The workplace AI market is moving from reactive chatbots to proactive digital assistants. Anthropic’s new Claude Duet is designed to take initiative, monitor work in the background, and handle tasks without waiting for constant prompts. It is currently available as a research preview for Pro and Max subscribers in the U.S. via a waitlist.
What Makes Claude Duet Different?
Unlike standard chatbots, Claude Duet is built as a proactive partner for professionals and small business owners. Powered by Claude Opus 5.5, it runs in its own virtual machine, so it can continue working even when the user’s laptop is closed or phone is turned off.
Key capabilities include:
- Daily briefings: A morning summary of important activity from connected apps.
- Continuous monitoring: Regular checks throughout the day with alerts for urgent matters, such as unpaid invoices or missed emails.
- Task execution: Support for scheduling meetings, drafting emails, preparing payroll in Gusto, creating financial forecasts, and even building web applications.
- Broad integrations: Connections to more than 6,000 apps and plugins, including Gmail, Slack, QuickBooks, Stripe, HubSpot, Canva, and Shopify.
Security and Human Oversight
Claude Duet is designed with human control in mind. The agent requires explicit user approval before high-stakes actions such as making payments, signing documents, or deleting files. Certain activities, including stock trading and transferring money between accounts, are blocked. A classifier also scans messages to help prevent the sharing of sensitive data such as credit card or government ID numbers.
An Early Version with Limitations
Claude Duet is still an early-stage product. It may occasionally misinterpret messages or misjudge urgency, and personal tasks such as booking tickets are currently less reliable than work-related tasks. Frequent background checks can also use up the usage limits of a Claude plan quickly.
The Bottom Line
Claude Duet points toward a future where AI works for users, not just with them. By shifting the user’s role from operator to reviewer, it aims to free up time for more ambitious work. It is still early, but the direction is clear: workplace AI is becoming more proactive, more integrated, and more autonomous — with humans kept in the loop for critical decisions.