Anthropic introduces Claude Duet, an AI agent that messages you first
Claude Duet runs in the background, watches connected apps and writes only when something matters.
On October 21 Anthropic opened a research preview of Claude Duet. It is an AI agent that runs in the background: every morning it sends a short brief on what matters in the connected apps, checks them every few hours and writes only when something needs attention.
Before sending a message, making a payment, signing a document or deleting anything, the agent shows the user what it proposes and waits for approval. For now it is open through a US waitlist for Claude Pro and Max subscribers.
It marks a shift from AI that answers questions to AI that watches and warns on its own. Nova Code builds similar solutions for directors: AI watches the company’s data and sends reports and warnings about problems to Telegram by itself.