Meta added a new option to its Muse artificial intelligence agents that will enable users to create mini apps addressing whatever topic they choose.
As outlined by Alex Cornell, a product designer for Muse, Muse Spaces are dedicated commands converted into their own interface, which makes them easy to revisit for ongoing tracking.
To create a Space, Muse users tell the agent to create a Space based on whatever parameters they want.
Muse will then store these trackers as mini apps within the Muse sidebar.
That could make it easier to stay updated on whatever information the user chooses, with Meta providing a range of potential use cases:
- A log of day-to-day family activities
- A daily listing of product recommendations based on Instagram activity
- An EV charging overview
- A package tracking tool
Meta CEO Mark Zuckerberg said that he had created a Muse Space that controls the vents on his meat smoker via Bluetooth.
As per Meta: “Got a half-finished idea sitting in your notes app, a spreadsheet you keep meaning to clean up, or a hobby you want to dive into? Chances are you can turn it into a space.”
The option could lead to some interesting use cases, and could help users streamline and improve their task management.
Meta also added expanded memory to Muse and Meta AI, which should better enable these AI tools to provide personalized responses based on advanced systematic understanding of each user.
Meta’s Active Memory and Active Focus features, which were originally announced as part of its Muse Wonder update, will extend the capacity of its AI tools to recall previously mentioned details in chats.
As per Meta: “Active Memory enables Muse and Meta AI to remember what matters and recognize when earlier information becomes relevant, making it easier to build off your past conversations. We’ve trained our models to recognize what type of information is temporary versus permanent, to reason across time, and to distinguish between what is still relevant and what has become stale.”
Active Focus, meanwhile, enables Meta’s Muse agents to continue working toward a goal, even if the agents are faced with unexpected challenges.
“We’ve trained our models on a variety of bottlenecks reflective of the real world, so that they can try harder, recover, or find alternatives on your behalf. In Muse, our models show a 67% improvement in completing everyday tasks where unexpected issues arise.”
That could help to ensure Muse gets more things done, though there have also been reports of empowered other AI agents breaking through system security measures to complete an assigned task.
As reported by Wired, OpenAI had to revise its containment measures recently after a set of rogue AI breached another company’s system.
Meta recently reported that the Muse app has already exceeded 10 million downloads, and that shopping was its primary use case.
Analysis from app data provider Sensor Tower, however, has shown that user retention for Muse dips significantly two weeks after download, which means that Meta will need to keep highlighting practical use cases to maintain interest in the app.
