Story perspectives

Anthropic Launches Claude Duet Agent for US Subscribers

By Drooid · Oct 22, 2026, 1:58 PM UTC · How we work

Summary

Story summary

Anthropic launched Claude Duet, a proactive AI agent, in the United States on October 21, 2026.
The preview is limited to U.S. Pro and Max subscribers on a waitlist.
Claude Duet runs on a cloud-based VM with its own browser and never accesses passwords.
Anthropic warns the preview may misjudge urgency or misread context and does not yet show its browser view.

History

Historical Context

OpenAI announced GPT-2 on 14 February 2019, releasing only a limited demo while citing misuse risks such as spam and fake news. The Allen Institute for AI responded with a detection tool, and skeptics like The Verge questioned the threat. After monitoring misuse, OpenAI unveiled the full 1.5-billion-parameter model on 5 November 2019, shaping today’s cautious AI-access policies.

1 of 2

Story summary

Anthropic launched Claude Duet, a proactive AI agent, in the United States on October 21, 2026.
The preview is limited to U.S. Pro and Max subscribers on a waitlist.
Claude Duet runs on a cloud-based VM with its own browser and never accesses passwords.
Anthropic warns the preview may misjudge urgency or misread context and does not yet show its browser view.

1 / 2