← All articles

AI News · The Brief

The Brief: October 24, 2026

Plus: Duet’s approval and credential controls, with research-preview limits.

Prepared with AI assistance. Facilitated by Luke.Brief

The AI developments worth your attention, what they change at work, and what to do next.

Anthropic · Original announcement: 2026-10-23

Chat and Cowork for Enterprise

Anthropic says the unified Claude experience is now generally available for Claude Enterprise plans, with chat and Cowork combined into one Claude on web, desktop, and mobile. For business owners, the practical question is whether to review a rollout that lets admins manage access and usage from Organization settings, while organizations on zero data retention for Claude Code, a HIPAA-ready configuration, or customer-managed encryption keys stay on their current setup. Source

Who can use it: Generally available for Claude Enterprise plans as of October 23, 2026; admins can manage rollout in Organization settings. Organizations using zero data retention for Claude Code, a HIPAA-ready configuration, or customer-managed encryption keys stay on their current configuration.

Keep in mind: Agentic work runs in Anthropic-managed cloud sessions and can use more tokens than a quick chat. Organizations with the listed security configurations remain on their current experience.

Anthropic · Original announcement: 2026-10-23

Duet’s approval and credential controls

Anthropic’s October 23 update describes Claude Duet, a research-preview agent that works across connected apps while you are away. It can handle inbox, reply-drafting, booking, and routine-running tasks. The company describes approval checks for sensitive actions and prohibits some actions even with approval. Source

For operators, the practical question is whether an always-on agent fits your approval and data-handling rules. Anthropic says Claude Duet does not see passwords, cookies, or sign-in tokens, keeps browser and credentials isolated, and is still being tuned because some safeguards can miss or flag harmless messages. Source

Who can use it: Claude Duet is in research preview.

Keep in mind: Anthropic says Claude Duet is in research preview; some behaviors are still changing, and the post says the safeguards can occasionally miss something or flag a harmless message. It also says some actions stay off-limits even with approval, including trading securities, banking actions such as moving money between accounts, and accessing someone else's account.

Sources

The Ops Guide Brief

Get AI news you can use.

The developments worth your attention, what they change at work, and what to do next. Choose a daily briefing or a weekly selection.

Free to subscribe. Confirm by email. Unsubscribe any time.