UncoverAI
AI Platforms

Agentic Workspaces

Anthropic Claude Duet

Trust card

Checked is the America/Los_Angeles day the cited vendor page was last recorded. A missing fact says Not stated by vendor.

Trains on your data

The safety article does not say whether Claude Duet conversations are used to train models.

Checked 13 Nov 2026https://claude.com/resources/articles/claude-duet-safety-privacy-and-security

Default retention

Not stated by vendor

Checked 13 Nov 2026https://claude.com/resources/articles/claude-duet-safety-privacy-and-security

Uptime SLA

Not stated by vendor

Checked 13 Nov 2026https://claude.com/product/claude-duet

Sub-processor notice period

Not stated by vendor

Checked 13 Nov 2026https://claude.com/resources/articles/claude-duet-safety-privacy-and-security

Objection window

Not stated by vendor

Checked 13 Nov 2026https://claude.com/resources/articles/claude-duet-safety-privacy-and-security

Last terms change

Not stated by vendor

Checked 15 Nov 2026https://claude.com/resources/articles/claude-duet-safety-privacy-and-security

SLA, uptime and data terms In Signals: Nov 2026

Related SLAs: xAI Grok Bot, ChatGPT Dots, Hermes Agent, Copilot Studio

Anthropic's product page says Claude Duet is a research preview on a waitlist for Pro, Max, and Ultra subscribers in the US, powered by Opus 5.5, and that Team and Enterprise plans are not included yet. The safety article says it works on a cloud computer and asks before sending a message, paying, or deleting.

Enterprise tier

No

The product page says the research preview is for Pro, Max, and Ultra in the US, and that Team and Enterprise plans are not included yet.

Checked 13 Nov 2026https://claude.com/product/claude-duet

Checked 13 Nov 2026https://claude.com/product/claude-duet

Data Changelog

Published changes. Draft edits stay off this page.
FieldFromToRecordedSource
Name—Anthropic Claude Duet16 Nov 2026https://privacy.claude.com/en/articles/10015870-what-certifications-has-anthropic-obtained
Category—agentic_workspace16 Nov 2026https://privacy.claude.com/en/articles/10015870-what-certifications-has-anthropic-obtained
Summary—Not stated16 Nov 2026https://privacy.claude.com/en/articles/10015870-what-certifications-has-anthropic-obtained
Native models—Not stated16 Nov 2026https://privacy.claude.com/en/articles/10015870-what-certifications-has-anthropic-obtained
Trains on customer data—Not stated16 Nov 2026https://privacy.claude.com/en/articles/10015870-what-certifications-has-anthropic-obtained
Data-training terms—Not stated16 Nov 2026https://privacy.claude.com/en/articles/10015870-what-certifications-has-anthropic-obtained
SOC 2—Stated16 Nov 2026https://privacy.claude.com/en/articles/10015870-what-certifications-has-anthropic-obtained
HIPAA—Stated16 Nov 2026https://privacy.claude.com/en/articles/10015870-what-certifications-has-anthropic-obtained
GDPR—Not stated16 Nov 2026https://privacy.claude.com/en/articles/10015870-what-certifications-has-anthropic-obtained
Inference SLA—Not stated16 Nov 2026https://privacy.claude.com/en/articles/10015870-what-certifications-has-anthropic-obtained
SLA—Not stated16 Nov 2026https://privacy.claude.com/en/articles/10015870-what-certifications-has-anthropic-obtained
Strengths—Not stated16 Nov 2026https://privacy.claude.com/en/articles/10015870-what-certifications-has-anthropic-obtained
Lock-in risks—Not stated16 Nov 2026https://privacy.claude.com/en/articles/10015870-what-certifications-has-anthropic-obtained
Pricing—Not stated16 Nov 2026https://privacy.claude.com/en/articles/10015870-what-certifications-has-anthropic-obtained
model_id—grok-build-0.116 Nov 2026https://privacy.claude.com/en/articles/10015870-what-certifications-has-anthropic-obtained
SummaryNot statedUsers approve important actions before Claude Duet sends potentially sensitive messages, makes payments, or deletes anything.16 Nov 2026https://claude.com/resources/articles/claude-duet-safety-privacy-and-security
StrengthsNot stated
  • Users approve important actions before Claude Duet sends potentially sensitive messages, makes payments, or deletes anything.
  • A new safety classifier running separately from the main model checks messages for sensitive information before they are sent.
  • Users have visibility into tasks and connected apps and can steer Claude Duet mid-task while keeping conversation history and memory until deleted.
  • Credentials including passwords, browser cookies, and sign-in tokens are isolated from the model and only pass through separate secure machines.
16 Nov 2026https://claude.com/resources/articles/claude-duet-safety-privacy-and-security
SummaryUsers approve important actions before Claude Duet sends potentially sensitive messages, makes payments, or deletes anything.Claude Duet is a proactive agent that understands what needs your attention.16 Nov 2026https://claude.com/product/claude-duet
Native modelsNot statedPowered by Opus 5.516 Nov 2026https://claude.com/product/claude-duet
Strengths
  • Users approve important actions before Claude Duet sends potentially sensitive messages, makes payments, or deletes anything.
  • A new safety classifier running separately from the main model checks messages for sensitive information before they are sent.
  • Users have visibility into tasks and connected apps and can steer Claude Duet mid-task while keeping conversation history and memory until deleted.
  • Credentials including passwords, browser cookies, and sign-in tokens are isolated from the model and only pass through separate secure machines.
  • Claude Duet is a proactive agent that understands what needs your attention.
  • It takes on tasks of all sizes, keeps working while you’re away, and waits for your ok when it matters.
  • It works with more than 6,000 connectors and plugins.
  • Your memory, skills, and apps from the Claude app are there from the first message, all in one thread.
16 Nov 2026https://claude.com/product/claude-duet
SummaryClaude Duet is a proactive agent that understands what needs your attention.Under approved zero data retention agreements Anthropic does not store inputs or outputs for eligible Anthropic APIs Anthropic products that use Commercial organization API key including Claude Code accessed via the API and Claude Code for Enterprise plans except where needed to comply with law or combat misuse or harm.16 Nov 2026https://privacy.claude.com/en/articles/8956058-i-have-a-zero-data-retention-agreement-with-anthropic-what-products-does-it-apply-to
Strengths
  • Claude Duet is a proactive agent that understands what needs your attention.
  • It takes on tasks of all sizes, keeps working while you’re away, and waits for your ok when it matters.
  • It works with more than 6,000 connectors and plugins.
  • Your memory, skills, and apps from the Claude app are there from the first message, all in one thread.
Under approved zero data retention agreements Anthropic does not store inputs or outputs for eligible Anthropic APIs Anthropic products that use Commercial organization API key including Claude Code accessed via the API and Claude Code for Enterprise plans except where needed to comply with law or combat misuse or harm.16 Nov 2026https://privacy.claude.com/en/articles/8956058-i-have-a-zero-data-retention-agreement-with-anthropic-what-products-does-it-apply-to