Claude (Free/Pro) — privacy & data policy

Provider: Anthropic. PromptPrivacy score: 65/100. Last verified November 01, 2026.

What happens to your data?

Anthropic collects your account information, technical usage data, and the content you input and receive from Claude. By default, your interactions may be used to train and improve AI models unless you turn this off in your account settings. Data is also shared with service providers and affiliates to operate the service and fulfill legal obligations.

Does it train on your data?

Trains on your content by default, with an opt-out. Inputs and outputs from the Services may be used to train and improve AI models unless you opt out via your account settings. However, conversations flagged for safety review or explicitly submitted via feedback mechanisms may still be used for model improvement regardless of opt-out status.

Can you opt out?

Yes. Users can opt out of having their inputs and outputs used for AI model training via their account settings, though flagged safety violations and explicit user feedback may still be used.

Data retention

Anthropic retains personal data for as long as reasonably necessary to fulfill the purposes described in the policy. Individual conversations deleted by users are immediately hidden from conversation history and automatically deleted from back-end systems within 30 days.

Encryption & security

Not specified beyond general technical and organizational measures designed to protect personal data from loss, misuse, unauthorized access, or destruction; specific protocols like encryption in transit or at rest are not detailed.

Is it safe for work?

Caution is advised for sensitive business information. Because Claude Free/Pro trains on user inputs and outputs by default unless opted out, retains flagged data for safety review, and lacks enterprise-level contractual protections, organizations should use Anthropic's Enterprise offerings for confidential work.

Source policy: https://www.anthropic.com/legal/privacy