Bluesky launches Attie: design your algorithm and feed with AI from text

Bluesky introduced a new application called Auntiean assistant based on artificial intelligence that promises to create personalized ‘feeds’ through natural language conversations. The novelty becomes relevant now that users are looking for alternatives to X (formerly Twitter): it offers advanced customization and opens a way for developers to build on the decentralized ecosystem.

The announcement was made at the Atmosphere conference by those responsible for the social network, including Jay Graber and Paul Frazee. According to the company, Auntie It operates as an independent application within the same technical universe, designed for people to describe what content they want to see and obtain a flow of publications adjusted to those preferences.

The key piece behind Attie is the language engine it uses: the tool is powered by Claude de Anthropic and leans on the protocolo ATthe open standard that underpins platforms like Bluesky. This allows the assistant to retrieve posts shared over networks that support that protocol.

In practice, users can type prompts in natural text—for example, “show me electronic music and experimental sounds from people I follow”—and Attie will generate a feed composed of related posts, taken from accounts and services connected to the protocol. The company clarifies that use is optional: those who prefer the traditional Bluesky experience can continue with it.

For now, Auntie is in phase beta and has already opened a registration list for interested parties. Toni Schneider, CEO of Bluesky, described the project as a starting point that could make it easier for more developers to build applications on top of the Atmosphere platform.

• Personalization: feeds created by textual instructions in natural language.
• Interoperability: content coming from services that use the protocolo AT.
• AI Engine: processed by Claude de Anthropicwhich improves understanding of preferences.
• Optionality: Users are not required to migrate from the classic Bluesky experience.
• Development: opens opportunities for third parties who want to build new social applications.

The practical implications for users are clear but not without questions: greater personalization may mean better content discovery, but it also raises questions about how data is shared and processed between Bluesky and AI providers.

Among the points to keep an eye on are the moderation and transparency of the algorithm: operating in a decentralized ecosystem, the way in which posts are selected and prioritized will depend on both the AI ​​model and the sources to which Attie has access. That adds a layer of complexity regarding responsibilities and privacy.

For those who use social networks as a source of information or promotion, Attie represents a tool with the potential to segment audiences and discover similar communities. For the general public, it could change the way news and trends are consumed, by allowing each user to define relevance criteria through a simple conversation.

In summary: Auntie comes as an extension of the Atmosphere ecosystem that combines AI and open protocols to build more personalized social experiences. It remains to be seen how companies will handle privacy, moderation, and interoperability as the beta progresses and more people try it.

Similar items

Diego Soler explores technological trends and digital innovations. You will enjoy practical tutorials and direct advice to better understand artificial intelligence, cybersecurity and new technologies.