What has Anthropic announced?
According to The Verge, Anthropic has introduced Claude Duet as a competitor to Meta’s Muse and OpenAI’s Dots. It is designed for professionals and small business owners and is launching as a research preview.
Claude Duet is built around Claude Opus 5.5. The report says it is available through standalone apps on iOS and Android, and through Claude on the web. Access is restricted: prospective users must join a waitlist and need to be Pro or Max subscribers in the US.
What could it mean for everyday business work?
Anthropic describes an agent that provides a morning brief from connected apps, monitors those apps during the day, and sends messages when it judges that attention is needed. Its stated tasks range from scheduling meetings to building dashboards, preparing financial or sales forecasts, prototyping, and building web apps.
Claude Duet draws on memory, skills, and connected apps from Claude. It also has a virtual machine with a browser for completing work. For a small business, the potential value is not just faster drafting. It is having an agent carry out parts of a workflow that would otherwise require someone to move between apps.
That makes task selection important. A useful pilot would focus on a repeated job with clear inputs and an output that a person can check. An internal briefing or draft report is easier to assess than an open-ended instruction to manage operations.
What limits and safeguards should owners understand?
The Verge reports that Claude Duet asks for approval before major actions, such as sending a payment or deleting something. Anthropic also describes a dedicated classifier and safety system prompt intended to reject requests to share its most restricted category of information, including credit card numbers.
Those controls are not a reason to remove human oversight. Anthropic says this early version can misunderstand a conversation or judge urgency incorrectly. It also says the agent is less capable at personal errands such as booking tickets.
For a business, approval should mean a real review of the recipient, amount, records, or proposed change. Forecasts and other analytical outputs should also be checked against their inputs. A confident-looking result is not evidence that the underlying work is correct.
What should a small business do next?
Check eligibility before planning a trial. The access conditions reported by The Verge require a US Pro or Max subscriber and a waitlist. Malaysian businesses should not assume local access from the announcement. They can still prepare a task list and review which information an agent would need.
Choose one low-risk workflow. Write down the allowed actions, the actions requiring approval, and the information that must remain out of scope. Record corrections and review time during the pilot. Expand only if the work is dependable and the checking effort is worthwhile.
Separately, Terabot gives businesses digital workers: AI workers with their own computers in private virtual machines, working in the apps a team already uses and around the clock. It is in private preview, by invitation only.
Key takeaways
- Claude Duet is a research-preview agent aimed at professionals and small business owners, with US subscriber and waitlist restrictions.
- Anthropic describes app monitoring and task execution, but also warns that the early version can misunderstand context and urgency.
- Start with a bounded, low-risk workflow and keep people responsible for approvals and checking results.
Written with AI assistance from the source linked above, and checked against it before publishing. Product names belong to their owners. Check the original source before relying on details.