticism for limiting transparency and reducing user control. Bluesky’s approach moves in a different direction — one that prioritises user agency over platform-led curation. By allowing individuals to define their own feeds, the platform is decentralising a key part of the social media experience. Users are no longer limited to a single algorithmic perspective; instead, they can create multiple feeds tailored to different interests or moods. This flexibility could reshape how people interact with content. Rather than scrolling through a single, continuous stream, users may begin to navigate between distinct environments — one focused on news, another on personal interests, and another on niche communities. In this sense, the social media feed becomes less of a fixed product and more of a configurable tool. At the same time, this shift raises practical questions around usability and behaviour. Greater control does not necessarily guarantee a better experience. The effectiveness of tools like Attie will depend on how intuitively users can articulate their preferences, and how accurately the AI can interpret and deliver on them. There is also the question of scale. While Bluesky’s model is designed to be open and user-driven, it remains to be seen how widely this approach will be adopted, particularly in comparison to larger platforms with established ecosystems and user bases. However, the direction itself is significant. As AI becomes more integrated into everyday digital experiences, its role is expanding from passive recommendation to active collaboration. Tools like Attie suggest a future where users are not just consumers of algorithmic content, but participants in shaping it. For Bluesky, this aligns with its broader positioning as an alternative to traditional social platforms — one that emphasises openness, transparency, and user control. Whether that model can scale remains an open question. But with Attie, the platform is clearly betting that the future of social media will be less about what the algorithm decides — and more about what users choose to see.