Anthropic’s Claude Duet Wants to Do Your Work Before You Ask

Anthropic's Claude Duet Wants to Do Your Work Before You Ask

Most AI assistants sit and wait for you to type something. With Claude Duet, Anthropic is trying something different: an assistant that comes to you first.

Anthropic opened a research preview of Claude Duet on 21 October. It’s a standalone app for iOS and Android, and it also works on the web at claude.ai. The company says it built Duet for professionals and small business owners who have more on their plates than they can keep track of.

Here’s the pitch. Every morning, Duet sends you a short brief on the most important items across your connected apps. During the day it keeps an eye on those apps and only messages you when something actually needs you. You can also hand it work, from simple jobs like scheduling meetings to bigger ones like building dashboards, putting together financial or sales forecasts, prototyping and building web apps.

Under the hood, Duet runs on Anthropic’s Opus 5.5 model. It works on its own computer in the cloud, a virtual machine with a browser, and it carries over your memory, skills and connected apps from Claude. Reuters reports it plugs into more than 6,000 connectors in the Claude Marketplace, including Gmail and Slack, and can make purchases through Stripe’s Link wallet and Shopify’s Shop Pay.

That’s a lot of reach, so the safety side matters. Anthropic says Duet is designed to ask before it sends a message, makes a payment, signs a document or deletes anything. The Verge notes it also has a dedicated classifier meant to refuse requests to share the most sensitive information, such as credit card numbers.

Anthropic is upfront about the rough edges. It says Duet might misread a thread or misjudge what’s urgent, and it isn’t as strong at personal errands like booking tickets. It also can’t show you its browser yet.

Access is limited for now. It’s waitlist-only for Pro and Max subscribers in the US, it uses the same usage limits as the rest of Claude, and Team and Enterprise plans aren’t included yet. Reuters points out that the launch lands as OpenAI and Meta race to build their own agents.

Why it matters

AI is moving from chatting to doing. An agent that can watch thousands of apps and act on them is genuinely useful, but it forces a business decision: what is it allowed to touch, and who approves what it does? For companies everywhere, including here in India where Duet isn’t available yet, now is a good time to write down those rules. When agents like this open up more widely, teams with clear permissions in place will get the benefit fastest and with the least risk.

Sources: Anthropic, Reuters via Investing.com, The Verge. Written by the xRapid Media LLP team.

Leave a Comment

Your email address will not be published. Required fields are marked *