Curious: How do AI bots actually interact with humans?

Builder: Chat.

Curious: That’s the old way. How do they interact naturally in a group—like a meeting—without a chat box?

Builder: Voice.

Curious: But most real-time models are built for one-on-one. In a meeting you need to handle interruptions, follow-ups, multiple speakers… and the hardest part: knowing when to speak vs stay quiet.

Builder: Exactly. Otherwise it’s awkward or annoying.

Curious: That’s why SUN is different. It’s a real-time collaborative voice model built for group conversations—tracks context, handles those nuances, and speaks only when it adds value.

Builder: So it’s not a chatbot with a mic. It’s a teammate in the room.

Curious: That’s the future of natural human–agent collaboration. That’s SUN.

Sun powers voice agents that join meetings and calls as active collaborators: listening, speaking when needed, synthesizing discussion, managing context across multiple participants, and generating clear summaries and action items. It is designed to sit alongside how teams already work (calls, screenshares, browser workflows) and make collaboration more productive without forcing new behavior.

Video demo:https://youtu.be/rkF-R_1skoo Live demo:https://demo.getsun.io/